JP3820719B2 - Biological condition measuring device - Google Patents

Description

【００６９】
ここで、減衰率をα，角周波数をωとしており、

である。そして、
Ａ1＝ＬＣ …（12）
Ａ2＝（Ｌ＋ＲcＲpＣ）／Ｒp …（13）
Ａ3＝（Ｒc＋Ｒp）／Ｒp …（14）
とおくと、式（10），式（11）は以下のように表わすことができる。
α＝（Ａ2／２Ａ1） …（15）
ω＝√｛（Ａ3／Ａ1）−α2｝ …（16）
【００７０】
このようにしてｓの値が確定し、微分方程式（4）を満足する解が得られる。
以上の知見に基づくことで、四要素集中定数モデルの応答波形に含まれる減衰振動成分を近似する式として、式（5）を用いることができる。
次に、大動脈起始部における圧力波形のモデリングを行う。一般に、大動脈起始部の圧力波形は図４の太線の如き波形であって、同図における時間ｔp は波形の１拍分の時間，時間ｔsは左心室の加圧時間である。四要素集中定数モデルでは、この圧力波形を図５に示す三角波で近似することにする。図５において近似波形の振幅と時間がＥo，Ｅm，ｔp，ｔp1で表わされるとすると、任意の時間ｔにおける大動脈圧ｖ1は以下の式で表わされる。ここで、Ｅoは最低血圧（拡張期血圧）、Ｅmは脈圧，（Ｅo＋Ｅm）は最高血圧（収縮期血圧），ｔpは１拍の時間、ｔp1は大動脈圧の立ち上がりから圧力が最低血圧値になるまでの時間である。
０≦ｔ＜ｔp1の区間：
ｖ1＝Ｅo＋Ｅm｛１−（ｔ／ｔp1）｝ …（17）
ｔp1≦ｔ＜ｔpの区間：
ｖ1＝Ｅo …（18）
【００７１】
そして、式（17），式（18）によって表わされる電圧ｖ1を図３（ａ）の等価回路へ入力した時の応答波形ｖp（即ち橈骨動脈波）は以下のようになる。
０≦ｔ＜ｔp1の区間：

ｔp1≦ｔ＜ｔpの区間：

【００７２】
ここで、Ｅminは、脈波検出装置１が測定する橈骨動脈波形における最低の血圧値（後述する 図１１を参照）である。
式（19）における右辺第３項および式（20）における右辺第２項が既述した式（5）の減衰振動成分であって、これらの項におけるαおよびωは式（15），式（16）により与えられている。なお、Ｂ，ｔb，Ｄm1，Ｄm2は後述する手順にしたがって算出される定数値である。
【００７３】
次に、式（19），式（20）の各定数のうち、既に確定したα，ω以外のものについて検討する。まず、式（17），式（19）を微分方程式（3）に代入すると、次式が得られる。

【００７４】
式（21）が成立するためには、以下の条件が必要となる。

【００７５】
なお、式（24）および（25）はαおよびωを拘束するものであるが、既に式（15），式（16）により得られたα，ωはこれらの式を満足する。
一方、式（18），式（20）を微分方程式（3）に代入すると、次式が得られる。

【００７６】
式（26）が成立するためには式（24），式（25）が成立することに加えて、次式が成立することが必要である。
Ｅo＝｛１＋（Ｒc／Ｒp）｝Ｅmin＝Ａ3Ｅmin …（27）
次に、微分方程式（3）が成立するための条件式（22）〜（25），式（27）に基づいて、式（19），式（20）の各定数を算定する。
【００７７】
まず、Ｅminは式（27）より次式のように得られる。
Ｅmin＝（ＥO／Ａ3） …（28）
また、式（23）よりＢは、
Ｂ＝（ｔbＥm）／（ｔp1Ａ3） …（29）
となる。また、式（22）に式（29）を代入して、ｔbについて解くと、
ｔb＝（ｔp1Ａ3＋Ａ2）／（Ａ3） …（30）となる。
【００７８】
さらに、残りの定数Ｄm1，Ｄm2，θ1，θ2は、橈骨動脈波形ｖpがｔ＝０，ｔp1，ｔpにおいて連続性を維持し得るような値、すなわち、下記に示す条件▲１▼〜▲４▼を満足する値が選ばれる。
▲１▼ 式（19）のｖp（ｔp1）と式（20）のｖp（ｔp1）とが一致すること
▲２▼ 式（20）のｖp（ｔp）と式（19）のｖp（０）とが一致すること
▲３▼ 式（19）および式（20）におけるｔ＝ｔp1の微分係数が一致すること
▲４▼ 式（19）のｔ＝０での微分係数と、式（20）のｔ＝ｔpでの微分係数が一致すること
【００７９】
すなわち、Ｄm1およびθ1は、
Ｄm1＝√（Ｄ112＋Ｄ122）／ω …（31）
θ1＝ｔａｎ-1（Ｄ11／Ｄ12） …（32）
なる値が選ばれる。ただし、
Ｄ11＝（ｖO1−Ｂ−Ｅmin）ω …（33）
Ｄ12＝（ｖO1−Ｂ−Ｅmin）α＋（Ｂ／ｔp）＋（ｉO1／Ｃ） …（34）
であり、ｖO1とｉO1はｔ＝０におけるｖpとｉcの初期値である。
【００８０】
また、Ｄm2およびθ2は、
Ｄm2＝√（Ｄ212＋Ｄ222）／ω …（35）
θ2＝ｔａｎ-1（Ｄ21／Ｄ22） …（36）
なる値が選ばれる。ただし、
Ｄ21＝（ｖO2−Ｅmin）・ω …（37）
Ｄ22＝（ｖO2−Ｅmin）・α＋（ｉO2／Ｃ） …（38）
であり、ｖO2とｉO2はｔ＝ｔp1でのｖpとｉcの初期値である。
【００８１】
このようにして、式（19），式（20）の各定数が得られた。
さて、式（16）の角周波数ωから逆算することにより、血管抵抗ＲCは、
Ｒc＝［Ｌ−２Ｒp√｛ＬＣ（１−ω2ＬＣ）｝］／ＣＲp …（39）
となる。ここで、Ｒcが実数でかつ正となる条件は、
｛４Ｒp2Ｃ｝／｛１＋（２ωＲpＣ）2｝≦Ｌ≦（１／ω2Ｃ） …（40）
である。
【００８２】
一般に、Ｒpは１０3［dyn・s/cm5］程度，Ｃは１０-4［cm5/dyn］程度であり、ωは脈波に重 畳している振動成分の角周波数であるから１０（rad/s）以上であるとみてよい。このため、式（40）の下限はほぼ１／（ω2Ｃ）と見なせる。そこで、簡略化のため、Ｌを近似的に、
Ｌ＝１／（ω2Ｃ） …（41）
とおくと、Ｒcは、
Ｒc＝Ｌ／（ＣＲp） …（42）
となる。
【００８３】
また、式（41），式（42）の関係より式（15）の減衰定数αは、
α＝１／（ＣＲp） …（43）
となる。
式（41）〜式（43）の関係を用いて、α，ω，Ｌによって四要素集中定数モデルの残りのパラメータを表わすと、
Ｒc＝αＬ …（44）
Ｒp＝（ω2Ｌ／α） …（45）
Ｃ＝１／（ω2Ｌ） …（46）
となる。これらの式（44）〜式（46）より、パラメータはα，ω，Ｌが得られることにより確定することが明らかである。
【００８４】
ここで、後述するようにα，ω，Ｂ，ｔbは橈骨動脈波の実測波形から得られ、Ｌは１回拍出量ＳＶに基づいて算出できる。以下に１回拍出量ＳＶに基づくＬの算出手順について説明する。
まず、大動脈起始部の圧力波の平均値Ｅ01は次式により与えられる。
Ｅ01＝｛Ｅoｔp＋（ｔp1Ｅm／２）｝／ｔp …（47）
一方、Ｒc，Ｒp，α，ω，Ｌの間には次式が成立する。
Ｒc＋Ｒp＝αＬ＋（ω2Ｌ／α）＝（α2＋ω2）Ｌ／α …（48）
【００８５】
そして、四要素集中定数モデルを流れる平均電流，すなわち平均値Ｅ01，を（Ｒc＋Ｒp）によって除算したものは、拍動により動脈を流れる血流の平均値（ＳＶ／ｔp）に相当するから、次式が成立する。

【００８６】
このようにして得られた式（49）をＬについて解くことにより、１回拍出量ＳＶからＬを求めるための式が次の通りに得られる。
Ｌ＝ α・｛Ｅoｔp＋（ｔp1Ｅm／２）｝／｛（α2＋ω2）ＳＶ｝ …（50）
なお、血流量を測定することにより式（49）中の平均電流（１／ｔp）｛Ｅoｔp＋（ｔp1Ｅm／２）｝に相当する値を求め、この結果に基づいてインダクタンスＬを算出してもよい。血流量を測定する装置としては、インピーダンス法によるもの，ドップラー法によるもの等が知られている。また、ドップラー法による血流量測定装置には、超音波を利用したもの，レーザを利用したもの等がある。
【００８７】
［４］ 循環動態パラメータの算出方法の原理説明
次に、五要素集中定数モデルに基づいた循環動態パラメータの算出方法の原理的な説明をおこなう。先に触れたように、循環動態パラメータの中のＲc，Ｒp，Ｃ，Ｌが、四要素集中定数モデルを用いて決定されるので、これらのパラメータをもとに静電容量Ｃcの値を決定する。そのために、図３（ｂ）における電流ｉ，電流ｉs，電圧ｖ1，電圧ｖp等を求める必要がある。
【００８８】
まず、左心室圧波形ｅを図４に示すような正弦波で近似する。すなわち、
ωs＝π／ｔs
とおいて、左心室圧波形ｅを次式で表わす。
ｅ＝Ｅm’ ｓｉｎωsｔ …（51）
ここで、Ｅm’は最高血圧であって、図５で言えば（Ｅm＋Ｅo）に相当する。以下、図４に示すように、時間ｔがｔ1≦ｔ＜ｔ2の収縮期とｔ2≦ｔ＜（ｔp＋ｔ1）の拡張期に場合分けして説明することとする。ここで、時刻ｔ1，時刻ｔ2は左心室圧波形と大動脈圧波形との交点における時刻である。
【００８９】
［４．１］ 収縮期
この場合、ｖ1＝ｅが成立するとともに、電圧ｖ1と電流ｉについてはそれぞれ式（1）と式（2）が成立する。したがって、式（1）〜式（3）と式（12）〜式（14），式（51）から、次に示す微分方程式が成立する。
Ａ1（ｄ2ｖp／ｄｔ2）＋Ａ2（ｄｖp／ｄｔ）＋Ａ3ｖp＝Ｅm’ｓｉｎωsｔ…（52）
【００９０】
そこでまず、四要素集中定数モデルと同様にして、この微分方程式の定常解ｖpstを求める。そのために、定常解ｖpstを次式のように仮定する。
ｖpst＝Ｅ1ｃｏｓωsｔ＋Ｅ2ｓｉｎωsｔ …（53）
式（53）を式（52）のｖpに代入して係数を比較することにより、次の２式が得られる。
（Ａ3・ωs2Ａ1）Ｅ1＋ωsＡ2Ｅ2＝０ …（54）
−ωsＡ2Ｅ1＋（Ａ3−ωs2Ａ1）Ｅ2＝Ｅm’ …（55）
【００９１】
これらの式を解くことにより、
Ｅ1＝（−ωsＡ2Ｅm’）／｛（ωsＡ2）2＋（Ａ3−ωs2Ａ1）2｝ …（56）
Ｅ2＝｛（Ａ3−ωs2Ａ1）Ｅm’｝／｛（ωsＡ2）2＋（Ａ3−ωs2Ａ1）2｝…（57）
が得られる。
【００９２】
次に、式（52）の微分方程式の過渡解ｖptrを求める。そのために、
ｖptr＝ｅｘｐ（λｔ）
とおいて、次式のｖpへ代入する。
Ａ1（ｄ2ｖp／ｄｔ2）＋Ａ2（ｄｖp／ｄｔ）＋Ａ3ｖp＝０ …（58）
これにより、次式が得られる。
Ａ1λ2＋Ａ2λ＋Ａ3＝０ …（59）
【００９３】
そこで、この式をλについて解くと次式が得られる。

【００９４】
ここで、｛Ａ2／（２Ａ1）｝2＜（Ａ3／Ａ1）とする（振動モード）と、次式が得られる。

【００９５】
このとき、
β1＝Ａ2／（２Ａ1） …（62）
ω1＝√（Ａ3／Ａ1−β12 …（63）
である。
【００９６】
ここで、さらに過渡解ｖptrを次式のように置く。
ｖptr＝（ａ1ｃｏｓω1ｔ＋ｊａ2ｓｉｎω1ｔ）ｅｘｐ（−β1ｔ）…（64）
すると、電圧ｖpは定常解と過渡解との和で表わされることから、式（53）と式（64）によって次式で与えられる。
ｖp＝（Ｅ1ｃｏｓωsｔ＋Ｅ2ｓｉｎωsｔ）＋（ａ1ｃｏｓω1ｔ＋ｊａ2ｓｉｎω1ｔ）ｅｘｐ（−β1ｔ）…（65）
【００９７】
また、電流ｉは、式（65）を式（2）へ代入することによって、次式のように得られる。

【００９８】
次に、ｔ＝ｔ1のときのｖp，ｉを各々ｖ02，ｉ0として次式の如く仮定する。
ｉ0＝Ｊ0＋（ａ1Ｊ1＋ｊａ2Ｊ2）ｅｘｐ（−β1ｔ1） …（68）
ｖ02＝Ｐ0＋（ａ1Ｐ1＋ｊａ2Ｐ2）ｅｘｐ（−β1ｔ1） …（69）
すると、式（65）〜式（69）より以下の式が成立する。
Ｊ0＝（Ｅ1／Ｒp＋ωsＣＥ2）ｃｏｓωsｔ1＋−ωsＣＥ1＋Ｅ2／Ｒp）・ｓｉｎωsｔ1 …（70）
Ｊ1＝｛（１−β1ＣＲp）／Ｒp｝ｃｏｓω1ｔ1−ω1Ｃｓｉｎω1ｔ1…（71）
Ｊ2＝ω1Ｃｃｏｓω1ｔ1＋｛（１−β1ＣＲp）／Ｒp｝ｓｉｎω1ｔ1…（72）
Ｐ0＝Ｅ1ｃｏｓωsｔ1＋Ｅ2ｓｉｎωsｔ1 …（73）
Ｐ1＝ｃｏｓω1ｔ1 …（74）
Ｐ2＝ｓｉｎω1ｔ1 …（75）
【００９９】
また、式（68）〜式（69）をａ1，ａ2について解くと、次のようになる。

【０１００】
さらに、式（71）〜式（75）から次式の関係が成立することがわかる。
Ｊ2Ｐ1−Ｊ1Ｐ2＝ω1Ｃ …（78）したがって、式（64）に式（76）〜式（77）を代入し、その際に式（78）を用いると、過渡解ｖptrとして次式が得られる。

【０１０１】
ここで、
Ｂ1tr＝ｖ02−Ｐ0 …（81）
ｔ’ ＝ｔ−ｔ1 …（82）
Ｂ2tr＝−｛（１−β1ＣＲp）（ｖ02−Ｐ0）−Ｒp（ｉ0−Ｊ0）｝／（ω1ＣＲp） …（83）
とおくと、
ｖptr＝（Ｂ1trｃｏｓω1ｔ’＋Ｂ2trｓｉｎω1ｔ’） ｅｘｐ（−β1ｔ’）…（84）
が得られる。
【０１０２】
結局、式（65）は次式のようになる。

【０１０３】
次に、前述した式（67）において
Ｄ1st＝（Ｅ1／Ｒp）＋ωsＣＥ2 …（86）
Ｄ2st＝−ωsＣＥ1＋（Ｅ2／Ｒp） …（87）
Ｄ1tr＝｛（１−β1ＣＲp）／Ｒp｝Ｂ1tr＋ω1ＣＢ2tr …（88）
Ｄ2tr＝−ω1ＣＢ1tr＋｛（１−β1ＣＲp）／Ｒp｝Ｂ2tr …（89）
とする。
【０１０４】
すると、電流ｉとしては、

が得られる。
【０１０５】
また、電流ｉsは、
ｉs＝Ｃc（ｄｖ1／ｄｔ）＋ｉ＝ωsＣcＥm’ｃｏｓωsｔ＋ｉ …（91）
として得られる。
【０１０６】
［４．２］ 拡張期
拡張期においては、ダイオードＤがオフとなって、左心室圧ｅがダイオードＤのカソード側の回路へ印加されなくなり、静電容量ＣCを流れる電流は、電流ｉと大きさが等しく、逆方向の電流となる。したがって、電圧ｖ1は上述した式（1）で表わされるとともに、電流ｉ，電流ｉcはそれぞれ以下の式で表わされる。
ｉ＝−Ｃc（ｄｖ1／ｄｔ） …（92）
ｉc＝Ｃ（ｄｖp／ｄｔ） …（93）
【０１０７】
したがって、電圧ｖpは、

となる。
【０１０８】
また、ｉ−ｉc＝ｉp＝ｖp／Ｒp であるから、
ｉ＝ｖp／Ｒp＋Ｃ（ｄｖp／ｄｔ） …（95）
となる。
【０１０９】
次に、式（1）に式（95）を代入して、得られた式の両辺を時間ｔで微分すると次式が得られ る。

【０１１０】
また、式（92）と式（95）から次式が導かれる。
ｄｖ1／ｄｔ＝−｛ｖp／Ｒp＋Ｃ（ｄｖp／ｄｔ）｝／Ｃc …（97）
そして、式（96）と式（97）から次式が得られる。

【０１１１】
したがって、この式を変形すると次式が得られる。
（ｄ3ｖp／ｄｔ3）＋Ａ1’（ｄ2ｖp／ｄｔ2）＋Ａ2’（ｄｖp／ｄt）＋Ａ3’ｖp＝０ …（99）
ここで、
Ａ1’＝（Ｌ＋ＣＲcＲp）／（ＬＣＲp） …（100）
Ａ2’＝（ＣcＲc＋ＣcＲp＋ＣＲp）／（ＬＣcＣＲp） …（101）
Ａ3’＝１／（ＬＣcＣＲp） …（102）
【０１１２】
次に、ｖp＝ｅｘｐ（λｔ）とおいて、これを式（99）へ代入すると次式が得られる。
（λ3＋Ａ1’λ2＋Ａ2’λ＋Ａ3’）ｅｘｐ（λｔ）＝０ …（103）
さらに、以下のような定義をおこなう。
ｐ＝（Ａ1’2／９）−（Ａ2’／３） …（104）
ｑ＝−Ａ1’3／２７＋（Ａ1’Ａ2’）／６−Ａ3’／２ …（105）
ｕ＝｛ｑ＋√（ｑ2−ｐ3）｝1/3 …（106）
ｖ＝｛ｑ−√（ｑ2−ｐ3）｝1/3 …（107）
α’＝−（ｕ＋ｖ）＋Ａ1’／３ …（108）
β2＝（ｕ＋ｖ）／２＋Ａ1’／３ …（109）
ω2＝（ｕ−ｖ）√（３）／２ …（110）
λ1＝−α’ …（111）
λ2＝−β2＋ｊω2 …（112）
λ3＝−β2−ｊω2 …（113）
なお、（ｑ2−ｐ3）＞０であれば振動モードである。
【０１１３】
そして、電圧ｖpをさらに次式のように仮定する。

【０１１４】
すると、式（95）へ式（114）を代入することにより、電流ｉは次式のように変形される。

【０１１５】
ここで、
ｇ0＝（１−α’ＣＲp）／Ｒp …（116）
ｇ1＝（１−β2ＣＲp）／Ｒp …（117）
ｇ2＝（ω2ＣＲp）／Ｒp …（118）
【０１１６】
したがって、電圧ｖ1は式（115）から次式のようになる。

【０１１７】
ここで、
ｆ0＝ｇ0／（α’Ｃc） …（120）
ｆ1＝（β2ｇ1−ω2ｇ2）／｛（β22＋ω22）Ｃc｝ …（121）
ｆ2＝（ω2ｇ1＋β2ｇ2）／｛（β22＋ω22）Ｃc｝ …（122）
次に、計算の簡略化を図るために、以後の説明においては、図４に示す時刻ｔ2をｔ＝０とおくことにする。そして、ｔ＝０における電圧ｖ1，電圧ｖp，電流ｉを各々ｖ01，ｖ02，ｉ0とすると、これらは式（119），式（114），式（115）のｔをｔ＝０とおくことで以下のように得られる。
ｖ01＝ｆ0ｂ1＋（ｆ1＋ｊｆ2）ｂ2＋（ｆ1−ｊｆ2）ｂ3 …（123）
ｖ02＝ｂ1＋ｂ2＋ｂ3 …（124）
ｉ0＝ｇ0ｂ1＋（ｇ1＋ｊｇ2）ｂ2＋（ｇ1−ｊｇ2）ｂ3 …（125）
【０１１８】
次に、式（114）における第２項と第３項を変形することで、電圧ｖp は次式のようになる。

【０１１９】
ここで、

であって、
ｋ1＝ｖ01−ｆ0ｖ02 …（130）
ｋ2＝ｉ0−ｇ0ｖ02 …（131）
ｋ3＝ｇ1−ｇ0 …（132）
ｋ4＝ｆ1−ｆ0 …（133）
である。
【０１２０】
次いで、式（126）のｖp を式（2）へ代入することにより、電流ｉは以下のようになる。

ここで、
Ｄ0＝｛（１−α’ＣＲp）／Ｒp｝Ｂ0 …（135）
Ｄ1＝｛（１−β2ＣＲp）／Ｒp｝Ｂ1＋ω2ＣＢ2 …（136）
Ｄ2＝−ω2ＣＢ1＋｛（１−β2ＣＲp）／Ｒp｝Ｂ2 …（137）
【０１２１】
したがって、式（134）から電圧ｖ1は、

となる。ここで、
Ｈ0＝Ｄ0／（α’Ｃc） …（139）
Ｈ1＝（β2Ｄ1＋ω2Ｄ2）／｛（β22＋ω22）Ｃc｝ …（140）
Ｈ2＝（−ω2Ｄ1＋β2Ｄ2）／｛（β22＋ω22）Ｃc｝ …（141）
【０１２２】
上述したように、以上の説明では時刻ｔ2をｔ＝０としていた。そこで、時間スケールを合わせるために、ｔ→（ｔ−ｔ2）の置き換えを行う。これにより、電圧ｖ1，電圧ｖp，電流ｉは各々式（138），式（126），式（134）から以下のように求められる。

【０１２３】
ここで、
Ｈm＝√（Ｈ12＋Ｈ22） …（145）
φ21＝ｔａｎ-1（Ｈ1／Ｈ2） …（146）
Ｂm＝√（Ｂ12＋Ｂ22） …（147）
φ22＝ｔａｎ-1（Ｂ1／Ｂ2） …（148）
Ｄm＝√（Ｄ12＋Ｄ22） …（149）
φ23＝ｔａｎ-1（Ｄ1／Ｄ2） …（150）
【０１２４】
ちなみに、拡張期における電流ｉsは「０」である。
次いで、１回拍出量ＳＶの理論値を求める。１回拍出量ＳＶは、収縮期における電流ｉsの面積で与えられることから、式（91）で示す電流ｉsを時刻ｔ1〜時刻ｔ2について積分することによって得られる。すなわち、

【０１２５】
［５］脈波解析装置の動作
次に、本実施形態による脈波解析装置の動作を図６ないし図１２を参照して説明する。
図６〜図１０に、第１実施形態における脈波解析装置の動作を示すフローチャートを示す。
【０１２６】
また、図１１に、上述した平均化処理により得られる平均波形の波形図を示す。
さらに図１２に、後述するパラメータ算出処理により得られる橈骨動脈波形と、平均化処理により得られた平均波形とを対比した波形図を示す。
以下、これらの図を参照して動作説明を行うこととする。
【０１２７】
▲１▼ 脈波の測定データ読込処理（ステップＳ１）
（ａ） 脈波読取処理
循環動態パラメータの評価を行うに際して、被験者の診断を担当する診断者は、図２に示すようにカフ帯Ｓ１及び圧力センサＳ２を被験者に装着させ、測定開始のコマンドをキーボード５から入力する。マイクロコンピュータ４はこのコマンドに応答して、脈波の測定指示を脈波検出装置１へ送出する。
この結果、脈波検出装置１が橈骨動脈波を検出して、この橈骨動脈波を表わす時系列デジタル信号をＡ／Ｄ変換器３が出力する。マイクロコンピュータ４は、このデジタル信号を一定時間（約１分間）にわたって内蔵の波形メモリへ取り込む。このようにして、波形メモリには複数拍分の橈骨動脈波形が取り込まれる。
【０１２８】
（ｂ） 平均化処理
次に、マイクロコンピュータ４は、複数拍分の橈骨動脈波形を１拍毎ごとに重ね合わせ、上記の一定時間における１拍当たりの平均波形を求める。そして、この平均波形を橈骨動脈波形の代表波形として内蔵メモリへ格納する。このようにして作成された平均波形の代表波形Ｗ１を、図１１に例示する。
【０１２９】
▲２▼ １回拍出量データ取込処理（ステップＳ２）
次いで、マイクロコンピュータ４は１回拍出量測定器２へ１回拍出量の測定指示を送る。この結果、１回拍出量測定器２が被験者の１回拍出量を測定し、その測定結果がマイクロコンピュータ４によって内蔵の一時記憶メモリへ取り込まれる。
【０１３０】
▲３▼ パラメータ算出処理（ステップＳ３）
つぎに、四要素集中定数モデルに基づいて、五要素集中定数モデルを構成する５つの循環動態パラメータのうち、静電容量Ｃcを除く４つの循環動態パラメータの決定を行う。
【０１３１】
マイクロコンピュータ４は、図７〜図８に示すパラメータ算出処理ルーチンを実行する。その際、当該ルーチンの実行に伴って、図９に示すα，ω算出処理ルーチンが実行され（ステップＳ１０９、Ｓ１１７）る。また、当該α，ω算出処理ルーチンの実行に伴って、図１０に示すω算出ルーチンが実行される（ステップＳ２０３）。
【０１３２】
以下、これらのルーチンの処理内容について説明する。
まず、マイクロコンピュータ４は、図１１に示すごとき橈骨動脈の平均波形について、血圧が最大となる第１ポイントＰ１に対応する時間ｔ1’と血圧値ｙ1，第１ポイントの後に血圧が一旦落込む第２ポイントに対応する時間ｔ2’と血圧値ｙ2，２番目のピーク点である第３ポイントＰ３に対応する時間ｔ3’と血圧値ｙ3，１拍分の時間ｔp，最低血圧値Ｅmin（上述した式（3）と式（4）の第１項に相当する）を求める（ステップＳ１０１）。
【０１３３】
なお、脈波が”なだらか”であって第２ポイントＰ２や第３ポイントＰ３を区別するのが困難であれば、第２ポイントと第３ポイントの時間を各々ｔ2’＝２ｔ1’、ｔ3’＝３ｔ1’と想定する。
次に、処理を簡略化するために、図１３に示すＡ点の血圧値ｙ0を用いて血圧値ｙ1〜ｙ3の正規化処理を行い（ステップＳ１０２，Ｓ１０３）、Ｂ点の値を（ｙ0／２）−０．１に初期設定する（ステップＳ１０４）。
【０１３４】
次いで、以下の手順に従ってＢ，ｔb，α，ωの最適値を決定する。
（ａ） まず、Ｂを
「（ｙ0／２）〜ｙ0」
の範囲で変化させ、同時に、ｔbを
「（ｔp／２）〜ｔp」
の範囲で変化させる。その際、Ｂとｔbは何れも＋０．１間隔で変化させるようにする。そして、Ｂ及びｔbの各々について、
｜ｖp（ｔ1’）−ｙ1｜，
｜ｖp（ｔ2’）−ｙ2｜，
｜ｖp（ｔ3’）−ｙ3｜
が最小となるα，ωを求める。
【０１３５】
（ｂ） （ａ）において求めたＢ，ｔb，α，ωの中で
｜ｖp（ｔ1’）−ｙ1｜，
｜ｖp（ｔ2’）−ｙ2｜，
｜ｖp（ｔ3’）−ｙ3｜
が最小となるＢ，ｔb，α，ωを求める。
【０１３６】
（ｃ） （ｂ）において求めたＢ，ｔbを基準にして、Ｂについては
Ｂ±０．０５，
ｔbについては
ｔb±０．０５
の範囲で、上記の（ａ），（ｂ）の処理を再実行する。
【０１３７】
（ｄ） 上記（ａ）〜（ｃ）の処理の際、αは３〜１０の範囲を０．１間隔で変化させ、各αについて最適なωを算出する。
またωは、各αにおいて、
ｄｖp（ｔ2’）／ｄｔ＝０
となる点について二分法を用いて求める（図１０のフローチャートを参照）。
【０１３８】
なお、上記の各処理におけるｖpの値の演算に際して、式（33）の初期値ｖo1は零とする。
以上のような処理によって、Ｂ，ｔb，α，ωが最終的に決定される。
【０１３９】
（ｅ） ｔp1，Ｅm，Ｅoを式（28）〜式（30），式（44）〜式（46）に基づいて算出する（ステップＳ１２３、Ｓ１２４）。
（ｆ） 式（50）を用いて、測定した１回拍出量ＳＶをもとにＬの値を算出し（ステップＳ１２５）、残りのパラメータＲc，Ｒp，Ｃを式（44）〜式（46）から求める（ステップＳ１２６）。
【０１４０】
次に、五要素集中定数モデルに基づいて、最後の循環動態パラメータである静電容量Ｃcを決定する。
その際、１回拍出量ＳＶの計算値と実測値が一致するように静電容量Ｃcを決定する方法と、計算脈波の最低血圧と実測脈波の最低血圧とが一致するように静電容量Ｃcを決定する方法とが考えられる。そこで、各々の方法について場合を分けて説明する。
【０１４１】
▲３▼−１ １回拍出量ＳＶの計算値と実測値とが一致するように静電容量Ｃc（大動脈コンプライアンス）を決定する方法
最初に、１回拍出量ＳＶの計算値と実測値とが一致するように静電容量Ｃcを決定するための具体的な方法について説明する。
まず初めに、静電容量Ｃcの値を、四要素集中定数モデルにより算出した静電容量Ｃをもとに、次式のように推定する。また、その他の循環動態パラメータ，すなわちＲc，Ｒp，Ｃ，Ｌの値は、四要素集中定数モデルで得られたものを用いる。
Ｃc＝１０・Ｃ …（153）
次いで、これらの循環動態パラメータを用いて、１回拍出量ＳＶの計算値を式（152）によって算出する。
【０１４２】
その際、左心室加圧時間ｔsは、四要素集中定数モデルによって得られた１拍の時間ｔpから、次式によって推定することとする。
ｔs＝（１．５２−１．０７９ｔp）ｔp …（154）
この関係式は、心エコーで左心室の収縮時間を測定した結果から得られた実験式であって、図１４に示すように、相関係数としては−０．８８２が得られている。また、最高血圧Ｅm’については、四要素集中定数モデルにより得られた値を用いる（式（22），式（28）を参照）。
【０１４３】
また、時刻ｔ1，時刻ｔ2に関しては、左心室内圧＝大動脈圧の関係から求めることができる。さらに、前述したようにｖ02とｉ0はｔ＝ｔ1におけるｖp，ｉの値であるから、式（85），式（90）に存在するｔへｔ1を代入することで、ｖ02とｉ0を得ることができる。
次に、上記のようにして求めた１回拍出量ＳＶ計算値が、１回拍出量測定器２から取り込んだ測定値と一致するように静電容量Ｃcの値を決定する。すなわち、静電容量Ｃcの値を式（153）で求めた初期値から所定の範囲内で変化させてゆく。そして、１回拍出量の測定値と、各静電容量Ｃcの値から計算された計算値とを比較して、測定値の整数部分と計算値の整数部分が一致するかどうかを調べる。もし整数部分に一致が見られれば、測定値と計算値とが一致したものと見なし、静電容量Ｃcが決定されてパラメータ算出処理が終了する。
【０１４４】
一方、静電容量Ｃcの値を調整しただけでは１回拍出量の測定値と計算値に一致が見られない場合には、調整した静電容量Ｃcの値の中で、１回拍出量の測定値と計算値との差分が最小であった静電容量Ｃcの値を最終的な値とする。次いで、最高血圧Ｅm’の値を±３ｍｍＨｇの範囲内で１ｍｍＨｇ毎に変化させて、上記と同様に１回拍出量の測定値と計算値との一致の有無を調べる。もし、一致が見られる最高血圧Ｅm’が存在すれば、その値を最終的な最高血圧Ｅm’として、パラメータ算出処理を終える。
【０１４５】
他方、最高血圧Ｅm’の値を調整しても、まだ１回拍出量の測定値と計算値に一致が見られない場合には、さらに抵抗Ｒpの値を調整する。そこで、調整した最高血圧値Ｅm’の値の中で、１回拍出量の測定値と計算値との差分が最小であった最高血圧値Ｅm’の値を最終的な値とする。次いで、抵抗Ｒpを例えば１０［dyn・s/cm5］刻みで増減させて、１回拍出量の測定値と計算値との差分が最も小さい値を最終的な抵抗Ｒpの値に決定する。
【０１４６】
以上説明した過程を実現するフローチャートの一例を図３１に示す。なお、プログラム中で所定の範囲内で変動されるパラメータに対しては、元々のパラメータ名に対して下添字の「ｖ」を付けた。
【０１４７】
▲３▼−２ 計算脈波の最低血圧と実測脈波の最低血圧とが一致するように静電容量Ｃcを決定する方法
次に、計算脈波の最低血圧と実測脈波の最低血圧とが一致するように静電容量Ｃcを決定する方法について説明する。
この場合において、従来は、収縮期時間ＱＴを初期値として用いて静電容量Ｃcを決定していた。
【０１４８】
この初期値として用いる収縮期時間ＱＴの算出方法としては、従来は、被験者の心電図より収縮期時間ＱＴを予め求めたり、心電図あるいは脈波波形より得た心拍数ＨＲから収縮期時間ＱＴを求める回帰式を用いて算出したりしていた。
そして、この予め得られた収縮期時間ＱＴに対して、左心室加圧時間ｔsvを「ＱＴ＋０．１〔sec〕」〜「ＱＴ＋０．２〔sec〕」の範囲で「０．０１〔sec〕」間隔で変化させ、同時に最高血圧Ｅmv’を「Ｅo＋Ｅm−２０〔ｍｍＨｇ〕」〜「Ｅo＋Ｅm＋２０〔ｍｍＨｇ〕」の範囲で「１ｍｍＨｇ」間隔で変化させる、すなわち、これら左心室加圧時間ｔsvおよび最高血圧Ｅmv’の各々に対して、４５１通りの組合せが想定され、これら各組合せにおいて、計算脈波の最低血圧と実測脈波の最低血圧とが一致するような静電容量Ｃcが計算する構成としていた。
【０１４９】
しかしながら、上記従来の心電図から収縮期時間ＱＴを求める方法においては、予め心電図を採取する必要があり、装置構成が大型化、複雑化してしまうという問題点があった。
また、心拍数ＨＲから回帰式を用いて収縮時間ＱＴを求める方法においては、正確な回帰式を求めることが困難であるという問題点があった。
【０１５０】
図３３に、一般的な脈波波形を示す。
ところで、脈波波形は、心臓の収縮・拡張によって生じる血液流の脈動を末梢部で測定したものであるから、その波形形状には、心臓の動きが反映されている。図中のＥＥＤ（Estimated Ejection Duration）は概略駆出期間と呼ばれ、１回の心拍中に心臓から血液が流れ出る時間、ひいては、収縮期時間ＱＴに対応している。
【０１５１】
そこで、本実施形態においては、この収縮期時間ＱＴに代えて、概略駆出時間ＥＥＤ（Estimated Ejection Duration）を用い、この概略駆出時間ＥＥＤに対して、左心室加圧時間ｔsvを「ＥＥＤ＋０．１〔sec〕」〜「ＥＥＤ＋０．２〔sec〕」の範囲で「０．０１〔sec〕」間隔で変化させ、同時に最高血圧Ｅmv’を「Ｅo＋Ｅm−２０〔ｍｍＨｇ〕」〜「Ｅo＋Ｅm＋２０〔ｍｍＨｇ〕」の範囲で「１ｍｍＨｇ」間隔で変化させる。
【０１５２】
すなわち、これら左心室加圧時間ｔsvおよび最高血圧Ｅmv’の各々に対して、４５１通りの組合せが想定されることになる。これら各組合せにおいて、計算脈波の最低血圧と実測脈波の最低血圧とが一致するような静電容量Ｃcが計算される。
この結果、各被験者自身の末梢部における圧脈波波形によるＥＥＤを用いて循環動態をパラメータを求めることができるため、心電計が不要となるなど装置構成を簡略化でき、被験者の循環動態を反映したより正確な循環動態パラメータを算出することができるのである。
【０１５３】
次に、各組合せにおける計算脈波のサンプリング値をＰ1（ｔ）とし、実測脈波のサンプリング値をＰ2（ｔ）としたとき、各組合せにおける波形平均誤差εは下式により求まる。そして、波形平均誤差εが最も小さい場合における静電容量Ｃc（大動脈コンプライアンス）が採用される。以上説明した過程を実現するフローチャートの一例を図３２に示す。
ε＝Σt=0tp（｜Ｐ2（ｔ）−Ｐ2（ｔ）｜）／（Ｎ） …（155）
【０１５４】
以上のようにして、１回拍出量の測定値と計算値が一致する循環動態パラメータが全て決定されたことになる。
ここで、３２歳の男性を被験者とした場合について橈骨動脈波形から算出した循環動態パラメータ等の値を以下に示す。
静電容量Ｃc ＝ ０．００１２１３〔cm5/dyn〕
電気抵抗Ｒc ＝ ９８．７６８〔dyn・s/cm5〕
インダクタンスＬ ＝ １５．９３０〔dyn・s2/cm5〕
静電容量Ｃ ＝ ０．０００１２４１〔cm5/dyn〕
電気抵抗Ｒp ＝ １３００．０５８〔dyn・s/cm5〕
左心室加圧時間ｔs ＝ ０．４９６〔s〕
１拍の時間ｔp ＝ ０．８９６〔s〕
１回拍出量ＳＶ ＝ ８３．６〔cc/拍〕
最高血圧Ｅm’ ＝ １１７．４４〔mmHg〕
また、図１２に示す通り、算出したパラメータから求めた橈骨動脈の計算波形と実測波形とは良く一致していることがわかる。
【０１５５】
▲５▼ データ算出処理（ステップＳ４）
さらに、循環動態パラメータＬ，Ｃ，Ｃc，Ｒc，Ｒpの値等をもとにして、大動脈圧波形が求められる。すなわち、収縮期にあっては式（51）を用い、拡張期にあっては式（142）を用いることにより、電圧ｖ1の波形を１拍分（すなわち、時刻０〜時刻ｔp或いは時刻ｔ1〜時刻（ｔ1＋ｔp））だけ計算する。
【０１５６】
そして、次に、得られた大動脈起始部の波形の時刻ｔ1における値をこれらの式から算出して、その算出結果を最低血圧値Ｅoとする。
次に先に求めた最高血圧値Ｅm‘に心拍数HR（＝６０／ｔp）を乗じることにより、心筋負荷指数ＷＰを算出する。

【０１５７】
▲６▼ データ出力処理（ステップＳ５）
次にマイクロコンピュータ４は、パラメータ算出処理により得られた循環動態パラメータＬ，Ｃ，Ｃc，Ｒc，Ｒpを出力装置６へ出力し、出力装置上に表示する。
また、得られた計算波形を出力装置６へ出力して大動脈圧波形の表示を行う。さらに最高血圧値Ｅm’、心筋負荷指数ＷＰを最低血圧値Ｅoと一緒に出力装置６へ送出して、これらの値を出力装置６上に表示させる。
【０１５８】
［６］ 第１実施形態のまとめ
ところで、従来の血圧測定装置では、橈骨動脈部，上腕部等の末梢側において血圧を測定しており、心臓の負担を間接的に測定する手法であると言える。ところが、心臓の負担の変化が末梢側の血圧に反映されているとは限らないのであって、心臓の負担を末梢側で見るということは、必ずしも的確なものとは言えない。
【０１５９】
このようなことから、本第１実施形態では、とりわけ中枢部の血圧波形が心臓の負担を見る上で重要であることに着目し、大動脈起始部（動脈系の中枢部）の血圧波形を末梢側で測定した脈波波形から推定して求めるようにしている。そして、推定された大動脈圧波形から、大動脈起始部における最高血圧値，最低血圧値並びに心筋負荷指数ＷＰを算出すれば、これらの値が心臓の負担を直接的に表わす指標となりうる。
【０１６０】
このように、本実施形態によれば、各種の循環動態パラメータとともに、中枢側の最高血圧，最低血圧，心筋負荷指数、大動脈圧波形を診断者や被験者に対して示すことができる。
【０１６１】
なお、式（51）は左心室圧波形そのものを示すことから、中枢部における圧波形として、上述した大動脈圧波形の代わりに左心室圧波形を出力装置６へ表示させるようにしても良い。
【０１６２】
第２実施形態
上記第１実施形態では、橈骨動脈波形と１回拍出量から循環動態パラメータの各値を算出することとした。しかるに、上述したように、１回拍出量の検出を行うには、被験者がカフ帯Ｓ１を装着する必要があるため、被験者にとって煩わしいものと言える。
【０１６３】
そこで、本第２実施形態では、橈骨動脈波形の形状によって大動脈圧が変化するという現象に着眼して、波形の形状をひずみ率で代表させて中枢側の血圧値等を推定するものである。すなわち、本実施形態では、橈骨動脈波形から得られるひずみ率ｄをもとにして循環動態パラメータを導出する。
【０１６４】
まず、マイクロコンピュータ４は、第１実施形態と同様にして、▲１▼脈波読み取り処理と▲２▼平均化処理を実施して、橈骨動脈波形の１拍分の平均波形を求める。次に、この平均波形に対して周知のＦＦＴ（高速フーリエ変換）処理を施すことによって、脈波のフーリエ解析を行う。そして、解析の結果として得られた周波数スペクトルから、基本波の振幅Ａ1，第２高調波の振幅Ａ2，第３高調波の振幅Ａ3，…，第ｎ高調波の振幅Ａnを求める。なお、ｎ（ｎは自然数）の値は、高調波の振幅の大きさを考慮して適宜決定するものとする。そして、これらの振幅値をもとにして、次式で定義されるひずみ率ｄを算出する。
ひずみ率ｄ＝（Ａ22＋Ａ32＋…＋Ａn2）1/2／Ａ1 …（156）
【０１６５】
次いで、得られたひずみ率ｄから循環動態パラメータを推定する。推定にあたっては、橈骨動脈波形のひずみ率と循環動態パラメータの各値の間に相当程度の相関関係があるという知見に基づいて行う。すなわち、予め多数の被験者についてひずみ率ｄと循環動態パラメータとを測定して、ひずみ率と各循環動態パラメータの間の関係式を導出しておく。ここで、ひずみ率ｄと循環動態パラメータＲC，Ｒp，Ｌ，Ｃの測定結果との相関関係の一例を、図２５〜図２８に示しておく。なお、大動脈コンプライアンスＣCに関しては図示していないが、他の四つのパラメータと同様に相関係数と関係式を求めることができる。
【０１６６】
そして、上記の式（156）で算出したひずみ率ｄと図２５〜図２８に各々図示した関係式に基づいて、循環動態パラメータＲc，Ｒp，Ｌ，Ｃ，Ｃcを計算する。
次いで、第１実施形態における▲５▼および▲６▼の出力処理と同様にして、算出した循環動態パラメータから、大動脈圧波形の１拍分の波形を求めるとともに、大動脈起始部における最低血圧値Ｅo、最高血圧値Ｅm‘及び心筋負荷指数ＷＰを算出して、これらを出力装置６上へ表示させる。
【０１６７】
第３実施形態
本第３実施形態は、大動脈起始部における最高血圧値、最低血圧値あるいは心筋負荷指数ＷＰに加え、上記のようにして求めた大動脈起始部の血圧波形から、心臓の仕事量（以下、心仕事量と呼ぶ）を算出して、これを表示させるものである。
【０１６８】
この心仕事量は、心臓の負担を表わす１指標であって、１回拍出量と大動脈圧との積で定義され、１分あたりの心拍出量を仕事量に換算したものである。
ここで、１回拍出量は、１回の拍動で心臓から送り出される血流量で定義され、心臓から出る血流波形の面積に相当するものである。この１回拍出量は、大動脈圧波形の収縮期の面積と相関があり、大動脈圧波形に対して収縮期面積法を適用することで１回拍出量を求めることができる。
【０１６９】
すなわち、まず、心臓の収縮期に対応する部分の脈波波形の面積Ｓを算出する。これを図２９の脈波波形で説明すると、脈波の立ち上がりの部分から窪み（ノッチ）に至る領域の面積，即ち同図でハッチングを付した部分が、面積Ｓに相当する。次いで、所定の定数をＫとすると、１回拍出量ＳＶを次式によって算出することができる。
１回拍出量ＳＶ［ｍｌ］＝面積Ｓ［ｍｍＨｇ・ｓ］×定数Ｋ
【０１７０】
一方、心拍出量は、心臓から１分間に送り出される血流量で定義される。したがって、心拍出量は１回拍出量を１分間に換算することで得られる。すなわち、心拍出量は、１回拍出量と心拍数の積によって求められる。
本実施形態では、第１実施形態又は第２実施形態の▲５▼の出力処理において、マイクロコンピュータ４が、算出された左心室圧波形をもとに心仕事量を算出して出力装置６へ表示する。その他の処理は、第１実施形態或いは第２実施形態と同じであり、その説明は省略する。
【０１７１】
ここで、マイクロコンピュータ４は、以下に示す手順によって心仕事量Ｗsを算出する。
まず、ｗsをｅ・ｉsで定義すると、これは式（51），式（90），式（91）から次式のように算出される。

【０１７２】
ここで、式（157）における第１項，第２項，第３項をそれぞれｗ1，ｗ2，ｗ3とすると、各々は以下の式のように変形される。

【０１７３】
次に、式（82）より

とおき、
ωsｔ−ω1ｔ’＝（ωs−ω1）ｔ＋ω1ｔ1＝ωbｔ＋Φ …（165）
とおく。
【０１７４】
すなわち、
ωa＝ωs＋ω1 …（166）
ωb＝ωs−ω1 …（167）
Φ ＝ω1ｔ1 …（168）
である。
【０１７５】
すると、式（163）は次式のようになる。

【０１７６】
次に、Ｗ1，Ｗ2，Ｗ3をそれぞれ以下のように定義し、式（161），式（162），式（169）から以下の式を導出する。

【０１７７】
仕事量Ｗsは、上記のＷ1，Ｗ2，Ｗ3の総和を”分”あたりに換算して得られることから、最終的に次式で表わされる。
Ｗs＝（Ｗ1＋Ｗ2＋Ｗ3）×１０-7×６０／ｔp〔Ｊ／分〕 …（176）
大動脈起始部における最高血圧値，最低血圧値及び心筋負荷指数ＷＰに加えて、以上説明したような心仕事量を表示する意味は次のようなものである。
【０１７８】
大動脈圧波形をもとに上述した心仕事量を求めることで、心臓の負担を表わす指標として、大動脈起始部の最高血圧値，最低血値値あるいは心筋負荷指数ＷＰとは別の有用な指標を提供することも可能となる。
ここで、心仕事量を算出することによる意義について以下に例を挙げて説明することとする。
【０１７９】
いま、患者へ降圧剤を投与して高血圧の治療を行う場合を考えてみる。通常、薬が効いているのであれば、橈骨動脈部で測定される最高血圧値，最低血圧値に変化が現れて薬の効果を確認することができる。
ところが、最高血圧値，最低血圧値に変化が見られない場合であっても、実際には薬が効いていて、心臓の負荷自体は軽くなっていることがある。
これは、降圧剤の役割としては動脈系のどこかで心臓の負荷を小さくしていれば良く、必ずしも橈骨動脈部における血圧が下がっている必要はないからである。
【０１８０】
このように、橈骨動脈部等の動脈系の末梢部における血圧値に顕著な変化が見られない場合であっても、大動脈起始部の血圧波形から求めた心仕事量を算出することで、真の心臓の負担を知ることが可能となるのである。
ところで、このような心臓の負担の変化は、大動脈起始部の血圧波形を子細に検討することで見い出せるのではあるが、心仕事量を算出することによって微妙な波形の変化を定量的に表現できるようになるのである。
【０１８１】
したがって、最高血圧値や最低血圧値ばかりでなく、心筋負荷指数ＷＰ及び心仕事量を求めてこれを表示することによって、降圧剤療法の評価をいっそうきめ細かく行うことが可能となるのである。
図２２〜図２４に上述した第１ないし第３のタイプの各脈波形状について心仕事量を算出した結果を示す。
【０１８２】
第４実施形態
以上の第１ないし第３実施形態においては、心筋負荷指数ＷＰを算出するに際しては、常に大動脈起始部血圧及び検出した心拍数に基づいていたが、本第４実施形態は末梢部血圧と大動脈起始部血圧との差を無視することが可能な心拍数範囲では、大動脈起始部血圧に代えて末梢部血圧を用い、末梢部血圧と大動脈起始部血圧との差を無視することができない心拍数範囲では、大動脈起始部血圧を用いることにより、全体として演算処理の軽減を図るための実施形態である。
【０１８３】
一般に心拍数が増加すると、大動脈起始部血圧と末梢部血圧との差は大きくなることが知られている。
そこで本第４実施形態においては、心拍数の変動範囲が所定の基準心拍数変動範囲内である場合には、末梢部血圧（最高血圧）及び検出した心拍数を用いて心筋負荷指数ＷＰを求め、心拍数の変動範囲が基準心拍数変動範囲以上となった場合には、大動脈起始部血圧（最高血圧）及び検出した心拍数を用いて心筋負荷指数を求めることとした。
【０１８４】
より具体的には、安静時におけるヒトの心拍数（もちろん、個人差を有する）である基準心拍数に対する心拍数変動は、±１０［％］程度である。
従って、例えば、実際の心拍数変動が安静時の基準心拍数に対して±１０［％］未満であるならば、末梢部血圧と大動脈起始部血圧との差を無視することが可能であると判断して末梢部血圧Ｐperi及び心拍数ＨＲに基づいて次式により心筋負荷指数ＷＰを算出する。
ＷＰ＝Ｐperi×ＨＲ
一方、実際の心拍数変動が安静時の基準心拍数に対して、測定誤差マージンを考慮した±１５［％］以上となった場合には、末梢部血圧と大動脈起始部血圧との差を無視することができないと判断して、大動脈起始部血圧Ｐcent及び心拍数ＨＲに基づいて次式により心筋負荷指数ＷＰを算出することとなる。
ＷＰ＝Ｐcent×ＨＲ
これにより本第４実施形態によれば、実際の心拍数変動が安静時の基準心拍数に対して所定範囲未満であれば、末梢部血圧を用いて心筋負荷指数ＷＰを算出することとなるので、常に大動脈起始部血圧を用いて心筋負荷指数を算出する場合と比較して、処理を簡略化し、処理速度の向上を図ることが可能となる。
【０１８５】
また、実際の心拍数変動が安静時の基準心拍数に対して所定範囲以上であれば、循環動態パラメータに基づいて算出した大動脈起始部血圧を用いて心筋負荷指数ＷＰを算出することとなるので、末梢部血圧を用いて心筋負荷指数ＷＰを算出する場合と比較してより正確な心筋負荷指数を算出することが可能となる。
このように本第４実施形態によれば、全体として処理を簡略化することができるにも拘わらず、常に正確な心筋負荷指数ＷＰを算出することが可能となる。
【０１８６】
第５実施形態
上記第１実施形態ないし第３実施形態においては、所定のタイミングで常時、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数ＷＰを算出する構成としていたが、大動脈起始部血圧があまり変化したとは考えられない場合には、必ずしも心筋負荷指数ＷＰを継続的に算出する必要はないと考えられる。
【０１８７】
そこで、本第５実施形態は、大動脈起始部血圧が大きく変化したと考えられる場合にのみ新たに心筋負荷指数ＷＰを算出し、大動脈起始部血圧があまり変化したとは考えられない場合には、心筋負荷指数ＷＰの算出を行わずに前回求めた心筋負荷指数ＷＰを保持、表示することにより演算処理量を低減するための実施形態である。
【０１８８】
ところで、大動脈起始部血圧を算出するためには、それに先だって循環動態パラメータの算出が必要である。
この場合において、今回求めた循環動態パラメータの前回（あるいは複数回前）に求めた循環動態パラメータに対する変化（変化率）が小さい場合には、今回求めた循環動態パラメータにより得られるであろう大動脈起始部血圧の前回に求めた循環動態パラメータにより得られるであろう大動脈起始部血圧に対す変化（変化率）も小さいと考えられる。
【０１８９】
そこで、本第５実施形態においては、今回求めた循環動態パラメータと前回求めた循環動態パラメータとを比較し、各循環動態パラメータの変化率が予め定めた基準変化率未満である場合には、大動脈起始部血圧の算出、ひいては、心筋負荷指数ＷＰの算出を行わずに前回（あるいは複数回前）に求めた心筋負荷指数を保持し、表示を継続する。
【０１９０】
また、今回求めた循環動態パラメータと前回求めた循環動態パラメータとを比較し、各循環動態パラメータの変化率が予め定めた基準変化率以上である場合には、今回求めた循環動態パラメータに基づいて大動脈起始部血圧の算出並びに心筋負荷指数ＷＰの算出を行うものである。
【０１９１】
より具体的には、今回もとめた循環動態パラメータと前回（あるいは複数回前）に求めた循環動態パラメータとを比較し、各循環動態パラメータの変化率が±５［％］以上である場合には大動脈起始部血圧の算出及びこの算出した大動脈起始部血圧に基づく心筋負荷指数ＷＰの算出を行う。
【０１９２】
一方、各循環動態パラメータの変化率が±５［％］未満の場合には、大動脈起始部血圧の算出及び心筋負荷指数ＷＰの算出を行わず、前回（あるいは複数回前）に求めた心筋負荷指数を保持し、表示を継続する。
この結果、本第５実施形態によれば、不必要な演算を行う必要がなくなり、演算処理量を低減し、処理を簡略化して、全体的な処理速度の向上を図ることが可能となる。
【０１９３】
実施形態の変形例
本発明は上述した実施形態に限定されるものではなく、例えば以下のように種々の変形が可能である。例えば、１回拍出量ＳＶの測定を行うことなく循環動態パラメータを求める形態も考えられる。
すなわち、この実施形態によれば、循環動態パラメータのうちのインダクタンスＬは固定値とすることとして、被験者から測定した橈骨動脈脈波の波形のみに基づいて、その他の循環動態パラ メータの値を算出するようにする。このようにすれば、図１の構成において必要とされた１回拍出量測定器２を、図１５に示す如く省略することが可能となる。したがって、この実施形態における測定の態様は、図１６に示されるように、図２で必要とされたカフ帯Ｓ１が不要となっている。
【０１９４】
ところで、このようにインダクタンスＬの値を固定してしまうと、実測した１回拍出量を用いた方法に比して、得られる循環動態パラメータの精度が低下する。そこでこの点を補うため、図１７に示すように、測定により得られた橈骨動脈波形（測定波形）Ｗ１と計算により得られた橈骨動脈波形（計算波形）Ｗ２とを重ねて出力装置６に表示させる。そして、まず、インダクタンスＬの値を上記の固定値に設定して計算波形Ｗ２を求め、この波形を出力装置６に表示させて測定波形Ｗ１との波形の一致の程度を見る。次に、診断者が、上記の固定値とは異なる適当な値をインダクタンスＬとして決めて、再度、計算波形Ｗ２を求めて測定波形Ｗ１との一致の程度を出力装置６上で見る。そして、以後は、診断者が上記と同様にインダクタンスＬの値を幾つか適当に決めて、それぞれのインダクタンスＬの値について計算波形Ｗ２を求め、出力装置６上で計算波形Ｗ２の各々と測定波形Ｗ１とを比較する。そして、これらの計算波形Ｗ２の中で測定波形Ｗ１と最も良く一致する波形を一つ選んで、その時のインダクタンスＬの値を最適値として決定する。
【０１９５】
なお、大動脈起始部の圧波形のモデルとしては、上述した三角波の代わりに台形波を使用することが考えられる。このようにすると、三角波で近似する場合に比べて実際の圧波形により近い波形となるため、さらに正確な循環動態パラメータを算出することができる。
【０１９６】
また、脈波や１回拍出量の測定箇所は、図２や図１６に示す場所に限られるものではなく、被験者の体の如何なる部位であっても良い。すなわち、上述した実施形態では、被験者の上腕部にカフ帯Ｓ１を装着させた測定態様としたが、被験者の利便を考えるとカフ帯を使用しない形態が好ましいと言える。
【０１９７】
その一例として、手首において橈骨動脈波形と１回拍出量の双方を測定する形態が考えられる。この種の構成例としては、図１８に示すように、血圧測定用のセンサおよび１回拍出量測定用のセンサからなるセンサ１２を腕時計１１のベルト１３に装着するとともに、脈波解析装置のうちセンサ１２以外の構成部分１０を腕時計１１の本体部分に内蔵させた構成が考えられる。そして、図に示すように、センサ１２が取り付け具１４によってベルト１３へ摺動自在に取り付けられており、被験者が腕時計１１を手首にはめることで、センサ１２が適度な圧力で橈骨動脈部へ押し当てられるようになっている。
【０１９８】
また、指において脈波と１回拍出量とを測定する形態も考えられるのであって、この形態による装置の構成例を図１９に示す。同図に示すように、血圧測定用のセンサおよび１回拍出量測定用のセンサからなるセンサ２２を指（この図の例では人差し指）の根元に取り付けるとともに、脈波解析装置のうちセンサ２２以外の構成部分１０を腕時計２１に内蔵させてリード線２３，２３を介してセンサ２２へ接続してある。
【０１９９】
さらに、これら２つの測定形態を組み合わせることによって、手首において１回拍出量を測定するとともに指において脈波を測定する形態，指において１回拍出量を測定するとともに手首において橈骨動脈波を測定する形態を実現することが可能となる。
【０２００】
そして、これらの如くカフ帯なしの構成とすることで被験者が腕をまくらずに済み、測定にあたって被験者の負担が軽減される。
他方、カフ帯だけを用いた形態として図２０に示す構成が考えられる。同図に示すように、血圧測定用のセンサおよび１回拍出量測定用のセンサからなるセンサ３２と、脈波解析装置のうちセンサ３２以外の構成部分１０とを、カフ帯によって被験者の上腕部へ固定させており、図２と比較しても簡易な構成となっていることがわかる。
【０２０１】
また、上記の実施形態においては、循環動態パラメータを算出するにあたって脈波を用いることとしたが、これに限定されるものではなく、その他の生体の状態を用いることが可能なことは言うまでもない。
以上の実施形態においては、算出した循環動態パラメータに基づく大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成としていたが、生体の末梢部血圧あるいは生体の末梢部の脈波波形に基づいて、生体の大動脈起始部血圧の推定値を算出すれば、算出方法の如何を問わず、同様に大動脈起始部血圧の推定値及び検出した心拍数に基づいて心筋負荷指数を算出することが可能である。
【０２０２】
例えば、循環動態パラメータに基づいて算出した大動脈起始部血圧に代えて、生体の末梢部の脈波波形からＧＴＦ（Genelal Transfer Function）等の予め求めた所定の伝達関数に基づいて生体の大動脈起始部血圧の推定値を算出し、この算出した大動脈起始部血圧の推定値及び検出した心拍数に基づいて心筋負荷指数を算出するように構成することも可能である。
この場合において、所定の伝達関数としては、万人に適用可能な一般的な伝達関数に限らず、特定の生体に固有の補正を加えた伝達関数を用いることも可能である。
【０２０３】
実施形態の効果
本実施形態によれば、算出した循環動態パラメータに基づく大動脈起始部血圧の推定値及び検出した心拍数に基づいて心筋負荷指数を算出するので、末梢部血圧を用いて心筋負荷指数を算出する場合と比較して、より広範な条件下で、最適な心筋負荷指数を算出することが可能となる。
【０２０４】
また、心拍数の変動率が基準心拍数変動率以上の場合に、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出するので、常に大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成と比較して、より処理を簡略化することができ、処理の高速化を図ることができる。
【０２０５】
さらに心拍数の変動率が基準心拍数変動率未満の場合に末梢部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成によれば、正確な心筋負荷指数を得られるにも拘わらず、処理を簡略化することが可能となる。
さらにまた、算出した循環動態パラメータの基準循環動態パラメータに対する変動率が予め設定したパラメータ基準変動率以上の場合に、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成によれば、心筋負荷指数があまり変化しない状態においては、不必要に演算処理を行うことがないので、処理の簡略化を図ることができる。
【０２０６】
また、循環動態パラメータの変動率が基準パラメータ変動率未満の場合に末梢部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成によれば、正確な心筋負荷指数を得られるにも拘わらず、処理を簡略化することが可能となる。
さらに生体の末梢部血圧あるいは生体の末梢部の脈波波形に基づいて、生体の大動脈起始部血圧の推定値を算出し、生体の心拍数を検出し、大動脈起始部血圧の推定値及び検出した心拍数に基づいて心筋負荷指数を算出する構成によれば、簡易、正確、かつ、迅速に心筋負荷指数を算出することが可能となる。
【０２０７】
さらにまた、生体の末梢部の脈波波形及び所定の伝達関数に基づいて生体の大動脈起始部血圧の推定値を算出し、生体の心拍数を検出し、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成によれば、簡易、正確、かつ、迅速に心筋負荷指数を算出することが可能となる。
【０２０８】
【発明の効果】
本発明によれば、循環動態パラメータを算出するに際し、前記概略駆出期間を初期値として算出した左心室加圧時間を用いることにより、心電計が不要となるなど装置構成を簡略化でき、被験者の循環動態を反映したより正確な循環動態パラメータを算出することができるので、得られた循環動態パラメータに基づいて生体の大動脈起始部血圧の推定値をより正確に算出することができる。
【０２０９】
また、生体の末梢部血圧あるいは生体の末梢部の脈波波形に基づいて、生体の大動脈起始部血圧の推定値を算出し、生体の心拍数を検出し、大動脈起始部血圧の推定値及び検出した心拍数に基づいて心筋負荷指数を算出するので、簡易、正確、かつ、迅速に心筋負荷指数を算出することが可能となる。
【０２１０】
また、生体の末梢部の脈波波形及び所定の伝達関数に基づいて生体の大動脈起始部血圧の推定値を算出し、生体の心拍数を検出し、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出するので、簡易、正確、かつ、迅速に心筋負荷指数を算出することが可能となる。
【０２１１】
さらに、算出した循環動態パラメータに基づく大動脈起始部血圧及び心電図などにより別個に検出した心拍数あるいは脈波波形から検出した心拍数に基づいて心筋負荷指数を算出するので、末梢部血圧を用いて心筋負荷指数を算出する場合と比較して、より広範な条件下で、最適な心筋負荷指数を算出することが可能となる。
【０２１２】
また心筋負荷指数算出手段は、心拍数の変動率が基準心拍数変動率以上の場合に、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出するので、常に大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出する構成と比較して、より処理を簡略化することができ、処理の高速化を図ることができる。
【０２１３】
さらにまた、心筋負荷指数算出手段は、心拍数の変動率が基準心拍数変動率未満の場合に末梢部血圧及び検出した心拍数に基づいて心筋負荷指数を算出するので、正確な心筋負荷指数を得られるにも拘わらず、処理を簡略化することが可能となる。
【０２１４】
また心筋負荷指数算出手段は、算出した循環動態パラメータの基準循環動態パラメータに対する変動率が予め設定したパラメータ基準変動率以上の場合に、大動脈起始部血圧及び検出した心拍数に基づいて心筋負荷指数を算出するので、心筋負荷指数があまり変化しない状態においては、不必要に演算処理を行うことがないので、処理の簡略化を図ることができる。
【０２１５】
さらに心筋負荷指数算出手段は、循環動態パラメータの変動率が基準パラメータ変動率未満の場合に末梢部血圧及び検出した心拍数に基づいて心筋負荷指数を算出するので、正確な心筋負荷指数を得られるにも拘わらず、処理を簡略化することが可能となる。
【図面の簡単な説明】
【図１】 本発明の第１実施形態による脈波解析装置の構成を示すブロック図である。
【図２】 第１実施形態における脈波検出装置，１回拍出量測定器を用いた測定態様を示す図である。
【図３】 （ａ）は、人体の動脈系をモデル化した四要素集中定数モデルを示す回路図、（ｂ ）は同じく五要素集中定数モデルを示す回路図である。
【図４】 左心室圧波形と大動脈起始部の血圧波形とを示す図である。
【図５】 大動脈起始部の血圧波形をモデル化した波形を示す図である。
【図６】 第１実施形態における脈波解析装置の動作の概要を示すフローチャートである。
【図７】 第１実施形態における脈波解析装置のパラメータ算出処理の動作を示すフローチャートである。
【図８】 第１実施形態における脈波解析装置のパラメータ算出処理の動作を示すフローチャートである。
【図９】 第１実施形態における脈波解析装置のα，ω算出処理の動作を示すフローチャートである。
【図１０】 第１実施形態における脈波解析装置のω算出処理の動作を示すフローチャートである。
【図１１】 第１実施形態における脈波解析装置の平均化処理により得られた橈骨動脈波形を例示する波形図である。
【図１２】 第１実施形態における脈波解析装置の演算処理により得られた橈骨動脈波形と平均化処理により得られた橈骨動脈波形とを重ね表示した波形図である。
【図１３】 第１実施形態における脈波解析装置の平均化処理により得られた橈骨動脈波形へ適用する正規化の処理内容を説明する図である。
【図１４】 左心室加圧時間ｔsと１拍の時間ｔpとの相関を示す図である。
【図１５】 第２実施形態による脈波解析装置の構成を示すブロック図である。
【図１６】 第２実施形態における脈波検出装置を用いた測定態様を示す図である。
【図１７】 第２実施形態において、出力装置に表示される橈骨動脈波の測定波形と計算波形の重ね表示を示す図である。
【図１８】 センサを除く脈波解析装置の構成部分を腕時計に内蔵させ、センサを腕時計のバンドへ装着させた形態の斜視図である。
【図１９】 センサを除く脈波解析装置の構成部分を腕時計に内蔵させ、センサを指の根元に装着させた形態の斜視図である。
【図２０】 センサを除く脈波解析装置の構成部分とセンサをカフ帯によって上腕部に取り付けた形態の構成図である。
【図２１】 本発明の第３実施形態による血圧測定装置の構成を示すブロック図である。
【図２２】 第１のタイプの脈波における大動脈圧波形（点線）と橈骨動脈波形（実線）の関係を表わす図である。
【図２３】 、第２のタイプの脈波における大動脈圧波形（点線）と橈骨動脈波形（実線）の関係を表わす図である。
【図２４】 第３のタイプの脈波における大動脈圧波形（点線）と橈骨動脈波形（実線）の関係を表わす図である。
【図２５】 中枢部血管抵抗Ｒcとひずみ率ｄの関係を表わす図である。
【図２６】 末梢部血管抵抗Ｒpとひずみ率ｄの関係を表わす図である。
【図２７】 血流による慣性Ｌとひずみ率ｄの関係を表わす図である。
【図２８】 コンプライアンスＣとひずみ率ｄの関係を表わす図である。
【図２９】 収縮期面積法を説明した図である。
【図３０】 従来の技術による血圧測定装置の構成を示す図である。
【図３１】 静電容量Ｃcを求めるプログラムのフローチャートである。
【図３２】 静電容量Ｃcを求める他のプログラムのフローチャートである。
【図３３】 脈波波形の説明図である。
【符号の説明】
１ 脈波検出装置
２ １回拍出量測定器
３ Ａ／Ｄ変換器
４ マイクロコンピュータ
５ キーボード
６ 出力装置
７ モニタ
６１ 実測血圧表示部
６２ 中枢部推定血圧表示部
６３ 警告表示部
６４ パラメータ表示部
６５ プリンタ
６６ プリント指令ボタン
６７ ＣＲＴディスプレイ
６８ 心筋負荷指数表示部[0001]
BACKGROUND OF THE INVENTION
  The present invention relates to a biological state measuring device suitable for measuring the state of a human body as a living body.InIn particular, a biological state measuring device suitable for calculating a myocardial load indexInRelated.
[0002]
[Prior art]
When diagnosing the state of the human circulatory system, blood pressure, heart rate, etc. are most commonly used.
However, in order to perform more detailed diagnosis, it is necessary to measure so-called circulatory dynamic parameters such as viscous resistance and compliance of blood vessels.
[0003]
By the way, when these circulatory dynamic parameters are modeled and expressed, a four-element lumped constant model is used as a model for describing the behavior of the arterial system.
On the other hand, in order to measure the circulatory dynamic parameters, it is necessary to measure the pressure waveform and blood flow volume at the aortic origin and notch. That is, a method of directly measuring by inserting a catheter into an artery is employed, or a method of indirectly measuring with an ultrasonic wave or the like.
[0004]
The above-described method of directly measuring by inserting a catheter into a pulse is an invasive measurement, so that there is a problem that the burden on the subject is large and the apparatus becomes large.
On the other hand, the method of measuring indirectly with ultrasonic waves can observe blood flow in blood vessels non-invasively and can reduce the burden on the subject. There was also a problem that the equipment for this would also be large.
[0005]
Accordingly, the present inventor has found a method of approximately calculating the parameters of the four-element lumped constant model by measuring the pulse waveform of the radial artery and the stroke volume. And by using this method, a pulse wave analysis device has been proposed that enables non-invasive and easy evaluation of circulatory dynamic parameters (Japanese Patent Laid-Open No. 6-205747, title of the invention: pulse wave analysis device). ).
[0006]
By the way, in the method of approximately calculating the parameters of the four-element lumped constant model by measuring the pulse waveform of the radial artery and the stroke volume, the compliance of the blood vessels is determined between the central part and the peripheral part of the arterial system. The model that is handled separately is not adopted. Therefore, when a hemodynamic agonist is administered to a patient during exercise or when a hemodynamic agonist is administered to a patient, the effect cannot be evaluated separately for the central part and the peripheral part.
[0007]
Next, the blood pressure measurement described above will be briefly described.
A conventional non-invasive blood pressure measuring device generally used has a cuff (arm band) attached to a subject's upper arm, etc., and pressure is applied to the cuff to detect the subject's pulse wave and blood pressure value Is measuring.
Thus, as an apparatus for measuring blood pressure in the peripheral part of a subject, for example, JP-A-4-276234 can be cited. That is, as shown in FIG. 29, the cuff 110 is wound around and attached to the upper arm of the subject's upper limb, the band 138 is wound around the wrist 140, and the pulse wave sensor 134 is brought into close contact with the subject's radial artery, The subject's pulse wave is detected. Then, after pressurizing the cuff 110, the systolic blood pressure value and the diastolic blood pressure value are measured by a well-known oscillometric method at the time of pressure reduction.
[0008]
By the way, when the blood pressure value on the central side and the blood pressure value on the peripheral side in the arterial system of the human body are actually measured, there is a difference between the blood pressure values on the central side and the peripheral side, particularly regarding the maximum blood pressure value. Moreover, the degree of this difference varies depending on the shape of the pulse wave observed on the peripheral side.
22 to 24 show how the blood pressure value varies depending on the shape of the pulse wave.
[0009]
In these figures, the central aortic pressure waveform and the maximum / minimum blood pressure value, and the radial artery pressure waveform and the maximum / minimum blood pressure value which are on the peripheral side are shown.
In the case of the first type of pulse waveform shown in FIG. 22, the maximum blood pressure values obtained from the aortic pressure waveform indicated by the dotted line and the radial artery waveform indicated by the solid line are substantially equal, although the radial artery side is slightly higher. Good.
[0010]
However, in the case of the second type pulse wave waveform shown in FIG. 23, the maximum blood pressure difference is 14.9 mmHg, which is considerably higher than that in the case of the first type pulse wave waveform shown in FIG. It gets bigger.
Furthermore, in the case of the third type of pulse waveform shown in FIG. 24, the maximum blood pressure difference is further increased to 26.1 mmHg, and on the contrary to the first or second type of pulse waveform, the aortic pressure waveform is increased. Overall exceeds the radial artery waveform.
[0011]
Incidentally, according to these FIG. 22 to FIG. 24, it can be seen that the minimum blood pressure value on the radial artery side is substantially the same regardless of the shape of the pulse wave.
Here, the first to third types of pulse waves described above will be briefly described.
The first type of pulse waveform is that of a normal healthy person, and the waveform is relaxed and relaxed, with a constant rhythm and less disturbance.
[0012]
In addition, the second type of pulse wave waveform is characterized in that it quickly rises after rising rapidly, the aortic incision is deeply cut, and the subsequent lagging peak is considerably higher than usual.
The third type of pulse wave waveform is characterized in that it rises sharply and does not fall immediately thereafter, and a high blood pressure state lasts for a certain period of time.
[0013]
It is derived from these illustrated figures that the blood pressure value at the origin of the aorta, that is, the central side, may be low even if the peripheral blood pressure value such as the rib or the upper arm is high, and conversely, the peripheral blood pressure value is low. Even if the blood pressure value on the side is low, the blood pressure value on the central side may be high. Such a relationship varies depending on the shape of the pulse wave waveform, and furthermore, these relationships appear in the shape of the pulse wave waveform.
[0014]
For example, suppose that a hypotensive agent is administered to a patient for the treatment of hypertension, and the effect of the drug is observed based on the blood pressure in the radial artery. In such a case, even if the blood pressure measured on the peripheral side decreases, the blood pressure on the central side may not actually decrease.
Therefore, it can be said that it may be difficult to correctly grasp the drug effect only from the blood pressure on the peripheral side.
[0015]
On the other hand, even if there is no change in the blood pressure on the peripheral side, if the aortic pressure waveform changes and the blood pressure on the central side decreases, the burden on the heart actually decreases. That is why. In such a case, even if the blood pressure on the peripheral side is not lowered forcibly, the effect of the medicine is sufficiently manifested, but it is difficult to judge this only from the blood pressure on the peripheral side.
[0016]
By the way, a myocardial load index (W-Product) has been used as an index for estimating the degree of load on the myocardium.
The myocardial load index WP is expressed as follows, assuming that the peripheral blood pressure is Pperi and the heart rate is HR.
WP = Pperi × HR
[0017]
[Problems to be solved by the invention]
As described above, even when the blood pressure value on the peripheral side is high, the blood pressure value on the aortic root, that is, the central side may be low. Further, on the contrary, even when the peripheral blood pressure value is low, the central blood pressure value may be high, and the peripheral blood pressure value is not necessarily linked to the central blood pressure value.
[0018]
By the way, the myocardial load index should be used as an index to see how much the heart burden is originally, but it is a blood pressure value (systolic blood pressure) measured on the peripheral side as is conventionally done. ) Based on), the heart burden may be overestimated, and conversely, it may be underestimated.
[0019]
  Accordingly, a first object of the present invention is to provide a biological state measurement device capable of accurately estimating central blood pressure with a simple configuration.TheIt is to provide.
[0020]
  Also, a second object of the present invention is a biological state measuring device capable of calculating a more accurate myocardial load index WP corresponding to an actual myocardial load.TheIt is to provide.
[0026]
[Means for Solving the Problems]
  To solve the above problem,Claim1The described configuration includes a measuring means for measuring the state of the living body based on a pulse waveform at a peripheral part of the living body, and a circulation of the arterial system from the central part of the living body to the peripheral part based on the state of the living body. Analysis means for calculating a circulatory dynamic parameter including viscoelasticity of the aorta as a circulatory dynamic parameter representing dynamics, and an aortic blood pressure calculating means for calculating an estimated value of the aortic origin blood pressure of the living body based on the circulatory dynamic parameter; , Heart rate detection means for detecting the heart rate of the living body, myocardial load index calculation means for calculating a myocardial load index based on the estimated value of the aortic origin blood pressure and the detected heart rate, and the detected heart rate Discriminating means for discriminating whether or not the fluctuation rate with respect to the reference heart rate, which is the number of resting heart rates, exceeds a preset reference heart rate fluctuation rateAnd comprisingThe myocardial load index calculating means calculates a myocardial load index based on the aortic origin blood pressure and the detected heart rate when the fluctuation rate is equal to or higher than the reference heart rate fluctuation rate based on the determination. It is characterized by.
  According to a second aspect of the present invention, in the configuration of the first aspect, there is provided peripheral blood pressure detecting means for noninvasively detecting the peripheral blood pressure of the living body, and the myocardial load index calculating means is configured to perform the determination. The myocardial load index is calculated based on the peripheral blood pressure and the detected heart rate when the fluctuation rate is less than the reference heart rate fluctuation rate.
  The structure according to claim 3 is a measuring means for measuring the state of the living body based on a pulse waveform at a peripheral portion of the living body, and an artery extending from the central portion of the living body to the peripheral portion based on the state of the living body. Analysis means for calculating a circulatory dynamic parameter including viscoelasticity of the aorta as a circulatory dynamic parameter representing the circulatory dynamics of the system, and an aortic blood pressure for calculating an estimated value of the aortic origin blood pressure of the living body based on the circulatory dynamic parameter Calculating means; heart rate detecting means for detecting a heart rate of the living body; and myocardial load index calculating means for calculating a myocardial load index based on the estimated value of the aortic origin blood pressure and the detected heart rate. The myocardial load index calculating means has a variation rate with respect to a reference hemodynamic parameter that is the hemodynamic parameter at a predetermined timing of the calculated hemodynamic parameter in advance. When boss was above parameters reference variation rate, it is characterized by calculating the cardiac muscle load index based on the heart rate the aortic root and the blood pressure and the detection.
  According to a fourth aspect of the present invention, in the configuration of the third aspect, the peripheral blood pressure detecting means for noninvasively detecting the peripheral blood pressure of the living body is provided, and the myocardial load index calculating means includes the discrimination The myocardial load index is calculated based on the peripheral blood pressure and the detected heart rate when the variation rate is less than the reference parameter variation rate.
[0027]
  Claim5The described configuration is claimed.5. The method according to any one of claims 1 to 4.In the structure, the hemodynamic parameters include blood vessel resistance due to blood viscosity at the central part, blood inertia, blood vessel resistance at the peripheral part, and viscoelasticity of the blood vessel at the peripheral part. .
[0028]
  Claim6The described configuration is defined in claims 1 to 5.5In the configuration described in any of the above, the blood pressure calculation means includes a diode corresponding to the aortic valve, a first resistance corresponding to vascular resistance due to blood viscosity in the central portion, and inertia of blood in the central portion. The first capacitance corresponding to the viscoelasticity of the aorta, the second resistance corresponding to the vascular resistance in the peripheral portion, and the viscoelasticity of the blood vessel in the peripheral portion. A model having a second capacitance, wherein a series circuit of the diode and the first capacitance is connected between a pair of input terminals, and the second capacitance and a pair of output terminals are connected between the pair of output terminals. A parallel circuit composed of the second resistor is inserted, and a series circuit composed of the first resistor and the inductance is inserted between both terminals of the first capacitance and the output terminal. The element lumped constant model By modeling the circulatory state of the pulse system, and determines the circulatory state parameters, the voltage waveform between the terminals of the first capacitance and the aortic pressure waveform, is characterized by.
[0029]
  Claim7The described configuration is defined in claims 1 to 5.6In the configuration according to any one of the above, the state of the living body is a pulse wave in the peripheral part of the arterial system, and the blood pressure calculating unit applies an electrical signal corresponding to the left ventricular pressure of the living body between the input terminals. The value of each element constituting the five-element lumped constant model is determined so that an electric signal corresponding to the waveform of the pulse wave is obtained from the output terminal.
[0030]
  Claim8The described configuration is defined in claims 1 to 5.6In the configuration according to any one of the above, the state of the living body is a pulse wave in a peripheral part of the arterial system, and has strain calculating means for calculating a distortion of the pulse wave from the waveform of the pulse wave, and the blood pressure calculation The means determines the hemodynamic parameter based on the correlation between the hemodynamic parameter and the distortion of the pulse wave.
[0031]
  Claim9The described configuration is defined in claims 1 to 5.8In the configuration according to any one of the above, there is provided a stroke volume detecting means for detecting a stroke volume of the living body, and the blood pressure calculating means is configured to calculate a stroke volume obtained from the aortic pressure waveform. The value of the hemodynamic parameter is adjusted so that the calculated value and the measured value of the stroke volume measured by the stroke volume measuring means coincide with each other.
[0032]
  Claim10The described configuration is defined in claims 1 to 5.9In the configuration described in any one of the above, it is characterized by having a work amount calculating means for calculating the work amount of the heart of the living body based on the aortic pressure waveform.
[0044]
DETAILED DESCRIPTION OF THE INVENTION
Next, preferred embodiments of the present invention will be described with reference to the drawings.
First embodiment
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.
[0045]
[1] Outline configuration of blood pressure measurement device
FIG. 1 shows a configuration block diagram of a blood pressure measurement device as a biological state measurement device according to the first embodiment.
In the first embodiment, the hemodynamic parameters of the arterial system of the human body are evaluated based on the information obtained from the human body by the noninvasive sensor, and the systolic blood pressure and the minimum blood pressure in the central part are evaluated based on the obtained hemodynamic parameters. The blood pressure, myocardial load index, and cardiac work are calculated. The specific contents of the hemodynamic parameters will be described later.
[0046]
As shown in FIG. 2, the pulse wave detection device 1 detects the radial artery waveform via a pressure sensor S2 attached to the wrist of the subject, and the subject via a cuff band S1 attached to the upper arm of the subject. Detect blood pressure.
Then, the measured radial artery waveform is calibrated by blood pressure and output as an analog electric signal. This analog signal is input to an A / D (analog / digital) converter 3 and converted into a digital signal at every predetermined sampling period.
[0047]
As shown in FIG. 2, the stroke volume measuring device 2 is connected to a cuff belt S1, and is the amount of blood that flows out of the heart in one stroke through the cuff belt S1. The stroke volume is measured, and the measurement result is output as a stroke volume data as a digital signal. As this type of measuring instrument, an apparatus for measuring by a so-called systolic area method can be used.
[0048]
The microcomputer 4 has a built-in waveform memory for storing the pulse waveform acquired from the A / D converter 3 and a temporary storage memory as a work area.
The microcomputer 4 performs various processes as shown in FIG. 6 according to commands input from the keyboard 5 as an input device, and outputs the results obtained from these processes to the output device 6.
[0049]
Here, only the outline of the processing will be described, and details of those processing will be described in detail when the operation is described.
(1) Pulse wave measurement data reading process (step S1)
a) The time-series digital signal of the radial artery waveform obtained via the A / D converter 3 is taken into a built-in waveform memory (not shown).
b) The radial artery waveform taken into the waveform memory is averaged for each “beat” to obtain a radial artery waveform corresponding to one beat (hereinafter referred to as an average waveform).
[0050]
(2) Measurement data reading process for stroke volume (step S2)
The stroke volume data is taken into a temporary storage memory built in the microcomputer 4.
(3) Parameter calculation processing (step S3)
A mathematical expression representing the radial artery waveform corresponding to one beat is obtained, and each parameter of the electrical model corresponding to the arterial system is calculated based on this mathematical expression.
[0051]
(4) Data calculation process (step S4)
From the obtained circulatory dynamic parameters, a pulse wave waveform at the aortic origin is obtained, and a maximum blood pressure value, a minimum blood pressure value, a myocardial load index WP, and a heart workload are calculated at the aortic origin.
(5) Output process (step S5)
The obtained circulatory dynamic parameters, systolic blood pressure value, diastolic blood pressure value, myocardial load index WP, and cardiac work are output to the output device 6.
[0052]
[2] Detailed configuration of output device
Next, details of the output device 6 will be described with reference to FIG.
The measured blood pressure display unit 61 displays the highest blood pressure, the lowest blood pressure, and the average blood pressure that are actually measured based on the radial artery waveform.
[0053]
The central part estimated blood pressure display unit 62 displays the average blood pressure E01, the maximum blood pressure Em ', and the minimum blood pressure Eo of the central part obtained by the processing described later.
The warning display unit 63 is configured by a plurality of LEDs arranged in a horizontal row, and these LEDs are lit according to the difference between the measured systolic blood pressure and the central systolic blood pressure Em '.
[0054]
That is, if the difference between the two is ± 10 mmHg or less, the “NORMAL” green LED is turned on, and if the difference exceeds ± 10 mmHg, the “CAUTION” red LED is turned on.
[0055]
The parameter display unit 64 receives the electrostatic capacitance Cc, electrical resistance Rc, inductance L, electrostatic capacitance C, electrical resistance Rp, left ventricular pressurization time ts, time of one beat tp, stroke volume SV from the microcomputer 4. When the mental work Ws is supplied, these parameters are displayed. Details of these parameters will be described later.
The CRT display 67 displays various waveforms such as a radial artery waveform, a left ventricular pressure waveform, and an aortic pressure waveform.
[0056]
When the print command button 66 is pressed, the printer 65 displays various data displayed on the measured blood pressure display unit 61, the central part estimated blood pressure display unit 62, the warning display unit 63, the parameter display unit 64, and the CRT display 67. Print out the displayed waveform on paper.
[0057]
The myocardial load index display unit 68 displays the myocardial load index WP obtained by the processing described later.
Here, the significance of warning display in the warning display unit 63 will be described.
[0058]
As described above with reference to FIGS. 22 to 24, there are three types of the maximum blood pressure difference between the estimated aortic pressure waveform and the radial artery waveform. The subject whose pulse waveform is the first type (FIG. 22) is likely to be a healthy person, and in the second and third types, the subject often has some disease.
[0059]
For example, the second type (FIG. 23) is caused by an abnormal blood flow state, and is likely to have diseases such as edema, hepatorenal disease, respiratory disease, gastrointestinal disease, and inflammatory disease. The third type is caused by an increase in the tension of the blood vessel wall, and is highly likely to have hepatobiliary disease, skin disease, hypertension, painful disease and the like.
[0060]
Therefore, in the present embodiment, when it is considered that the maximum blood pressure difference is abnormal, a red LED is turned on to display a warning.
In the above example, the diagnosis is performed based on the maximum blood pressure difference between the aortic pressure waveform and the radial artery waveform, but the minimum blood pressure difference or the average blood pressure difference may be used instead of the maximum blood pressure difference. Furthermore, it goes without saying that the diagnosis may be performed using all of the maximum blood pressure difference, the minimum blood pressure difference, and the average blood pressure difference.
[0061]
[3] About the five-element lumped constant model
In the first embodiment, a “five element lumped constant model” is newly adopted as an electrical model of the arterial system.
Among the factors that determine the behavior of the circulatory system of the human body, this five-element lumped constant model is adopted in the four-element lumped constant model disclosed in Japanese Patent Laid-Open No. 6-205747 (name of invention: pulse wave analyzer). In addition to the four parameters of blood inertia in the central part, vascular resistance (viscosity resistance) due to blood viscosity in the central part, vascular compliance (viscoelasticity) in the peripheral part, vascular resistance (viscosity resistance) in the peripheral part, Aortic compliance is added as a parameter, and these five parameters are modeled as an electric circuit. The compliance is an amount representing the softness of the blood vessel.
[0062]
FIG. 3 (a) shows a circuit diagram of a four-element lumped constant model, and FIG. 3 (b) shows a circuit diagram of a five-element lumped constant model. The correspondence between each element and each parameter constituting the five-element lumped constant model is shown below.
Capacitance Cc: Aortic compliance [cm5 / dyn]
Electrical resistance Rc: Vascular resistance due to blood viscosity at the central part of the arterial system [dyn · s / cm5]
Inductance L: Blood inertia in the central part of the arterial system [dyn · s2 / cm5]
Capacitance C: Blood vessel compliance in the peripheral part of the arterial system [cm5 / dyn]
Electrical resistance Rp: Vascular resistance due to blood viscosity at the peripheral arterial system [dyn · s / cm5]
[0063]
Here, the currents i, ip, ic, and is flowing through each part in the electric circuit correspond to the blood flow [cm3 / s] flowing through each corresponding part. In particular, the current i is the aortic blood flow, and the current is is the blood flow pumped from the left ventricle. The input voltage e corresponds to the left ventricular pressure [dyn / cm2], and the voltage v1 corresponds to the pressure at the aortic origin [dyn / cm2]. Further, the terminal voltage vp of the capacitance C corresponds to the pressure [dyn / cm2] at the radial artery. In addition, the diode D shown in FIG. 3B corresponds to an aortic valve, and is on (valve opened) during a period corresponding to the systole and off (valve) during the period corresponding to the diastole. Is closed).
[0064]
As will be described later, in the first embodiment, these five parameters are not calculated at once, but parameters other than the capacitance Cc are calculated using the four-element lumped constant model disclosed in the above-mentioned document. After using and calculating, the capacitance Cc is determined. Therefore, first, a theoretical explanation will be given on the behavior of the four-element lumped constant model shown in FIG.
[0065]
In the four-element lumped constant model shown in FIG.
v1 = Rci + L (di / dt) + vp (1)
Here, the current i is
i = ic + ip = C (dvp / dt) + (vp / Rp) (2)
Therefore, equation (1) can be transformed into the following equation.
v1 = LC (d2vp / dt2) + {RcC + (L / Rp)} (dvp / dt) + {1+ (Rc / Rp)} vp (3)
[0066]
As is well known, the general solution of the second-order constant coefficient ordinary differential equation expressed by equation (3) is a special solution (stationary solution) that satisfies equation (3) and a transient solution that satisfies the following equation: Given by the sum of
0 = LC (d2vp / dt2) + {RcC + (L / Rp)} (dvp / dt) + {1+ (Rc / Rp)} vp (4)
[0067]
Next, the solution of the differential equation (4) is obtained as follows. First, a damped oscillation waveform represented by the following equation is assumed as a solution of the differential equation (4).
vp = exp (st) (5)
Substituting equation (5) into equation (4), equation (4) is transformed as follows.
[LCs2 + {RcC + (L / Rp)} s + {1+ (Rc / Rp)}] vp = 0 (6)
[0068]
Solving equation (6) for s,
s = [-{RcC + (L / Rp)} ± √ {{RcC + (L / Rp)} 2-4LC {1+ (Rc / Rp)}}] / 2LC (7)
It becomes.
In equation (7),
{RcC + (L / Rp)} 2 <4LC {1+ (Rc / Rp)} (8)
, The second term root sign √ is negative, and s is as follows.

[0069]
Here, the attenuation rate is α, the angular frequency is ω,

It is. And
A1 = LC (12)
A2 = (L + RcRpC) / Rp (13)
A3 = (Rc + Rp) / Rp (14)
In other words, equations (10) and (11) can be expressed as follows.
α = (A2 / 2A1) (15)
ω = √ {(A3 / A1) -α2} (16)
[0070]
In this way, the value of s is determined, and a solution satisfying the differential equation (4) is obtained.
Based on the above knowledge, Expression (5) can be used as an expression for approximating the damped vibration component included in the response waveform of the four-element lumped constant model.
Next, the pressure waveform at the aortic root is modeled. In general, the pressure waveform at the beginning of the aorta is a waveform as shown by the thick line in FIG. 4, where time tp is the time for one beat of the waveform and time ts is the pressurization time of the left ventricle. In the four-element lumped constant model, this pressure waveform is approximated by a triangular wave shown in FIG. In FIG. 5, if the amplitude and time of the approximate waveform are represented by Eo, Em, tp, and tp1, the aortic pressure v1 at an arbitrary time t is represented by the following equation. Here, Eo is the lowest blood pressure (diastolic blood pressure), Em is the pulse pressure, (Eo + Em) is the highest blood pressure (systolic blood pressure), tp is the time of one beat, tp1 is the lowest blood pressure value from the rise of the aortic pressure It is time to become.
Section where 0 ≦ t <tp1:
v1 = Eo + Em {1- (t / tp1)} (17)
tp1 ≦ t <tp interval:
v1 = Eo (18)
[0071]
The response waveform vp (namely, radial artery wave) when the voltage v1 represented by the equations (17) and (18) is input to the equivalent circuit of FIG. 3 (a) is as follows.
Section where 0 ≦ t <tp1:

tp1 ≦ t <tp interval:

[0072]
Here, Emin is the lowest blood pressure value in the radial artery waveform measured by the pulse wave detection device 1 (see FIG. 11 described later).
The third term on the right side in Equation (19) and the second term on the right side in Equation (20) are the damped oscillation components of Equation (5) described above, and α and ω in these terms are Equations (15) and ( 16). B, tb, Dm1, and Dm2 are constant values calculated according to the procedure described later.
[0073]
Next, of the constants in the equations (19) and (20), those other than the already determined α and ω will be examined. First, when the equations (17) and (19) are substituted into the differential equation (3), the following equation is obtained.

[0074]
The following conditions are necessary for the expression (21) to hold.

[0075]
Equations (24) and (25) constrain α and ω, but α and ω already obtained by equations (15) and (16) satisfy these equations.
On the other hand, when the equations (18) and (20) are substituted into the differential equation (3), the following equation is obtained.

[0076]
In order for the expression (26) to be satisfied, in addition to the expressions (24) and (25) being satisfied, the following expression must be satisfied.
Eo = {1+ (Rc / Rp)} Emin = A3 Emin (27)
Next, the constants of the equations (19) and (20) are calculated based on the conditional equations (22) to (25) and the equation (27) for establishing the differential equation (3).
[0077]
First, Emin is obtained from the equation (27) as follows.
Emin = (EO / A3) (28)
Also, from the equation (23), B is
B = (tbEm) / (tp1A3) (29)
It becomes. Also, substituting equation (29) into equation (22) and solving for tb,
tb = (tp1A3 + A2) / (A3) (30)
[0078]
Further, the remaining constants Dm1, Dm2, θ1, and θ2 are values such that the radial artery waveform vp can maintain continuity at t = 0, tp1, and tp, that is, the conditions (1) to (4) shown below. A value that satisfies is selected.
(1) vp (tp1) in equation (19) matches vp (tp1) in equation (20)
(2) vp (tp) in equation (20) matches vp (0) in equation (19)
(3) The differential coefficient of t = tp1 in equation (19) and equation (20) must match.
(4) The differential coefficient at t = 0 in equation (19) matches the differential coefficient at t = tp in equation (20).
[0079]
That is, Dm1 and θ1 are
Dm1 = √ (D112 + D122) / ω (31)
θ1 = tan −1 (D11 / D12) (32)
Is chosen. However,
D11 = (vO1−B−Emin) ω (33)
D12 = (vO1-B-Emin) [alpha] + (B / tp) + (iO1 / C) (34)
VO1 and iO1 are initial values of vp and ic at t = 0.
[0080]
Dm2 and θ2 are
Dm2 = √ (D212 + D222) / ω (35)
θ2 = tan −1 (D21 / D22) (36)
Is chosen. However,
D21 = (vO2−Emin) · ω (37)
D22 = (vO2-Emin) .alpha. + (IO2 / C) (38)
VO2 and iO2 are initial values of vp and ic at t = tp1.
[0081]
In this way, the constants of Equation (19) and Equation (20) were obtained.
By calculating backward from the angular frequency ω in equation (16), the vascular resistance RC is
Rc = [L-2Rp√ {LC (1-ω2LC)}] / CRp (39)
It becomes. Here, the condition that Rc is real and positive is:
{4Rp2C} / {1+ (2ωRpC) 2} ≦ L ≦ (1 / ω2C) (40)
It is.
[0082]
In general, Rp is about 10 3 [dyn · s / cm 5], C is about 10 −4 [cm 5 / dyn], and ω is 10 (rad /) because it is the angular frequency of the vibration component superimposed on the pulse wave. s) It may be considered that the above is true. For this reason, the lower limit of the equation (40) can be regarded as approximately 1 / (ω2C). Therefore, for simplicity, L is approximately
L = 1 / (ω2C) (41)
Rc is
Rc = L / (CRp) (42)
It becomes.
[0083]
In addition, from the relationship between Equation (41) and Equation (42), the attenuation constant α in Equation (15) is
α = 1 / (CRp) (43)
It becomes.
Using the relations of the equations (41) to (43), the remaining parameters of the four-element lumped constant model are expressed by α, ω, and L.
Rc = αL (44)
Rp = (ω2L / α) (45)
C = 1 / (ω2L) (46)
It becomes. From these equations (44) to (46), it is clear that the parameters are determined by obtaining α, ω, and L.
[0084]
Here, as will be described later, α, ω, B, and tb are obtained from the actually measured waveform of the radial artery wave, and L can be calculated based on the stroke volume SV. A procedure for calculating L based on the stroke volume SV will be described below.
First, the average value E01 of the pressure wave at the aortic root is given by the following equation.
E01 = {Eotp + (tp1Em / 2)} / tp (47)
On the other hand, the following equation holds between Rc, Rp, α, ω, and L.
Rc + Rp = αL + (ω2L / α) = (α2 + ω2) L / α (48)
[0085]
The average current flowing through the four-element lumped constant model, that is, the average value E01, divided by (Rc + Rp) corresponds to the average value (SV / tp) of the blood flow flowing through the artery due to pulsation. Is established.

[0086]
By solving the equation (49) thus obtained for L, an equation for obtaining L from the stroke volume SV is obtained as follows.
L = α · {Eotp + (tp1Em / 2)} / {(α2 + ω2) SV} (50)
A value corresponding to the average current (1 / tp) {Eotp + (tp1Em / 2)} in equation (49) is obtained by measuring the blood flow volume, and the inductance L may be calculated based on this result. . As an apparatus for measuring a blood flow rate, an impedance method, a Doppler method, and the like are known. In addition, blood flow measurement devices using the Doppler method include those using ultrasonic waves and those using lasers.
[0087]
[4] Explanation of the principle of the calculation method of hemodynamic parameters
Next, the principle of the calculation method of the circulatory dynamic parameter based on the five-element lumped constant model will be described. As mentioned above, since Rc, Rp, C, and L in the circulation dynamic parameters are determined using a four-element lumped constant model, the value of the capacitance Cc is determined based on these parameters. To do. Therefore, it is necessary to obtain the current i, current is, voltage v1, voltage vp, etc. in FIG.
[0088]
First, the left ventricular pressure waveform e is approximated by a sine wave as shown in FIG. That is,
ωs = π / ts
The left ventricular pressure waveform e is expressed by the following equation.
e = Em ′ sin ωst (51)
Here, Em ′ is the systolic blood pressure and corresponds to (Em + Eo) in FIG. Hereinafter, as shown in FIG. 4, the time t is divided into a systole when t1 ≦ t <t2 and a diastole when t2 ≦ t <(tp + t1). Here, time t1 and time t2 are times at the intersection of the left ventricular pressure waveform and the aortic pressure waveform.
[0089]
[4.1] systole
In this case, v1 = e holds, and equations (1) and (2) hold for voltage v1 and current i, respectively. Therefore, the following differential equation is established from the equations (1) to (3) and the equations (12) to (14) and (51).
A1 (d2vp / dt2) + A2 (dvp / dt) + A3vp = Em'sinωst (52)
[0090]
First, the steady solution vpst of this differential equation is obtained in the same manner as in the four-element lumped constant model. For this purpose, the steady solution vpst is assumed as follows.
vpst = E1 cos ωst + E2 sin ωst (53)
By substituting equation (53) into vp of equation (52) and comparing the coefficients, the following two equations are obtained.
(A3 ・ ωs2A1) E1 ＋ ωsA2E2 = 0 (54)
−ωsA2E1 + (A3−ωs2A1) E2 = Em ′ (55)
[0091]
By solving these equations,
E1 = (− ωsA2Em ′) / {(ωsA2) 2+ (A3−ωs2A1) 2} (56)
E2 = {(A3−ωs2A1) Em ′} / {(ωsA2) 2+ (A3−ωs2A1) 2} (57)
Is obtained.
[0092]
Next, the transient solution vptr of the differential equation of equation (52) is obtained. for that reason,
vptr = exp (λt)
And substitute for vp in the following equation.
A1 (d2vp / dt2) + A2 (dvp / dt) + A3vp = 0 (58)
As a result, the following equation is obtained.
A1λ2 + A2λ + A3 = 0 (59)
[0093]
Therefore, when this equation is solved for λ, the following equation is obtained.

[0094]
Here, when {A2 / (2A1)} 2 <(A3 / A1) is set (vibration mode), the following equation is obtained.

[0095]
At this time,
β1 = A2 / (2A1) (62)
ω1 = √ (A3 / A1-β12 (63)
It is.
[0096]
Here, the transient solution vptr is set as follows.
vptr = (a1 cos ω1t + ja2sin ω1t) exp (−β1t) (64)
Then, since the voltage vp is expressed by the sum of the steady solution and the transient solution, it is given by the following equation using Equation (53) and Equation (64).
vp = (E1 cos ωst + E2 sin ωst) + (a1 cos ω1t + ja2 sin ω1t) exp (−β1t) (65)
[0097]
Further, the current i is obtained by substituting Equation (65) into Equation (2) as follows:

[0098]
Next, it is assumed that vp and i at t = t1 are v02 and i0, respectively, as in the following equation.
i0 = J0 + (a1J1 + ja2J2) exp (−β1t1) (68)
v02 = P0 + (a1P1 + ja2P2) exp (-. beta.1t1) (69)
Then, the following formulas are established from formulas (65) to (69).
J0 = (E1 / Rp + ωsCE2) cosωst1 + −ωsCE1 + E2 / Rp) · sinωst1 (70)
J1 = {(1-β1CRp) / Rp} cosω1t1−ω1Csinω1t1 (71)
J2 = ω1C cos ω1t1 + {(1-β1CRp) / Rp} sin ω1t1 (72)
P0 = E1 cos ωst1 + E2 sin ωst1 (73)
P1 = cos ω1t1 (74)
P2 = sinω1t1 (75)
[0099]
Further, when the equations (68) to (69) are solved for a1 and a2, the following is obtained.

[0100]
Furthermore, it can be seen from the equations (71) to (75) that the relationship of the following equation is established.
J2P1−J1P2 = ω1C (78) Accordingly, when the equations (76) to (77) are substituted into the equation (64) and the equation (78) is used at that time, the following equation is obtained as the transient solution vptr.

[0101]
here,
B1tr = v02−P0 (81)
t '= t-t1 (82)
B2tr =-{(1-.beta.1CRp) (v02-P0) -Rp (i0-J0)} / (. Omega.1CRp) (83)
After all,
vptr = (B1 trcos ω1t ′ + B2 trsin ω1 t ′) exp (−β1t ′) (84)
Is obtained.
[0102]
Eventually, equation (65) becomes:

[0103]
Next, in equation (67) described above,
D1st = (E1 / Rp) + ωsCE2 (86)
D2st = -ωsCE1 + (E2 / Rp) (87)
D1tr = {(1-β1CRp) / Rp} B1tr + ω1CB2tr (88)
D2tr = -ω1CB1tr + {(1-β1CRp) / Rp} B2tr (89)
And
[0104]
Then, as the current i,

Is obtained.
[0105]
Also, the current is is
is = Cc (dv1 / dt) + i = ωsCcEm′cosωst + i (91)
As obtained.
[0106]
[4.2] Diastole
In the diastole, the diode D is turned off, and the left ventricular pressure e is not applied to the circuit on the cathode side of the diode D, and the current flowing through the capacitance CC is equal in magnitude to the current i and is in the reverse direction. It becomes current. Therefore, the voltage v1 is expressed by the above-described equation (1), and the current i and the current ic are respectively expressed by the following equations.
i = -Cc (dv1 / dt) (92)
ic = C (dvp / dt) (93)
[0107]
Therefore, the voltage vp is

It becomes.
[0108]
Since i-ic = ip = vp / Rp,
i = vp / Rp + C (dvp / dt) (95)
It becomes.
[0109]
Next, substituting equation (95) into equation (1) and differentiating both sides of the obtained equation with time t yields the following equation.

[0110]
Further, the following expression is derived from Expression (92) and Expression (95).
dv1 / dt =-{vp / Rp + C (dvp / dt)} / Cc (97)
Then, the following expression is obtained from Expression (96) and Expression (97).

[0111]
Therefore, when this equation is modified, the following equation is obtained.
(D3vp / dt3) + A1 '(d2vp / dt2) + A2' (dvp / dt) + A3'vp = 0 (99)
here,
A1 '= (L + CRcRp) / (LCRp) (100)
A2 '= (CcRc + CcRp + CRp) / (LCcCRp) (101)
A3 '= 1 / (LCcCRp) (102)
[0112]
Next, when vp = exp (λt) is substituted into equation (99), the following equation is obtained.
(Λ3 + A1′λ2 + A2′λ + A3 ′) exp (λt) = 0 (103)
In addition, the following definitions are made.
p = (A1'2 / 9)-(A2 '/ 3) (104)
q = -A1'3 / 27 + (A1'A2 ') / 6-A3' / 2 (105)
u = {q + √ (q2-p3)} 1/3 (106)
v = {q-√ (q2-p3)} 1/3 (107)
α ′ = − (u + v) + A1 ′ / 3 (108)
β2 = (u + v) / 2 + A1 ′ / 3 (109)
ω2 = (u−v) √ (3) / 2 (110)
λ1 = −α ′ (111)
λ2 = −β2 + jω2 (112)
λ3 = −β2−jω2 (113)
If (q2-p3)> 0, the vibration mode is set.
[0113]
The voltage vp is further assumed as follows:

[0114]
Then, by substituting equation (114) into equation (95), current i is transformed as follows.

[0115]
here,
g0 = (1-α'CRp) / Rp (116)
g1 = (1-β2CRp) / Rp (117)
g2 = (ω2CRp) / Rp (118)
[0116]
Therefore, the voltage v1 is expressed by the following equation from the equation (115).

[0117]
here,
f0 = g0 / (α'Cc) (120)
f1 = (β2g1−ω2g2) / {(β22 + ω22) Cc} (121)
f2 = (ω2g1 + β2g2) / {(β22 + ω22) Cc} (122)
Next, in order to simplify the calculation, the time t2 shown in FIG. 4 is set to t = 0 in the following description. If the voltage v1, the voltage vp, and the current i at t = 0 are v01, v02, and i0, respectively, these are obtained by setting t in the equations (119), (114), and (115) to t = 0. It is obtained as follows.
v01 = f0b1 + (f1 + jf2) b2 + (f1-jf2) b3 (123)
v02 = b1 + b2 + b3 (124)
i0 = g0b1 + (g1 + jg2) b2 + (g1-jg2) b3 (125)
[0118]
Next, by modifying the second term and the third term in the equation (114), the voltage vp becomes as follows.

[0119]
here,

Because
k1 = v01−f0v02 (130)
k2 = i0-g0v02 (131)
k3 = g1-g0 (132)
k4 = f1-f0 (133)
It is.
[0120]
Next, by substituting vp in the equation (126) into the equation (2), the current i becomes as follows.

here,
D0 = {(1-α'CRp) / Rp} B0 (135)
D1 = {(1-β2CRp) / Rp} B1 + ω2CB2 (136)
D2 =-. Omega.2CB1 + {(1-.beta.2CRp) / Rp} B2 (137)
[0121]
Therefore, from equation (134), the voltage v1 is

It becomes. here,
H0 = D0 / (α'Cc) (139)
H1 = (β2D1 + ω2D2) / {(β22 + ω22) Cc} (140)
H2 = (− ω2D1 + β2D2) / {(β22 + ω22) Cc} (141)
[0122]
As described above, the time t2 is set to t = 0 in the above description. Therefore, t → (t−t 2) is replaced in order to adjust the time scale. As a result, the voltage v1, the voltage vp, and the current i are obtained as follows from the equations (138), (126), and (134), respectively.

[0123]
here,
Hm = √ (H12 + H22) (145)
φ21 = tan-1 (H1 / H2) (146)
Bm = √ (B12 + B22) (147)
φ22 = tan-1 (B1 / B2) (148)
Dm = √ (D12 + D22) (149)
φ23 = tan-1 (D1 / D2) (150)
[0124]
Incidentally, the current is in the diastole is “0”.
Next, a theoretical value of the stroke volume SV is obtained. Since the stroke volume SV is given by the area of the current is in the systole, it is obtained by integrating the current is shown in the equation (91) from time t1 to time t2. That is,

[0125]
[5] Operation of pulse wave analyzer
Next, the operation of the pulse wave analyzer according to the present embodiment will be described with reference to FIGS.
6 to 10 are flowcharts showing the operation of the pulse wave analyzer in the first embodiment.
[0126]
FIG. 11 shows a waveform diagram of an average waveform obtained by the averaging process described above.
Further, FIG. 12 shows a waveform diagram comparing the radial artery waveform obtained by the parameter calculation process described later and the average waveform obtained by the averaging process.
Hereinafter, the operation will be described with reference to these drawings.
[0127]
(1) Pulse wave measurement data reading process (step S1)
(A) Pulse wave reading process
When evaluating the circulatory dynamic parameters, a diagnostician in charge of diagnosis of the subject wears the cuff belt S1 and the pressure sensor S2 on the subject as shown in FIG. In response to this command, the microcomputer 4 sends a pulse wave measurement instruction to the pulse wave detector 1.
As a result, the pulse wave detection device 1 detects the radial artery wave, and the A / D converter 3 outputs a time series digital signal representing this radial artery wave. The microcomputer 4 takes this digital signal into a built-in waveform memory for a certain time (about 1 minute). In this way, the radial artery waveform for a plurality of beats is captured in the waveform memory.
[0128]
(B) Averaging process
Next, the microcomputer 4 superimposes the radial artery waveforms for a plurality of beats for each beat, and obtains the average waveform per beat for the predetermined time. Then, this average waveform is stored in the built-in memory as a representative waveform of the radial artery waveform. A representative waveform W1 of the average waveform created in this way is illustrated in FIG.
[0129]
(2) Stroke volume data capture processing (step S2)
Next, the microcomputer 4 sends a stroke volume measurement instruction to the stroke volume measuring device 2. As a result, the stroke volume measuring device 2 measures the stroke volume of the subject, and the measurement result is taken into the built-in temporary storage memory by the microcomputer 4.
[0130]
(3) Parameter calculation processing (step S3)
Next, based on the four-element lumped constant model, among the five circulatory dynamic parameters constituting the five-element lumped constant model, four circulatory dynamic parameters excluding the capacitance Cc are determined.
[0131]
The microcomputer 4 executes a parameter calculation processing routine shown in FIGS. At that time, in accordance with the execution of the routine, the α, ω calculation processing routine shown in FIG. 9 is executed (steps S109 and S117). Further, along with the execution of the α, ω calculation processing routine, the ω calculation routine shown in FIG. 10 is executed (step S203).
[0132]
Hereinafter, processing contents of these routines will be described.
First, for the average waveform of the radial artery as shown in FIG. 11, the microcomputer 4 temporarily reduces the blood pressure once after the time t1 ′ and the blood pressure value y1 corresponding to the first point P1 at which the blood pressure becomes maximum. Time t2 'corresponding to two points and blood pressure value y2, time t3' corresponding to the third point P3 which is the second peak point, blood pressure value y3, time tp for one beat, minimum blood pressure value Emin (the above-mentioned formula (Corresponding to the first term of equation (4) and equation (4)) is obtained (step S101).
[0133]
If the pulse wave is “smooth” and it is difficult to distinguish the second point P2 and the third point P3, the times of the second point and the third point are set to t2 ′ = 2t1 ′, t3 ′ = Assume 3t1 ′.
Next, in order to simplify the process, blood pressure values y1 to y3 are normalized using the blood pressure value y0 at point A shown in FIG. 13 (steps S102 and S103), and the value at point B is set to (y0 / 2) Initially set to -0.1 (step S104).
[0134]
Next, optimum values of B, tb, α, and ω are determined according to the following procedure.
(A) First, B
"(Y0 / 2)-y0"
At the same time, tb
“(Tp / 2) to tp”
Change in the range. At that time, both B and tb are changed at intervals of +0.1. And for each of B and tb
| Vp (t1 ')-y1 |,
| Vp (t2 ')-y2 |,
| Vp (t3 ')-y3 |
Find α and ω that minimizes.
[0135]
(B) Among B, tb, α, and ω obtained in (a)
| Vp (t1 ')-y1 |,
| Vp (t2 ')-y2 |,
| Vp (t3 ')-y3 |
B, tb, α, and ω that minimize the value are obtained.
[0136]
(C) Based on B and tb obtained in (b),
B ± 0.05,
About tb
tb ± 0.05
Within the range, the above processes (a) and (b) are re-executed.
[0137]
(D) During the processes (a) to (c) above, α is changed in the range of 3 to 10 at 0.1 intervals, and the optimal ω is calculated for each α.
In addition, ω is
dvp (t2 ') / dt = 0
Is obtained using a bisection method (see the flowchart of FIG. 10).
[0138]
Note that the initial value vo1 in equation (33) is set to zero when calculating the value of vp in each of the above processes.
B, tb, α, and ω are finally determined by the above processing.
[0139]
(E) tp1, Em and Eo are calculated based on the equations (28) to (30) and (44) to (46) (steps S123 and S124).
(F) Using equation (50), the value of L is calculated based on the measured stroke volume SV (step S125), and the remaining parameters Rc, Rp, and C are represented by equations (44) to (44). 46) (step S126).
[0140]
Next, based on the five-element lumped constant model, the capacitance Cc which is the last circulatory dynamic parameter is determined.
At this time, the method of determining the capacitance Cc so that the calculated value of the stroke volume SV and the actually measured value coincide with each other and the minimum blood pressure of the calculated pulse wave and the minimum blood pressure of the actually measured pulse wave are static. A method for determining the capacitance Cc is conceivable. Therefore, each method will be described separately.
[0141]
(3) -1 Method of determining the capacitance Cc (aortic compliance) so that the calculated value of the stroke volume SV and the actual measurement value match.
First, a specific method for determining the capacitance Cc so that the calculated value of the stroke volume SV coincides with the actually measured value will be described.
First, the value of the capacitance Cc is estimated as follows based on the capacitance C calculated by the four-element lumped constant model. Other circulatory dynamic parameters, that is, values of Rc, Rp, C, and L are obtained by a four-element lumped constant model.
Cc = 10 · C (153)
Next, using these circulatory dynamic parameters, a calculated value of the stroke volume SV is calculated by Expression (152).
[0142]
At that time, the left ventricular pressurization time ts is estimated by the following formula from the time tp of one beat obtained by the four-element lumped constant model.
ts = (1.52-1.079tp) tp (154)
This relational expression is an experimental expression obtained from the result of measuring the contraction time of the left ventricle by echocardiography, and as shown in FIG. 14, a correlation coefficient of −0.882 is obtained. For systolic blood pressure Em ', a value obtained by a four-element lumped constant model is used (see equations (22) and (28)).
[0143]
Further, time t1 and time t2 can be obtained from the relationship of left ventricular pressure = aortic pressure. Furthermore, since v02 and i0 are the values of vp and i at t = t1, as described above, v02 and i0 are obtained by substituting t1 into t existing in equations (85) and (90). Can do.
Next, the value of the capacitance Cc is determined so that the calculated stroke volume SV calculated as described above matches the measured value taken from the stroke volume measuring device 2. That is, the value of the capacitance Cc is changed within a predetermined range from the initial value obtained by the equation (153). Then, the measured value of stroke volume and the calculated value calculated from the value of each capacitance Cc are compared to check whether the integer part of the measured value matches the integer part of the calculated value. If there is a match in the integer part, it is considered that the measured value and the calculated value match, the capacitance Cc is determined, and the parameter calculation process ends.
[0144]
On the other hand, if the measured value of the stroke volume does not match the calculated value only by adjusting the value of the capacitance Cc, the stroke of the adjusted capacitance Cc is The value of the capacitance Cc in which the difference between the measured value of the quantity and the calculated value is the minimum is taken as the final value. Next, the value of the systolic blood pressure Em ′ is changed for each 1 mmHg within a range of ± 3 mmHg, and whether or not there is a match between the measured value of the stroke volume and the calculated value is examined in the same manner as described above. If there is a systolic blood pressure Em 'where a match is found, the value is set as the final systolic blood pressure Em', and the parameter calculation process ends.
[0145]
On the other hand, even if the value of the systolic blood pressure Em 'is adjusted, if the measured value of the stroke volume still does not match the calculated value, the value of the resistance Rp is further adjusted. Therefore, among the adjusted systolic blood pressure value Em ', the final systolic blood pressure value Em' that has the smallest difference between the measured value of the stroke volume and the calculated value is set as the final value. Next, the resistance Rp is increased or decreased, for example, in increments of 10 [dyn · s / cm5], and the value with the smallest difference between the measured value and the calculated value of the stroke volume is determined as the final resistance Rp value.
[0146]
An example of a flowchart for realizing the process described above is shown in FIG. For parameters that vary within a predetermined range in the program, the subscript “v” is added to the original parameter name.
[0147]
(3) -2 Method of determining the capacitance Cc so that the minimum blood pressure of the calculated pulse wave and the minimum blood pressure of the measured pulse wave match.
Next, a method for determining the capacitance Cc so that the minimum blood pressure of the calculated pulse wave and the minimum blood pressure of the measured pulse wave coincide with each other will be described.
In this case, conventionally, the capacitance Cc is determined using the systolic time QT as an initial value.
[0148]
As a method for calculating the systolic time QT used as the initial value, conventionally, a regression is obtained in which the systolic time QT is obtained in advance from the electrocardiogram of the subject or the systolic time QT is obtained from the heart rate HR obtained from the electrocardiogram or pulse waveform. Or calculated using an equation.
The left ventricular pressurization time tsv is “0.01 [sec]” in the range of “QT + 0.1 [sec]” to “QT + 0.2 [sec]” with respect to the previously obtained systolic time QT. At the same time, the systolic blood pressure Emv ′ is changed at an interval of “1 mmHg” in the range of “Eo + Em−20 [mmHg]” to “Eo + Em + 20 [mmHg]”. For each of ′, 451 combinations are assumed, and in each of these combinations, the capacitance Cc is calculated such that the lowest blood pressure of the calculated pulse wave and the lowest blood pressure of the measured pulse wave match.
[0149]
However, in the conventional method for obtaining the systolic time QT from the electrocardiogram, it is necessary to collect an electrocardiogram in advance, and there is a problem that the apparatus configuration becomes large and complicated.
Further, in the method of obtaining the contraction time QT from the heart rate HR using the regression equation, there is a problem that it is difficult to obtain an accurate regression equation.
[0150]
FIG. 33 shows a general pulse wave waveform.
By the way, since the pulse wave waveform is obtained by measuring the pulsation of the blood flow caused by the contraction / expansion of the heart at the peripheral portion, the waveform shape reflects the motion of the heart. EED (Estimated Ejection Duration) in the figure is called an approximate ejection period, and corresponds to the time during which blood flows from the heart during a single heartbeat, and thus the systolic time QT.
[0151]
Therefore, in the present embodiment, instead of the systolic time QT, an approximate ejection time EED (Estimated Ejection Duration) is used, and the left ventricular pressurization time tsv is set to “EED + 0. In the range of 1 [sec] to “EED + 0.2 [sec]”, the blood pressure is changed at intervals of “0.01 [sec]”, and at the same time, the maximum blood pressure Emv ′ is changed from “Eo + Em−20 [mmHg]” to “Eo + Em + 20 [mmHg]”. In the range of “1 mmHg”.
[0152]
That is, 451 combinations are assumed for each of the left ventricular pressurization time tsv and systolic blood pressure Emv ′. In each of these combinations, the capacitance Cc is calculated such that the minimum blood pressure of the calculated pulse wave matches the minimum blood pressure of the measured pulse wave.
As a result, since the parameters of the circulatory dynamics can be obtained using the EED based on the pressure pulse waveform at the peripheral part of each subject, the device configuration can be simplified such that an electrocardiograph is not required, The reflected circulatory dynamic parameters can be calculated more accurately.
[0153]
Next, when the sampling value of the calculated pulse wave in each combination is P1 (t) and the sampling value of the measured pulse wave is P2 (t), the waveform average error ε in each combination is obtained by the following equation. Then, the capacitance Cc (aortic compliance) when the waveform average error ε is the smallest is employed. An example of a flowchart for realizing the process described above is shown in FIG.
ε = Σt = 0tp (| P2 (t) −P2 (t) |) / (N) (155)
[0154]
As described above, all the circulatory dynamic parameters in which the measured value and the calculated value of the stroke volume coincide with each other are determined.
Here, the values of hemodynamic parameters and the like calculated from the radial artery waveform when a 32-year-old male is the subject are shown below.
Capacitance Cc = 0.001213 [cm5 / dyn]
Electrical resistance Rc = 98.768 [dyn · s / cm5]
Inductance L = 15.930 [dyn · s2 / cm5]
Capacitance C = 0.0001241 [cm5 / dyn]
Electrical resistance Rp = 1300.058 [dyn · s / cm5]
Left ventricular pressurization time ts = 0.496 [s]
One beat time tp = 0.896 [s]
Stroke volume SV = 83.6 [cc / beat]
Maximum blood pressure Em ′ = 117.44 [mmHg]
Also, as shown in FIG. 12, it can be seen that the calculated waveform of the radial artery obtained from the calculated parameters and the measured waveform are in good agreement.
[0155]
(5) Data calculation process (step S4)
Further, an aortic pressure waveform is obtained based on the values of the circulatory dynamic parameters L, C, Cc, Rc, and Rp. That is, by using the equation (51) in the systole and using the equation (142) in the diastole, the waveform of the voltage v1 is equivalent to one beat (that is, time 0 to time tp or time t1 to Only time (t1 + tp)) is calculated.
[0156]
Then, the value at the time t1 of the waveform of the obtained aortic root is calculated from these equations, and the calculation result is set as the minimum blood pressure value Eo.
Next, the myocardial load index WP is calculated by multiplying the previously obtained maximum blood pressure value Em ′ by the heart rate HR (= 60 / tp).

[0157]
(6) Data output process (step S5)
Next, the microcomputer 4 outputs the circulatory dynamic parameters L, C, Cc, Rc, Rp obtained by the parameter calculation process to the output device 6 and displays them on the output device.
Further, the obtained calculation waveform is output to the output device 6 to display the aortic pressure waveform. Further, the maximum blood pressure value Em ′ and the myocardial load index WP are sent to the output device 6 together with the minimum blood pressure value Eo, and these values are displayed on the output device 6.
[0158]
[6] Summary of the first embodiment
By the way, it can be said that the conventional blood pressure measuring device measures the blood pressure on the peripheral side of the radial artery, the upper arm, etc., and is a technique for indirectly measuring the burden on the heart. However, changes in the burden on the heart are not always reflected in the blood pressure on the peripheral side, and viewing the burden on the heart on the peripheral side is not necessarily accurate.
[0159]
Therefore, in the first embodiment, focusing on the fact that the blood pressure waveform at the central part is particularly important when viewing the burden on the heart, the blood pressure waveform at the aortic root (the central part of the arterial system) is calculated. It is estimated and determined from the pulse waveform measured on the peripheral side. Then, if the maximum blood pressure value, the minimum blood pressure value, and the myocardial load index WP at the aortic origin are calculated from the estimated aortic pressure waveform, these values can serve as an index directly representing the burden on the heart.
[0160]
As described above, according to the present embodiment, the maximum systolic blood pressure, the minimum blood pressure, the myocardial load index, and the aortic pressure waveform on the central side can be shown to the diagnostician and the subject along with various hemodynamic parameters.
[0161]
Since equation (51) indicates the left ventricular pressure waveform itself, the left ventricular pressure waveform may be displayed on the output device 6 instead of the aortic pressure waveform described above as the pressure waveform in the central portion.
[0162]
Second embodiment
In the first embodiment, each value of the hemodynamic parameter is calculated from the radial artery waveform and the stroke volume. However, as described above, since it is necessary for the subject to wear the cuff belt S1 in order to detect the stroke volume, it can be said that the subject is troublesome.
[0163]
Therefore, in the second embodiment, focusing on the phenomenon that the aortic pressure changes depending on the shape of the radial artery waveform, the waveform shape is represented by the strain rate to estimate the central blood pressure value and the like. That is, in this embodiment, the circulatory dynamic parameter is derived based on the strain rate d obtained from the radial artery waveform.
[0164]
First, the microcomputer 4 performs (1) pulse wave reading processing and (2) averaging processing in the same manner as in the first embodiment to obtain an average waveform for one beat of the radial artery waveform. Next, a Fourier analysis of the pulse wave is performed by performing a well-known FFT (Fast Fourier Transform) process on the average waveform. Then, from the frequency spectrum obtained as a result of the analysis, the amplitude A1 of the fundamental wave, the amplitude A2 of the second harmonic, the amplitude A3 of the third harmonic,..., The amplitude An of the nth harmonic are obtained. Note that the value of n (n is a natural number) is appropriately determined in consideration of the amplitude of the harmonics. Then, based on these amplitude values, a distortion rate d defined by the following equation is calculated.
Strain rate d = (A22 + A32 +... + An2) 1/2 / A1 (156)
[0165]
Next, a circulation dynamic parameter is estimated from the obtained strain rate d. The estimation is based on the knowledge that there is a considerable degree of correlation between the distortion rate of the radial artery waveform and each value of the hemodynamic parameter. That is, the strain rate d and the circulatory dynamic parameters are measured in advance for a large number of subjects, and a relational expression between the strain rate and each circulatory dynamic parameter is derived. Here, an example of the correlation between the strain rate d and the measurement results of the circulation dynamic parameters RC, Rp, L, and C is shown in FIGS. Although the aortic compliance CC is not shown, the correlation coefficient and the relational expression can be obtained in the same manner as the other four parameters.
[0166]
Then, the circulation dynamic parameters Rc, Rp, L, C, and Cc are calculated based on the strain rate d calculated by the above equation (156) and the relational expressions shown in FIGS.
Next, in the same manner as the output processing of (5) and (6) in the first embodiment, a waveform for one beat of the aortic pressure waveform is obtained from the calculated circulatory dynamic parameters, and the minimum blood pressure value at the aortic origin is obtained. Eo, systolic blood pressure value Em ′, and myocardial load index WP are calculated and displayed on the output device 6.
[0167]
Third embodiment
In the third embodiment, in addition to the maximum blood pressure value, the minimum blood pressure value, or the myocardial load index WP at the aortic origin, the work volume of the heart (hereinafter, referred to as the blood pressure waveform at the aortic origin is calculated as described above). This is calculated and displayed.
[0168]
This cardiac work is an index representing the burden on the heart, and is defined by the product of stroke volume and aortic pressure, and the cardiac output per minute is converted into work volume.
Here, the stroke volume is defined by the blood flow volume sent out from the heart in one beat and corresponds to the area of the blood flow waveform coming out from the heart. This stroke volume has a correlation with the systolic area of the aortic pressure waveform, and the stroke volume can be obtained by applying the systolic area method to the aortic pressure waveform.
[0169]
That is, first, the area S of the pulse wave waveform corresponding to the systole of the heart is calculated. This will be described with reference to the pulse waveform of FIG. 29. The area of the region from the rising portion of the pulse wave to the depression (notch), that is, the hatched portion in FIG. Next, when the predetermined constant is K, the stroke volume SV can be calculated by the following equation.
Stroke volume SV [ml] = area S [mmHg · s] × constant K
[0170]
On the other hand, the cardiac output is defined as the blood flow delivered from the heart in one minute. Therefore, the cardiac output can be obtained by converting the stroke volume to 1 minute. That is, the cardiac output is obtained by the product of the stroke volume and the heart rate.
In the present embodiment, in the output process (5) of the first embodiment or the second embodiment, the microcomputer 4 calculates the cardiac work based on the calculated left ventricular pressure waveform and outputs it to the output device 6. indicate. Other processes are the same as those in the first embodiment or the second embodiment, and the description thereof is omitted.
[0171]
Here, the microcomputer 4 calculates the cardiac work Ws by the following procedure.
First, when ws is defined by e · is, it is calculated from the equations (51), (90), and (91) as follows.

[0172]
Here, if the first term, the second term, and the third term in the equation (157) are w1, w2, and w3, respectively, they are transformed as the following equations.

[0173]
Next, from equation (82)

Toki,
ωst−ω1t ′ = (ωs−ω1) t + ω1t1 = ωbt + Φ (165)
far.
[0174]
That is,
ωa = ωs + ω1 (166)
ωb = ωs−ω1 (167)
Φ = ω1t1 (168)
It is.
[0175]
Then, Formula (163) becomes like the following formula.

[0176]
Next, W1, W2, and W3 are respectively defined as follows, and the following expressions are derived from Expressions (161), (162), and (169).

[0177]
Since the work amount Ws is obtained by converting the total sum of W1, W2, and W3 per "minute", it is finally expressed by the following equation.
Ws = (W1 + W2 + W3) × 10 −7 × 60 / tp [J / min] (176)
In addition to the maximum blood pressure value, the minimum blood pressure value, and the myocardial load index WP at the beginning of the aorta, the meaning of displaying the cardiac work as described above is as follows.
[0178]
A useful index different from the highest blood pressure value, the lowest blood value value, or the myocardial load index WP at the beginning of the aorta as an index representing the burden on the heart by obtaining the above-described cardiac work based on the aortic pressure waveform Can also be provided.
Here, the significance of calculating the mental work will be described below with an example.
[0179]
Consider the case where hypertension is treated by administering an antihypertensive agent to a patient. Normally, if the drug is effective, changes appear in the maximum blood pressure value and the minimum blood pressure value measured at the radial artery, and the effect of the drug can be confirmed.
However, even when there is no change in the maximum blood pressure value or the minimum blood pressure value, the drug is actually effective, and the heart load itself may be lightened.
This is because the role of the antihypertensive agent is to reduce the load on the heart somewhere in the arterial system, and it is not always necessary to lower the blood pressure in the radial artery.
[0180]
Thus, even when there is no significant change in the blood pressure value in the peripheral part of the arterial system such as the radial artery, by calculating the cardiac work obtained from the blood pressure waveform of the aortic root, It is possible to know the true heart burden.
By the way, although such a change in the burden on the heart can be found by careful examination of the blood pressure waveform at the beginning of the aorta, the subtle change in the waveform can be quantitatively calculated by calculating the cardiac work. It becomes possible to express.
[0181]
Therefore, not only the maximum blood pressure value and the minimum blood pressure value, but also the myocardial load index WP and the cardiac work amount are obtained and displayed, whereby the evaluation of the antihypertensive agent therapy can be performed more finely.
FIGS. 22 to 24 show the results of calculating the cardiac work for each of the first to third types of pulse wave shapes described above.
[0182]
Fourth embodiment
In the first to third embodiments, the myocardial load index WP is always based on the aortic origin blood pressure and the detected heart rate. However, in the fourth embodiment, the peripheral blood pressure and the aorta are calculated. In the heart rate range where the difference from the origin blood pressure can be ignored, the peripheral blood pressure can be used instead of the aortic origin blood pressure, and the difference between the peripheral blood pressure and the aortic origin blood pressure can be ignored. In the heart rate range where it is not possible, the calculation processing is reduced as a whole by using the aortic origin blood pressure.
[0183]
In general, it is known that as the heart rate increases, the difference between the aortic origin blood pressure and the peripheral blood pressure increases.
Therefore, in the fourth embodiment, when the heart rate fluctuation range is within a predetermined reference heart rate fluctuation range, the myocardial load index WP is obtained using the peripheral blood pressure (maximum blood pressure) and the detected heart rate. When the heart rate fluctuation range is equal to or greater than the reference heart rate fluctuation range, the myocardial stress index is obtained using the aortic origin blood pressure (maximum blood pressure) and the detected heart rate.
[0184]
More specifically, the heart rate fluctuation with respect to the reference heart rate, which is a human heart rate at rest (of course, has individual differences), is about ± 10 [%].
Therefore, for example, if the actual heart rate fluctuation is less than ± 10 [%] with respect to the resting reference heart rate, the difference between the peripheral blood pressure and the aortic origin blood pressure can be ignored. Based on the peripheral blood pressure Pperi and the heart rate HR, the myocardial load index WP is calculated by the following formula.
WP = Pperi × HR
On the other hand, when the actual heart rate variability is more than ± 15 [%] considering the measurement error margin with respect to the resting reference heart rate, the difference between the peripheral blood pressure and the aortic origin blood pressure is It is determined that it cannot be ignored, and the myocardial load index WP is calculated by the following formula based on the aortic origin blood pressure Pcent and the heart rate HR.
WP = Pcent × HR
Thus, according to the fourth embodiment, if the actual heart rate fluctuation is less than the predetermined range with respect to the resting reference heart rate, the myocardial load index WP is calculated using the peripheral blood pressure. As compared with the case where the myocardial load index is always calculated using the aortic origin blood pressure, the processing can be simplified and the processing speed can be improved.
[0185]
If the actual heart rate fluctuation is greater than or equal to a predetermined range with respect to the resting reference heart rate, the myocardial load index WP is calculated using the aortic origin blood pressure calculated based on the circulatory dynamic parameters. Therefore, a more accurate myocardial load index can be calculated as compared with the case where the myocardial load index WP is calculated using the peripheral blood pressure.
As described above, according to the fourth embodiment, although the processing can be simplified as a whole, it is possible to always calculate an accurate myocardial load index WP.
[0186]
Fifth embodiment
In the first to third embodiments, the myocardial load index WP is always calculated based on the aortic origin blood pressure and the detected heart rate at a predetermined timing. In the case where it is not considered that there has been much change, it is considered that it is not always necessary to continuously calculate the myocardial load index WP.
[0187]
Therefore, in the fifth embodiment, the myocardial load index WP is newly calculated only when the aortic origin blood pressure is considered to have changed significantly, and when the aortic origin blood pressure is not considered to have changed much. Is an embodiment for reducing the amount of calculation processing by holding and displaying the myocardial load index WP obtained last time without calculating the myocardial load index WP.
[0188]
By the way, in order to calculate the aortic origin blood pressure, it is necessary to calculate the circulatory dynamic parameter prior to that.
In this case, when the change (rate of change) of the circulatory parameter obtained this time with respect to the circulatory parameter obtained last time (or several times before) is small, the aortic origin that would be obtained from the circulatory parameter obtained this time It is considered that the change (rate of change) with respect to the aortic origin blood pressure that would be obtained from the circulatory dynamic parameter obtained last time of the initial blood pressure is also small.
[0189]
Therefore, in the fifth embodiment, the circulatory dynamic parameter obtained this time is compared with the circulatory dynamic parameter obtained last time, and when the change rate of each circulatory dynamic parameter is less than a predetermined reference change rate, the aorta The myocardial load index obtained last time (or a plurality of times before) without holding the calculation of the starting part blood pressure and hence the myocardial load index WP is held, and the display is continued.
[0190]
In addition, the circulatory dynamic parameter obtained this time is compared with the circulatory dynamic parameter obtained last time, and if the change rate of each circulatory dynamic parameter is equal to or higher than a predetermined reference change rate, the circulatory dynamic parameter obtained this time is used. The calculation of the aortic origin blood pressure and the myocardial load index WP are performed.
[0191]
More specifically, the circulatory dynamic parameters obtained this time are compared with the circulatory dynamic parameters obtained in the previous time (or several times before), and when the change rate of each circulatory dynamic parameter is ± 5 [%] or more. The calculation of the aortic origin blood pressure and the myocardial load index WP based on the calculated aortic origin blood pressure are performed.
[0192]
On the other hand, when the rate of change of each circulatory dynamic parameter is less than ± 5 [%], the calculation of the aortic origin blood pressure and the calculation of the myocardial load index WP is not performed, but the myocardium obtained last time (or a plurality of times before). Hold the load index and continue displaying.
As a result, according to the fifth embodiment, it is not necessary to perform unnecessary calculations, the amount of calculation processing can be reduced, the processing can be simplified, and the overall processing speed can be improved.
[0193]
Modification of the embodiment
The present invention is not limited to the above-described embodiment, and various modifications can be made as follows, for example. For example, the form which calculates | requires a circulatory dynamics parameter without measuring the stroke volume SV is also considered.
That is, according to this embodiment, assuming that the inductance L of the circulatory dynamic parameters is a fixed value, the values of the other circulatory dynamic parameters are calculated based only on the waveform of the radial artery pulse wave measured from the subject. To do. In this way, the stroke volume measuring device 2 required in the configuration of FIG. 1 can be omitted as shown in FIG. Therefore, the measurement mode in this embodiment does not require the cuff band S1 required in FIG. 2, as shown in FIG.
[0194]
By the way, if the value of the inductance L is fixed in this way, the accuracy of the obtained circulatory dynamic parameter is lowered as compared with the method using the measured stroke volume. In order to compensate for this, as shown in FIG. 17, the radial artery waveform (measurement waveform) W1 obtained by measurement and the radial artery waveform (calculation waveform) W2 obtained by calculation are displayed on the output device 6 in an overlapping manner. Let First, the value of the inductance L is set to the above fixed value to obtain a calculated waveform W2, and this waveform is displayed on the output device 6 to see the degree of coincidence of the waveform with the measured waveform W1. Next, the diagnostician determines an appropriate value different from the above fixed value as the inductance L, obtains the calculated waveform W2 again, and looks on the output device 6 for the degree of coincidence with the measured waveform W1. Thereafter, the diagnostician appropriately determines several values of the inductance L in the same manner as described above, obtains the calculated waveform W2 for each value of the inductance L, and each of the calculated waveforms W2 and the measured waveform on the output device 6. Compare with W1. Then, one of the calculated waveforms W2 that best matches the measured waveform W1 is selected, and the value of the inductance L at that time is determined as the optimum value.
[0195]
Note that it is conceivable to use a trapezoidal wave instead of the above-described triangular wave as a model of the pressure waveform at the aortic root. In this way, since the waveform is closer to the actual pressure waveform than when approximated by a triangular wave, a more accurate circulatory dynamic parameter can be calculated.
[0196]
Further, the measurement location of the pulse wave and stroke volume is not limited to the location shown in FIG. 2 or FIG. 16, and may be any location on the subject's body. That is, in the above-described embodiment, the measurement mode is such that the cuff belt S1 is attached to the upper arm of the subject. However, it is preferable that the cuff belt is not used in consideration of the convenience of the subject.
[0197]
As an example, a form in which both the radial artery waveform and the stroke volume are measured at the wrist can be considered. As an example of this type of configuration, as shown in FIG. 18, a sensor 12 comprising a blood pressure measurement sensor and a stroke volume measurement sensor is mounted on a belt 13 of a wristwatch 11, and a pulse wave analysis device Of these, a configuration in which the component 10 other than the sensor 12 is built in the main body of the wristwatch 11 is conceivable. Then, as shown in the figure, the sensor 12 is slidably attached to the belt 13 by the attachment 14, and when the subject puts the wristwatch 11 on the wrist, the sensor 12 is pushed to the radial artery with an appropriate pressure. It is supposed to be applied.
[0198]
Moreover, since the form which measures a pulse wave and stroke volume in a finger | toe is also considered, the structural example of the apparatus by this form is shown in FIG. As shown in the figure, a sensor 22 composed of a blood pressure measurement sensor and a stroke volume measurement sensor is attached to the base of a finger (index finger in this example), and the sensor 22 of the pulse wave analyzer is also included. The other components 10 are built in the wristwatch 21 and connected to the sensor 22 via lead wires 23 and 23.
[0199]
Furthermore, by combining these two measurement forms, the stroke volume is measured at the wrist and the pulse wave is measured at the finger, the stroke volume is measured at the finger, and the radial artery wave is measured at the wrist It becomes possible to realize the form to do.
[0200]
Further, by adopting a configuration without a cuff belt as described above, the subject does not have to wear his arm, and the burden on the subject is reduced in the measurement.
On the other hand, the configuration shown in FIG. 20 can be considered as a form using only the cuff belt. As shown in the figure, a sensor 32 composed of a blood pressure measurement sensor and a stroke volume measurement sensor, and a component 10 other than the sensor 32 in the pulse wave analysis device are connected to the upper arm of the subject by a cuff band. It can be seen that the structure is simple even when compared with FIG.
[0201]
In the above embodiment, the pulse wave is used for calculating the circulation dynamic parameter. However, the present invention is not limited to this, and it is needless to say that other biological states can be used.
In the above embodiment, the myocardial load index is calculated based on the aortic origin blood pressure based on the calculated circulatory dynamic parameter and the detected heart rate, but the peripheral blood pressure of the living body or the pulse of the peripheral portion of the living body is used. If the estimated value of the aortic origin blood pressure of the living body is calculated based on the waveform, regardless of the calculation method, the myocardial load index is similarly calculated based on the estimated value of the aortic origin blood pressure and the detected heart rate. Can be calculated.
[0202]
For example, instead of the aortic origin blood pressure calculated based on the circulatory dynamic parameters, the aortic origin of the living body is determined based on a predetermined transfer function such as GTF (Genelal Transfer Function) from the pulse waveform of the peripheral part of the living body. It is also possible to calculate the estimated value of the initial blood pressure and calculate the myocardial load index based on the calculated estimated value of the aortic starting blood pressure and the detected heart rate.
In this case, the predetermined transfer function is not limited to a general transfer function that can be applied to all people, and a transfer function with corrections specific to a specific living body can be used.
[0203]
Effects of the embodiment
According to the present embodiment, since the myocardial load index is calculated based on the estimated value of the aortic origin blood pressure based on the calculated circulatory dynamic parameter and the detected heart rate, the myocardial load index is calculated using the peripheral blood pressure. Compared to the case, the optimal myocardial load index can be calculated under a wider range of conditions.
[0204]
In addition, when the heart rate fluctuation rate is equal to or higher than the reference heart rate fluctuation rate, the myocardial load index is calculated based on the aortic origin blood pressure and the detected heart rate, so the aortic origin blood pressure and the detected heart rate are always calculated. As compared with the configuration for calculating the myocardial load index based on the above, the processing can be further simplified and the processing can be speeded up.
[0205]
Furthermore, according to the configuration in which the myocardial load index is calculated based on the peripheral blood pressure and the detected heart rate when the heart rate fluctuation rate is less than the reference heart rate fluctuation rate, the accurate myocardial load index can be obtained despite the fact that an accurate myocardial load index can be obtained. The processing can be simplified.
Furthermore, the myocardial load index is calculated based on the aortic origin blood pressure and the detected heart rate when the fluctuation rate of the calculated hemodynamic parameter with respect to the reference hemodynamic parameter is equal to or higher than a preset parameter reference fluctuation rate. Therefore, in the state where the myocardial load index does not change so much, the calculation process is not performed unnecessarily, so that the process can be simplified.
[0206]
In addition, according to the configuration in which the myocardial load index is calculated based on the peripheral blood pressure and the detected heart rate when the fluctuation rate of the hemodynamic parameter is less than the reference parameter fluctuation rate, an accurate myocardial load index can be obtained. Therefore, the process can be simplified.
Further, based on the peripheral blood pressure of the living body or the pulse waveform of the peripheral portion of the living body, the estimated value of the aortic origin blood pressure of the living body is calculated, the heart rate of the living body is detected, the estimated value of the aortic starting blood pressure and According to the configuration in which the myocardial load index is calculated based on the detected heart rate, the myocardial load index can be calculated easily, accurately, and quickly.
[0207]
Furthermore, the estimated value of the aortic origin blood pressure of the living body is calculated based on the pulse waveform and the predetermined transfer function of the peripheral part of the living body, the heart rate of the living body is detected, the aortic starting blood pressure and the detected heart rate According to the configuration in which the myocardial load index is calculated based on the number, it is possible to calculate the myocardial load index simply, accurately, and quickly.
[0208]
【The invention's effect】
According to the present invention, when calculating the circulatory dynamic parameters, by using the left ventricular pressurization time calculated with the approximate ejection period as an initial value, it is possible to simplify the device configuration, such as eliminating the need for an electrocardiograph, Since a more accurate hemodynamic parameter that reflects the circulatory dynamics of the subject can be calculated, the estimated value of the aortic origin blood pressure of the living body can be calculated more accurately based on the obtained hemodynamic parameter.
[0209]
Also, based on the peripheral blood pressure of the living body or the pulse waveform of the peripheral part of the living body, the estimated value of the aortic origin blood pressure of the living body is calculated, the heart rate of the living body is detected, and the estimated value of the aortic starting blood pressure is calculated. Since the myocardial load index is calculated based on the detected heart rate, the myocardial load index can be calculated easily, accurately, and quickly.
[0210]
In addition, an estimated value of the aortic origin blood pressure of the living body is calculated based on the pulse waveform of the peripheral part of the living body and a predetermined transfer function, the heart rate of the living body is detected, the aortic starting blood pressure and the detected heart rate Therefore, the myocardial load index can be calculated simply, accurately, and quickly.
[0211]
Furthermore, since the myocardial load index is calculated based on the heart rate detected from the heart rate detected from the aortic origin blood pressure and the electrocardiogram based on the calculated circulatory parameters or the electrocardiogram, the peripheral blood pressure is used. Compared with the case where the myocardial load index is calculated, the optimal myocardial load index can be calculated under a wider range of conditions.
[0212]
The myocardial load index calculating means calculates the myocardial load index based on the aortic origin blood pressure and the detected heart rate when the heart rate fluctuation rate is equal to or higher than the reference heart rate fluctuation rate. Compared with the configuration in which the myocardial load index is calculated based on the blood pressure and the detected heart rate, the processing can be further simplified, and the processing speed can be increased.
[0213]
Furthermore, the myocardial load index calculating means calculates the myocardial load index based on the peripheral blood pressure and the detected heart rate when the heart rate fluctuation rate is less than the reference heart rate fluctuation rate. Despite being obtained, the processing can be simplified.
[0214]
The myocardial load index calculating means is configured to calculate a myocardial load index based on the aortic origin blood pressure and the detected heart rate when the calculated rate of change of the hemodynamic parameter with respect to the reference hemodynamic parameter is equal to or greater than a preset parameter reference rate of change. Therefore, in the state where the myocardial load index does not change so much, unnecessary calculation processing is not performed, so that the processing can be simplified.
[0215]
Furthermore, since the myocardial load index calculating means calculates the myocardial load index based on the peripheral blood pressure and the detected heart rate when the fluctuation rate of the circulatory dynamic parameter is less than the reference parameter fluctuation rate, an accurate myocardial load index can be obtained. Nevertheless, the processing can be simplified.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a pulse wave analysis device according to a first embodiment of the present invention.
FIG. 2 is a diagram showing a measurement mode using a pulse wave detection device and a stroke volume measuring device in the first embodiment.
3A is a circuit diagram showing a four-element lumped constant model that models the arterial system of a human body, and FIG. 3B is a circuit diagram showing a five-element lumped constant model. FIG.
FIG. 4 is a diagram showing a left ventricular pressure waveform and a blood pressure waveform at the beginning of the aorta.
FIG. 5 is a diagram showing a waveform obtained by modeling a blood pressure waveform at the aortic origin.
FIG. 6 is a flowchart showing an outline of the operation of the pulse wave analyzer according to the first embodiment.
FIG. 7 is a flowchart showing an operation of a parameter calculation process of the pulse wave analyzer in the first embodiment.
FIG. 8 is a flowchart showing an operation of a parameter calculation process of the pulse wave analyzer in the first embodiment.
FIG. 9 is a flowchart showing an operation of α, ω calculation processing of the pulse wave analysis apparatus in the first embodiment.
FIG. 10 is a flowchart showing an operation of ω calculation processing of the pulse wave analysis apparatus according to the first embodiment.
FIG. 11 is a waveform diagram illustrating a radial artery waveform obtained by the averaging process of the pulse wave analyzer according to the first embodiment.
FIG. 12 is a waveform diagram in which the radial artery waveform obtained by the calculation process of the pulse wave analysis device according to the first embodiment and the radial artery waveform obtained by the averaging process are displayed in an overlapping manner.
FIG. 13 is a diagram for explaining normalization processing contents applied to the radial artery waveform obtained by the averaging processing of the pulse wave analysis apparatus according to the first embodiment.
FIG. 14 is a diagram showing a correlation between a left ventricular pressurization time ts and a time tp of one beat.
FIG. 15 is a block diagram showing a configuration of a pulse wave analysis device according to a second embodiment.
FIG. 16 is a diagram illustrating a measurement mode using the pulse wave detection device according to the second embodiment.
FIG. 17 is a diagram showing a superimposed display of the measured waveform and the calculated waveform of the radial artery wave displayed on the output device in the second embodiment.
FIG. 18 is a perspective view of a form in which the component part of the pulse wave analysis apparatus excluding the sensor is built in the wristwatch and the sensor is attached to the wristband of the wristwatch.
FIG. 19 is a perspective view of a form in which the components of the pulse wave analysis apparatus excluding the sensor are built in a wristwatch and the sensor is attached to the base of a finger.
FIG. 20 is a configuration diagram of the configuration of the pulse wave analysis device excluding the sensor and a form in which the sensor is attached to the upper arm portion by a cuff belt.
FIG. 21 is a block diagram showing a configuration of a blood pressure measurement device according to a third embodiment of the present invention.
FIG. 22 is a diagram showing a relationship between an aortic pressure waveform (dotted line) and a radial artery waveform (solid line) in the first type of pulse wave.
FIG. 23 is a diagram illustrating a relationship between an aortic pressure waveform (dotted line) and a radial artery waveform (solid line) in a second type of pulse wave.
FIG. 24 is a diagram showing a relationship between an aortic pressure waveform (dotted line) and a radial artery waveform (solid line) in a third type of pulse wave.
FIG. 25 is a graph showing the relationship between central vascular resistance Rc and strain rate d.
FIG. 26 is a graph showing the relationship between peripheral vascular resistance Rp and strain rate d.
FIG. 27 is a diagram showing the relationship between inertia L due to blood flow and strain rate d.
FIG. 28 is a diagram illustrating the relationship between compliance C and strain rate d.
FIG. 29 is a diagram illustrating a systolic area method.
FIG. 30 is a diagram illustrating a configuration of a blood pressure measurement device according to a conventional technique.
FIG. 31 is a flowchart of a program for obtaining a capacitance Cc.
FIG. 32 is a flowchart of another program for obtaining the capacitance Cc.
FIG. 33 is an explanatory diagram of a pulse wave waveform.
[Explanation of symbols]
1 Pulse wave detector
2 Stroke volume measuring device
3 A / D converter
4 Microcomputer
5 Keyboard
6 Output device
7 Monitor
61 Measured blood pressure display
62 Central blood pressure indicator
63 Warning display
64 Parameter display section
65 Printer
66 Print command button
67 CRT display
68 Myocardial stress index display

Claims (10)

Measuring means for measuring the state of the living body based on the pulse wave waveform of the peripheral part of the living body;
Based on the state of the living body, as a circulatory kinetic parameter representing the circulatory kinetics of the arterial system from the central part to the peripheral part of the living body, an analysis means for calculating a circulatory kinetic parameter including viscoelasticity of the aorta,
Aortic blood pressure calculating means for calculating an estimated value of the aortic origin blood pressure of the living body based on the circulatory dynamic parameter;
Heart rate detecting means for detecting the heart rate of the living body;
Myocardial load index calculating means for calculating a myocardial load index based on the estimated value of the aortic origin blood pressure and the detected heart rate;
A discriminating means for discriminating whether or not a fluctuation rate of the detected heart rate with respect to a reference heart rate which is a resting heart rate exceeds a preset reference heart rate fluctuation rate ;
The myocardial load index calculating means calculates the myocardial load index based on the aortic origin blood pressure and the detected heart rate when the fluctuation rate is equal to or higher than the reference heart rate fluctuation rate based on the determination. A biological state measuring device characterized by calculating. The biological state measuring device according to claim 1 ,
Comprising peripheral blood pressure detecting means for noninvasively detecting the peripheral blood pressure of the living body ,
The myocardial load index calculating means calculates a myocardial load index based on the peripheral blood pressure and the detected heart rate when the fluctuation rate is less than the reference heart rate fluctuation rate based on the determination. Biological condition measuring device. Measuring means for measuring the state of the living body based on the pulse wave waveform of the peripheral part of the living body;
Based on the state of the living body, as a circulatory kinetic parameter representing the circulatory kinetics of the arterial system from the central part to the peripheral part of the living body, an analysis means for calculating a circulatory kinetic parameter including viscoelasticity of the aorta,
Aortic blood pressure calculating means for calculating an estimated value of the aortic origin blood pressure of the living body based on the circulatory dynamic parameter;
Heart rate detecting means for detecting the heart rate of the living body;
Myocardial load index calculating means for calculating a myocardial load index based on the estimated value of the aortic origin blood pressure and the detected heart rate;
The myocardial load index calculating means includes the aorta when the variation rate of the calculated hemodynamic parameter with respect to a reference hemodynamic parameter, which is the hemodynamic parameter, at a predetermined timing is greater than or equal to a preset parameter reference variation rate. A biological state measuring device, wherein a myocardial load index is calculated based on a starting blood pressure and the detected heart rate. The biological state measuring device according to claim 3 ,
Comprising peripheral blood pressure detecting means for noninvasively detecting the peripheral blood pressure of the living body,
The myocardial load index calculating means calculates a myocardial load index based on the peripheral blood pressure and the detected heart rate when the fluctuation rate is less than the reference parameter fluctuation rate based on the determination. Biological condition measuring device. In the biological condition measuring device according to any one of claims 1 to 4 ,
The circulatory dynamic parameters include vascular resistance due to blood viscosity at the central part, blood inertia, vascular resistance at the peripheral part, and viscoelasticity of the blood vessel at the peripheral part.
The biological state measuring apparatus characterized by the above-mentioned. In the biological condition measuring device according to any one of claims 1 to 5 ,
The aortic blood pressure calculation means corresponds to the diode corresponding to the aortic valve, the first resistance corresponding to the vascular resistance due to blood viscosity in the central part, the inductance corresponding to the inertia of blood, and the viscoelasticity of the aorta A model having a first capacitance, a second resistance corresponding to the vascular resistance in the peripheral portion, and a second capacitance corresponding to the viscoelasticity of the blood vessel in the peripheral portion, A series circuit of the diode and the first capacitance is connected between a pair of input terminals, and a parallel circuit including the second capacitance and the second resistor is inserted between a pair of output terminals, The circulatory dynamics of the arterial system is modeled by a five-element lumped constant model in which a series circuit composed of the first resistor and the inductance is inserted between both terminals of the first capacitance and the output terminal. The circulation And determines the status parameter, the voltage waveform between the terminals of the first capacitance and the aorta pressure waveform,
The biological state measuring apparatus characterized by the above-mentioned. In the biological condition measuring device according to any one of claims 1 to 6 ,
The state of the living body is a pulse wave in the peripheral part of the arterial system,
The blood pressure calculation means is configured to obtain an electrical signal corresponding to the waveform of the pulse wave from the output terminal when an electrical signal corresponding to the left ventricular pressure of the living body is applied between the input terminals. A biological state measurement device characterized in that the value of each element constituting a five-element lumped constant model is determined. In the biological condition measuring device according to any one of claims 1 to 6 ,
The state of the living body is a pulse wave in the peripheral part of the arterial system,
Strain calculation means for calculating the distortion of the pulse wave from the waveform of the pulse wave;
The blood pressure calculation unit determines the circulatory dynamic parameter based on a correlation between the circulatory dynamic parameter and the distortion of the pulse wave. The biological state measuring device according to any one of claims 1 to 8 ,
Having a stroke volume detecting means for detecting stroke volume of the living body;
The blood pressure calculation means is such that the calculated value of stroke volume obtained from the aortic pressure waveform and the measured value of stroke volume measured by the stroke volume measuring means match. The biological state measuring device characterized by adjusting the value of the circulatory dynamic parameter. In the biological condition measuring device according to any one of claims 1 to 9 ,
A biological condition measuring device comprising: a work amount calculating means for calculating a work amount of the heart of the living body based on the aortic pressure waveform.

Images