JP4711583B2 - Ultrasonic imaging device - Google Patents

Description

ここに、音速ｃ_１はレンズの材質によって決まり、また実測可能であるので既知のものである。また、音速ｃ_２，ｃ_３も以下の説明において、経験的に近似値として既知であるとの条件の下で以下の説明を進め、既知ではなくそれを計測によって求める方法については他の実施形態として後に説明する。なお、音速ｃ_１は超音波探触子２０のメモリ２２内に、また、音速ｃ_２，ｃ_３は超音波撮像装置本体１０のメモリ１２内に保持されているものとする。
本発明では、この問題の数値的な解法として２つの方法を提案する。まず第１の方法について以下に説明する。式１からｓｉｎを消去すると次の式２のようになる。

しかし、式２は、このままでは解析的に解くことはできない。ここでは、解析的な解を求めるために式２を３つの連立方程式で解くことにする。ここで変数を落とすときに残す変数として、レンズ層中での横方向の移動量ｘ_１、脂肪層中での横方向の移動量ｘ_２、生体中での横方向の移動量ｘ_３のどれを残すかについての選択肢があるが、誤差の大きさを考えた時に多くの条件下で一番有利となるのはｘ_３を残す方法である。これは、音速がレンズ層、脂肪層、生体組織中と段々速くなるため、各層の中を通る音波の経路が段々横に寝てくること、および多くの撮像条件下ではこのことと併せて、厚さｄ_３が一番大きいことによる。故に実際の系では生体組織中で、音の経路は横方向移動量が一番大きくなる。一番大きな量を計算で直接求めるのが相対的な誤差を一番小さくする方法であるから、通常の撮像条件下ではｘ_３を求めればよくなる。しかし、条件によらずレンズ層が一番薄いが、脂肪層の厚さと生体組織の厚さは撮像条件、撮像対象部位の深さによって異なることがある。よって脂肪層の厚さと焦点距離の関係で、ｘ_２を残すか、ｘ_３を残すか条件判定してから一変数の多項式を解くことを考えるアルゴリズムも有り得る。しかし、ここではその方法の説明は省略してｘ_３を求める方法についてのみ説明することにする。
ｘ_１，ｘ_２を連立方程式から消去して式３を得る。式３では、ｘ_３を求めるのであるから、以下、これを単にｘと略記するとともに、ｒ_２＝（ｃ_３×ｃ_３）／（ｃ_２×ｃ_２）と置き換えている。

実際の超音波撮像装置における超音波ビームフォーマの計算処理部では、前述のようにＭＰＵもしくはＤＳＰを使っている。このうちＤＳＰを用いている場合には、計算は全て積和に換算される。代表的なＤＳＰにおいては、平方根は積和３２回に相当する。割り算は値に依存するので、計算前には計算回数がわからない。よって式３を平方根、割り算のない形に変形することはＤＳＰを用いる場合に重要であり、その操作をした結果が式４である。

ガロアの理論により、因数分解によって次数を落とすことの出来ない多項式は５次以上の場合、解の公式は存在しないことが知られている。このことから、式４をいきなり解くことは不可能である。ただし、式４は１２次方程式とはいえ、有理多項式であり、全範囲で微分可能であることから、十分に解の近傍では直線として近似することが可能である。常に解の近傍にあるという条件さえ満たすｘがあればよい。
解の近傍での式４の振る舞いを解析してみると、式３から式４への変換において平方根を除くために両辺自乗したわけであるが、これは式３の前項をｇ_１、後項をｇ_２とすると、ｇ_１（ｘ）−ｇ_２（ｘ）の解を探していたのがｇ_１（ｘ）×ｇ_１（ｘ）−ｇ_２（ｘ）×ｇ_２（ｘ）＝（ｇ_１（ｘ）−ｇ_２（ｘ））×（ｇ_１（ｘ）＋ｇ_２（ｘ））の解を探していることになることから、式４の解には、本来のｇ_１（ｘ）−ｇ_２（ｘ）の解に加え、ｇ_１（ｘ）＋ｇ_２（ｘ）の解も含まれている。この両辺自乗と言う操作は２回行われているので、関係ない解が２つ余計に入っていることになるが、２回目の自乗によって式４に新たに含まれた解は、十分に本来の解から離れているので、ここでは問題とならない。ｇ_１（ｘ）＋ｇ_２（ｘ）の解も持つことで式４は解近傍で変曲点を持つので、近似的な値を求める位置はこの変曲点より真の解に近いことが必要である。その条件に適するものとして、図中の表記を用いてｘ＋ｐ×ｄ_３／Ｆを近似解の出発点として用いる。そうすると振動子位置Ｎでのｘであるｘ_Ｎは次の式５により求まる。ここからは、式４の左辺をｆ（ｘ）と表記する。

ここからは、この漸化式を用いてリアルタイムに振動子毎の遅延時間を計算する方法についてフローチャートを交えて説明する。最初に、図８のフローチャートを用いて、リニア型探触子またはコンベックス型探触子のように、焦点と口径の中心振動子を結ぶ線分と送波面が直交する探触子の遅延時間を計算する場合について説明する。この条件では、焦点が口径の正面にあるので、各振動子の遅延時間は口径内で左右対称である。つまり、片側半分だけ計算すれば十分である。
探触子に関するパラメータは図５のステップ１１で既に与えられており、焦点深度の値は図５のステップ１４において既に与えられている。ここでは、口径の中心の振動子から始めて、一つずつ外側に向かって順に遅延時間を計算する。まず、ステップ２１において、計算対象を中心の振動子とする。中心の振動子から焦点に向かう超音波は層の界面に垂直に入射し、界面では屈折しないのでｘ_０＝０である。また、振動子番号Ｎを１（中心の振動子に対する隣の振動子）とする。
次に、ステップ２２において、ｘ_０＝０であることから、振動子番号Ｎ＝１に対するｘ_１を漸化式式５により計算する。次に、ステップ２３に進み、ステップ２２で求めたｘ_１をもとに、振動子番号Ｎ＝１の振動子に対する遅延時間τを次の一連の式、すなわち式６により求める。

次に、ステップ２４に進み、振動子番号を１つ増やし、その隣の振動子に移り、ステップ２５からステップ２２に戻って、再び漸化式（式５）でｘ_２を計算し、そのｘ_２をもとに式６を用いて振動子番号２の振動子に対する遅延時間τを計算する。この操作を口径内振動子全てについて行う。ステップ２５の判定が「ＹＥＳ」であれば、この焦点深度に関しては全振動子分の遅延時間を計算したことになるので、ステップ２６に進み、計算結果を演算・制御回路から遅延回路へ出力する。次の計算すべき焦点深度が超音波ビームフォーマに与えられたら、同様にしてまた全口径内振動子について計算する。これを全焦点深度について繰り返すことで屈折によるずれを補正した撮像が行われる。
次に、図９のフローチャートを用いて、オブリーク型探触子またはセクタ型探触子、フェイズドアレイ型探触子のように、焦点と口径の中心振動子を結ぶ線分と送波面が直交しないタイプの探触子に対して遅延時間を計算する方法、もしくは焦点と口径の中心振動子を結ぶ線分と送波面のなす角が変化していくようなタイプの探触子に対して遅延時間を計算する方法について説明する。この時、焦点位置と口径、振動子列の関係は図１０に略示するように、オブリーク角度θの大きさによって、
（１） 焦点Ｆから振動子の並んだ面に降ろした垂線の足Ａが送波口径内に入る場合
（２） 前記Ａが送波口径内に入らない場合
の二通りがある。後者の場合はＡから口径の端の振動子までの間を素子ピッチで刻み仮想的な振動子列を考える。そしてこの考えに基づき、遅延時間の計算に入る前、すなわち図５でいうと、ステップ１３の次の段階で式５によってＡの位置にある仮想的な振動子を漸化式の第一項として逐次ｘを計算して、口径の端にある振動子でのｘを求めておく。オブリーク角度などの探触子に関するパラメータは図５のステップ１１で既に与えられており、焦点深度や焦点方向等の値は図５のステップ１４において既に与えられている。
図９のステップ３１では、撮像すべき範囲内での各焦点深度、方向に関して、その焦点から送波面に降ろした垂線の足（図１０中の点Ａ）が口径内にあるか否かを計算する。口径内に有るときは、ステップ３９に進み、その垂線の足から口径の左右の端までの振動子数を計算し、ステップ４０において、その足に一番近い位置にある振動子をＮ＝１，ｘ_０＝０とする。続いて、ステップ３３において漸化式（式５）によりｘを求め、ステップ３４において式６によりｘから遅延時間τを求める処理を、振動子番号を増加しながら口径の端の位置まで行う（Ｓ２６，Ｓ２７）。この場合、対称性は使えないので、ステップ３７、ステップ４１、ステップ４０のように、Ｎ＝０の振動子から両側に口径の端まで計算していく必要がある。これはＦｃｏｓθが振動子ピッチの整数倍となるとは限らないからである。
焦点から送波面に降ろした垂線の足が口径の中にない場合は、ステップ３１からステップ３２に進み、口径の端の振動子でのｘを読み込む。この点Ａを遅延時間の計算の開始点とするので、焦点距離ＦはＦｃｏｓθ、ｄ_３は本来のＦを用いて、ｄ_３＝Ｆｃｏｓθ−ｄ_１−ｄ_２と置換する必要がある。
次に、口径の端の位置でのｘをｘ_０に入力し、Ｎ＝１とする。その後、ステップ３３において漸化式（式５）によりｘを求め、ステップ３４において式６によりｘから遅延時間τを求める。次に、ステップ３５で隣の振動子に移り、ステップ３６で口径の端の振動子に達したと判定されるまでステップ３３からステップ３５の処理を繰り返す。これで、この焦点深度、焦点方向に関しては全振動子分の遅延時間を計算したことになるので、ステップ３８において計算結果を演算・制御回路から遅延回路へ出力する。次の計算すべき焦点深度、焦点方向が超音波ビームフォーマに与えられたら、同様にしてまた全口径内振動子について計算する。これを全焦点深度、焦点方向について繰り返すことで屈折によるずれを補正した撮像が行われる。
前述の式５の漸化式を用いることで、遅延時間の近似値は十分な精度（１／中心周波数の１／１０精度）で求めることが可能となる。この方法は従来の近似式に比べると十分な精度を有し、且つ精度を重視してイテレーションで求める方法に比べて、アルゴリズム中にループを持たないので、計算速度はかなり速い。このアルゴリズムと近年の演算処理装置の高速化と相俟って、本発明のような脂肪層による屈折まで含めたリアルタイム高速遅延時間計算アルゴリズムが実現できたわけである。さらに一般の漸化式の計算に比べると、誤差が蓄積していかない点で著しく有利である。これは、近似解を求めるためのｘには毎回誤差が含まれていても、ｆ（ｘ）は近似を含まない関数であるために、ｘとｆ（ｘ）、ｆ’（ｘ）の間には誤差が存在しないため、式５にはｘ_Ｎ−１での誤差がｘ_Ｎに反映されないためである。
この考え方はＤＳＰに特化した解き方であるが、もしもその制約が無い場合にはより計算は容易になる。式３は解近傍で直線とみなせるので、微分係数から傾きを求める必要がなく、解を明らかに挟む二点として、隣の振動子でのｘ＋ｐ×ｄ_３／Ｆ、ｘ＋ｐを用い、内分点として次式のように解が求まる。したがって、漸化式（式５）の代わりに漸化式（式７）を用いることで処理速度を向上することができる。

次に、図１１を用いて本発明による超音波ビームフォーマの他の実施形態を説明する。図１１では、振動子位置Ｘ_Ｎにある振動子Ｎからレンズ層内に角度θ_Ｎで出た音が、焦点Ｆに到達すると考える。この角度θ_Ｎを求めると、各層での屈折角が求まるので、音の経路が決定され、遅延時間を決定することができる。この遅延時間を求めるために従来は、まず適当な角度で音を出し、焦点のどちら側を音が通るか計算する。もし音が焦点より遠くを通ればθ_Ｎを大きくし、焦点の手前を通ればθ_Ｎを小さくする。音が焦点を通るまでこの作業を繰り返すことで、θ_Ｎを求めている。しかし、この方法では収束させるまでにループをかなりの回数繰り返さざるを得ず、必要な速度に計算が間に合わない。速い収束を達成するには、ずれの大きさに応じてθ_Ｎの変化のさせ方を大きくすることが必要である。
本実施形態による超音波ビームフォーマで用いるアルゴリズムは、ニュートン・ラフソン法を改良し、明示的な関数を持たない系にも適応出来るようにしたものである。すなわち、次のように考える。まず適当な角度θで超音波が出たとすると、ずれの大きさはΔＸ（θ）となる。このΔＸが０になるθが求めるべき解であるから、ΔＸの微分係数を求めるためにθ＋ｄθの角度で放射されたときのずれΔＸ（θ＋ｄθ）を求める。図１２に示すように、横軸θ、縦軸ΔＸの空間で、この二点を通る直線がｘ軸を横切るときのθは、元のθに比べより良い近似解となる。これを式で表すと式８になる。

ずれ量としては、ΔＸを使う以外にΔＹを使う方法、ΔＸとΔＹの自乗和を使う方法も考えることができる。いずれも本発明で使うことはできるが、この実施形態ではΔＸを使う場合を説明する。
この操作を繰り返すことで、θは真の解に近づいていく。このとき最初に用いるθと真の解の差が小さいほど計算の繰り返しが少なくなるので、計算時間の点から有利となる。本発明では、ここでθの初期値として、Ｎ番目の振動子の遅延時間を求めるときにはＮ−１番目の振動子で求まったθ_Ｎ−１を用いる。この方法では、実用的な計算精度では、ループ２回で収束することがわかった。よってループの回数を２回に決めて計算する。ループの回数が従来の方法に比べ格段に少ないので従来の方法に比べ格段に優れている。Ｎ番目の振動子についてθが求まると、このθを初期値にしてＮ＋１番目の振動子についてのθを求める。この方法の場合も前記方法と同様焦点が振動子の正面にある場合は屈折がないのでθ＝０となり、それを計算の出発点として用いることができる。
図１３のフローチャートを用いて、リニア型探触子またはコンベックス型探触子のように、焦点と口径の中心振動子を結ぶ線分と送波面が直交する探触子の遅延時間を計算する方法について説明する。この条件では、焦点が口径の正面にあるので、各振動子の遅延時間は口径内で左右対称である。つまり、片側半分だけ計算すれば十分である。探触子に関するパラメータは図５のステップ１１で既に与えられており、焦点深度の値は図５のステップ１４において既に与えられている。
図１３のステップ５１において、口径の中心にある振動子から出た超音波はレンズ層と脂肪層の境界面、及び脂肪層と生体組織の境界面に垂直に入射し、屈折することなく焦点に達するので、振動子番号０の振動子に対してθ＝０とし、Ｎ＝１とする。次に、ステップ５２に進み、番号Ｎ＝１の振動子から角度θで超音波を放射したときの経路と、目標とする焦点位置とのずれΔＸ（θ）を求める。ここで、角度θとしては、隣の振動子（番号０の振動子）に対する角度θを用いる。また、ステップ５３において、番号１の振動子から角度θ＋ｄθで超音波を放射したときの超音波の経路と目標とする焦点位置とのずれΔＸ（θ＋ｄθ）を求める。
ここで、コンベックス型の探触子の場合、リニア型探触子とは異なり、以下のように計算される。コンベックス型探触子で、被検体を観測するときは、探触子を被検体へ押し付けて観測するので、脂肪層も探触子に沿って曲がる。脂肪層をコンベックス探触子の構造と同心円であると仮定すると、音の経路は図１５に示すようになる。素子を出発して焦点近傍を通るときのずれΔを求める方法と、焦点を出発して素子近傍を通るときのずれΔを求める二通りの方法は原理的には同じであるが、全反射条件や、音の経路が図１５の左半分側に行くことはあり得ないことからこの系は対称ではなく、前者の解き方は解の安定性が良くないので、後者の方法で解くことが望ましい。この音の経路を各層毎に分けて解析的に解くと、音が層に入る点及び角度から層を出て行く点の位置及び角度を求めれば、各層順に計算することで最後に到達する点が求まり、ずれΔも求めることができる。例として脂肪層の外側に生体中での音の経路を求めると、音の出発点の位置をＦ＝（ｘ_４、ｙ_４）出るときの位置をＰ_３＝（ｘ_３、ｙ_３）、音が出て行く側の層界面の曲率をＲ、音が入るときの角度をθとする。するとｘ_４、ｙ_４は式９の連立式からｘ_４を消去した二次方程式の±２通りある解のうち常に正となる側の解を解くことで求まる。

このときθ_４に関しては三角形ＯＦＰ_３を考えることで求まる。すなわち、今求まった（ｘ_３、ｙ_３）からθ_ｐ３が求まるので、θ_４＝θ＋θ_ｐ３の式から求まる。後はθ_４からスネルの法則を用いてθ_３が求まり、同様にして順に或る層を通過するときの経路が求まる。この例では三層問題を例に説明したが、全く同じ方法で層数がより多いときにも適用可能である。また脂肪層と脂肪層外の生体との界面が探触子の同心円で表せなくとも、ｆ（ｘ_３）の形で表せれば、式９を変形した式１０を用い、θ_４はθ＋ａｒｃｔａｎ（−ｆ’（ｘ_３））を代わりに用いることで求めることが可能となる。

次にステップ５４において、式８より外挿で新しいθを求める。続いて、ステップ５５からステップ５２に戻り、ステップ５４で求めた新しいθを用いてずれΔＸ（θ）を求める。また、ステップ５３において、新しい角度θ＋ｄθで超音波を放射したときの超音波の経路と目標とする焦点位置とのずれΔＸ（θ＋ｄθ）を求める。次に、ステップ５４において、式８より外挿で新しいθを求める。
次に、ステップ５６に進み、求めたθから次の式１１によって遅延時間τを計算する。次に、振動子番号を１つ増加し、ステップ４２からの処理を繰り返す。ステップ５８において、使用口径内全振動子についての計算が終了したと判定されれば、ステップ５９に進んで計算結果を出力する。

図１３に示したフローチャートは、図８に示したフローチャートにおける漸化式の計算処理を、焦点位置からのずれの計算処理と、外挿による新しいθの計算処理と、この部分のループ処理をするためのループ回数の判断処理とで置き換えたものに相当する。
この第２の方法に関しても、焦点が送波口径の正面にない場合について、撮像前に予め送波口径に端の振動子でのθを求め、それを記憶しておく。送波口径の端の振動子でのθは、異なる皮下脂肪層の厚さに対して複数求めておく。撮像中にこの条件になったとき、このθを読み込み、漸化式の第１項として漸化式を解くのに用いる。この計算の流れを図１４のフローチャートに示す。この詳細は、第１の方法で図８の計算処理と図９の計算処理の関係と、図１３の計算処理と図１４の計算処理の関係が全く同じであることから容易に説明される。
この口径の端の振動子でのパラメータを予め計算する方法は、現在の中級超音波撮像装置に搭載のＤＳＰもしくはＭＰＵに余裕がないための方策であって、高速の演算処理部を搭載できる高級機では、リアルタイムで計算するときも焦点から口径面に降ろした垂線の足の位置にある仮想的口径の中心から無駄になる計算も含めて全て撮像とリアルタイムに計算することも可能である。その場合は、撮像前の計算が不要になる。もちろん、この場合、仮想的な振動子に関しても実際の振動子と同じ振動子間隔で並べて計算する必要はないので、仮想的な部分のみ振動子間隔を粗くすることは計算を速くする上で有効である。この撮像前の計算は脂肪層の厚さが変わると毎回計算する必要が生じるのでその度に撮像が止まってしまう。その問題を解決するには以下のような２つの方法がある。
その第１の方法は、考えられる範囲で脂肪層厚さ毎の計算を全て事前に行い、この撮像前の計算結果をすべてデータとして超音波探触子２０のメモリ２２に記憶させておく方法である。そして、超音波探触子２０を超音波撮像装置本体１０に接続したときに、その内容を超音波撮像装置本体１０のメモリ１２に転送する。撮像中にはこのパラメータと、制御部１１により与えられた焦点深度に対し、超音波ビームフォーマ１４で遅延時間を計算し、その遅延時間を振動子アレイ２１に与え、被検体３０に対し送波する。受波もこの遅延時間分で整相し、制御部１１で診断画像を計算し、画像表示部１３に出力する。
一方、第２の方法は、予め計算したデータをＣＤ−ＲＯＭ等のメディアに入れておき、超音波探触子に付属させておく。そして、前もって、あるいは探触子を使用する際にそのＣＤ−ＲＯＭ等からデータを超音波撮像装置本体１０のメモリ１２にインストールする方法である。インストール以後の操作は前記第１の方法と全く同様である。
図１６は、本発明による超音波撮像装置の他の例を示す概略構成図である。図１６において、図１と同じ機能部分には図１と同じ符号を付し、重複する説明を省略する。今までの説明では、皮下脂肪層の厚さの入力は診断者が行っていた。すなわち、図６に示すように、診断装置の診断画面上で診断者が脂肪層の始まりの部分と終わりの部分を指示することで、画面上にその距離が表示される。この値を診断者が入力していた。
図１６に示した超音波撮像装置は、制御部１１中に脂肪層厚計算部１６を備え、皮下脂肪層に関する画面出力を直接、超音波ビームフォーマ１４に入力できるようにしたものである。脂肪層厚計算モードにおいて、診断者が図６に示すように、ポインティングデバイス等で皮下脂肪層が始まる箇所Ａと終わる箇所Ｂを指定すると、脂肪層厚計算部１６はその間の距離を計算し、計算結果をビームフォーマ１４に送信する。
次に、本発明の超音波ビームフォーマ搭載のアルゴリズムを生かして、撮像中にアダプティブに画質を改良する方法について説明する。これまで説明してきた計算アルゴリズムには、脂肪層は均質、厚さ一定であり、個人や対象部位によって脂肪層中での音速が異なることはないという前提があった。この前提は、遅延時間計算に一定の誤差があるときは、その誤差に埋もれて影響が解らなかった。しかし、本発明によって高精度な撮像が可能となると、この前提を取り除くことで画質が改善されるのがはっきりする。前記前提を取り除くには、撮像しながら微妙に脂肪層の音速を変えることができるようにしておけばよい。
図１７は、本発明による超音波撮像装置の他の例を示す概略構成図である。図１７において、図１と同じ機能部分には図１と同じ符号を付し、重複する説明を省略する。この超音波撮像装置は、脂肪層に関するパラメータを微調整することを可能にし、均質、厚さ一定でない脂肪層に対応できるようにしたものである。具体的には、図１に示す構成の超音波撮像装置へ入力装置及びパラメータ計算部１７を備え、超音波ビームフォーマが保持する脂肪層の音速を可変できるようにした。この装置では、入力装置及びパラメータ計算部１７において入力値に対し対応するパラメータを計算し、その値を超音波ビームフォーマ１４に入力する。 このパラメータを変えることの効果は次のように説明される。まず屈折がない場合を考えると式１２のようにして、各振動子から焦点まで到達するのに要する時間が計算される。これをグラフに表すと図１８のようになる。

このとき、式１２中のパラメータすなわち音速、ピッチｐ、Ｆのいずれか、もしくは複数を変化させると、図１８に図示するようにグラフの曲率が変化する。このことは、今回の問題のようにレンズ層、脂肪層、脂肪層以外の生体組織と３層からなる屈折を考慮すべき問題でも事情が複雑になるだけで原理的には同じである。遅延時間は各振動子から焦点まで到達するのに要する時間で決定されるので、「縦軸：各振動子から焦点まで到達するのに要する時間」対「横軸：各振動子の位置」の関係の曲線で遅延時間は表現される。よって、先に述べたように脂肪層の音速が場所によって一定でないようなときは、この曲線が歪むので、それにカーブフィッティングするように諸パラメータを変更することで、焦点のぼけを改善することが可能となる。カーブフィッティングは撮像画面が最適になるように調整するのが最も好ましく、パラメータは外部つまみなどによって制御すると操作性が良い。このとき脂肪層の音速に代えて脂肪層の厚さ、もしくは振動子ピッチ、これらの複数のパラメータを適切に連動させて変えて調整することも有効である。なお、図１８において、（ａ）は振動子ピッチ、脂肪層の音速に手を加えない場合、（ｂ），（ｃ）は振動子ピッチ，脂肪層音速を調整した場合を示す。
ここからは，被検体の表面に層構造が存在した場合の例についての実施例を説明する。まず、診察中に被検体表面部の層構造の厚さ、音速を求める具体例について示す。
図１９に検者がそれらを選択可能とした実施例を示す。図１９（ｂ）に示す厚さのテーブル、及び音速のテーブルを予め用意し装置内へ組み込み、検者が、モニタ１２０に表示された画像１７０を観察し、深度スケール１６０を参照して得た結果から層の厚さを読み取り、スイッチＳＷ２（１４０）等で近い厚さを選択し、音速も同様でスイッチＳＷ１（１５０）により選択可能とする。この場合スイッチＳＷ１とスイッチＳＷ２とは独立に操作が可能として、それぞれの値として検者がふさわしいと考える値を選択できるようにする。音速に関しては、数値でも良いし、脂肪層、筋肉層という選択でも良い。例えば、図１９の画面で脂肪層１７０が、画面に記されている深度方向のスケール１６０から概略２ｃｍと読めた場合、コンソール１３０に設けた層厚み設定スイッチ１４０により２ｃｍを選択する。すると、画面上に厚さ（Ｔｈｉｃｋｎｅｓｓ）が２ｃｍと表示される。次に，音速設定スイッチ１５０を所定方向へ操作して、音速を選択する。それらの選択値を確認するために、それらの入力値を画面に表示すると良い。その表示は，１４５０ｍ／ｓのように数値でも良いが，硬い筋肉質、筋肉質、標準、脂肪、高脂肪のような表現でも良い。また，スイッチは回転式でもなんでも良い。もちろんタッチパネルでも良い。
以上の説明は、生体の層構造から生ずる超音波の屈折の影響を除去して、超音波画像の画質を向上する実施形態について説明したが、超音波画像の画質を更に向上するためには、生体内での超音波の伝播速度を正確に把握して、そのデータを超音波ビームフォーマでの上記屈折補正制御に反映させることが必要である。
このことから、次に、本発明の他の実施形態を図２０に示す。これは層構造を有する生体の音速を求め、その音速により前述の実施形態の屈折補正処理を行うものである。図２０は超音波撮像装置において音速を求める部分の実施例を示すブロック図である。図において、２００は探触子で受信した複数の超音波信号を遅延制御し、受信ビーム信号を出力するデジタル制御可能なデジタル遅延部で、受信に供した振動子数に対応したチャンネル数の回路を有している。２１０はデジタル遅延部２００で遅延させた複数の信号を入力し、演算により遅延制御に供したデジタル遅延データの真の遅延データに対する誤差を推定する遅延データ誤差推定部、２２０はデジタル遅延部２００の各チャンネルの動作制御を行うデジタル遅延制御部である。
以上の構成へ加え、本実施例の超音波撮像装置装置は、さらに、予め複数の媒質音速による遅延時間を蓄えておく音速対応遅延時間記録部２３０と、遅延誤差推定部２１０で得た遅延誤差から新たな遅延時間を算出し、それを音速対応遅延時間記録部２３０に記憶されている値と照合し、その記憶値と最も近いデータを出力する遅延時間比較部２４０と、音速対応遅延時間記録部２３０に蓄えられた遅延時間データがいかなる媒質音速によるものかを記録しておく音速データ記録部２５０と、遅延時間比較部２４０の出力と一致する遅延時間の記録場所から遅延時間記録部を参照し媒質音速を選択する媒質音速選択部２６０とを備えている。そして、媒質音速選択部２６０の出力ラインは図２に示す演算・制御回路４１へ接続され、またデジタル遅延制御部２２０は演算・制御回路４１により制御されるように接続されている。
このように、図２に示す構成において、遅延回路をデジタル制御可能なものとし、その出力へ遅延誤差推定部２１０を設けるとともに、演算・制御回路４１へデジタル遅延制御部２２０を付加し、さらに上記の如く音速対応遅延時間記録部２３０と遅延時間比較部２４０と音速データ記録部２５０と媒質音速選択部２６０とを新たに設けることにより、本実施例の超音波撮像装置は真の媒質音速にほぼ等しい音速による屈折補正が可能となる。
次に、上記の如く構成された部分の動作の説明を行う。まず、生体内の音速を例えば生体の平均的な値の等音速と仮定して求めた遅延時間データを演算・制御回路４２よりデジタル遅延部２へ出力して超音波を被検体内へ送信する。このときの送波の焦点は適宜な深度に設定する。そして、送波と同じ音速に基づく遅延時間データＤを演算・制御回路４１よりデジタル遅延部２２０を介してデジタル遅延部２００へ供給してこの受信信号を遅延制御する。
そして、デジタル遅延部２００の各チャンネルの遅延制御された信号であって整相のために加算処理される前の信号が遅延誤差推定部２１０へ入力され、使用した遅延時間データＤに対する誤差ΔＤ_１を演算により推定値として求め、Ｄ_ｃ１＝Ｄ＋ΔＤ_１を出力する。この演算手法としては前述の特開平８−３１７９２３号公報に開示された相関処理法を用いることができる。遅延誤差推定部２１０から出力されたＤ_ｃ１は遅延時間比較部２４０へ入力される。すると、遅延時間比較部２４０は音速対応遅延時間記録部２３０に記録されたデータと入力したデータＤ_ｃ１とを比較し、Ｄ_ｃ１に最も近い遅延時間データを選び出し、それを媒質音速選択部２６０へ出力する。データが入力すると、媒質音速選択部２６０は入力したデータの音速が幾らであるかを選択する。この選択は、音速対応遅延時間記録部２３０のデータと音速記録部２５０の音速データとの記憶アドレスを対応させておくことで可能である。したがって、音速対応遅延時間記録部２３０と音速記録部２５０とを一つに纏めることもできる。
そして、媒質音速選択部２６０で選択された音速データは演算・制御回路４１を介してデジタル遅延制御部へフィードバックされる。このフィードバック回路は正確な音速を求めるために、上記動作を繰り返すためのもので、必要に応じて、演算・制御回路４１から上記動作の繰り返し実行指令を発する。以上の動作により生体内の音速を推定値としてではあるが測定することができる。
なお、遅延時間比較部２４０での比較動作を簡便にするために、遅延誤差推定部２１０が出力するデータＤ_ｃ１と音速対応遅延時間記録部２３０に記録されたデータとを２次元分布データとし、カーブフィッティング手法を用いることができるようにすると扱う情報量を少なくできるため有用である。さらに、遅延時間分布の階差を求め該階差遅延時間列に対して１次直線を当て嵌めるようにすると扱う情報量がさらに減少するので有用である。
次に、図２０に示す構成で層構造を有す生体の各層の音速を求める方法を説明する。図２１は超音波撮像装置のモニタに表示された被検体の超音波断層像である。その画像中の１７０は脂肪層、１９０は脂肪層１７０より深い組織領域である。まず脂肪層１７０の音速を求める。このために、脂肪層１７０の内部の適宜な深さ位置、例えば脂肪層１７０と組織領域の境界に近い計測位置に焦点位置を前述のキャリパのカーソルを用いて設定し、その値を計測するとともに演算・制御回路４１へ焦点位置を与える信号を供給する。その後、探触子より超音波パルスを送信し、その反射信号を受信する。脂肪層１７０の音速を求めるためには脂肪層からの反射信号のみを図２０に示す音速測定部へ取り込む必要がある。このために演算・制御回路４１にゲーティング機能を持たせる。このゲーティング機能は前記キャリパ機能に連動するようにすると良い。そして、このゲーティング機能によって脂肪層１７０からの反射信号をデジタル遅延部２００へ取り込み、前述の図２０に示す構成の動作説明に従って音速を求める。求める音速はある１点のものでも良いが、複数の点についての音速を求めて、それらから平均音速を求めることが望ましい。
次に、脂肪層１７０より深い組織領域の音速を求める。このときには、計測のための焦点位置を図２１に示す１９０の位置に設定する。この設定にも前述のキャリパ機能を用いる。そして演算・制御回路４１へ焦点位置を与える信号を供給し、その後、探触子より超音波パルスを送信し、その反射信号を受信する。組織領域の音速を求めるためには組織領域からの反射信号のみを取り込めれば良いのであるが、この場合の反射波は脂肪層１７０を通過せざるを得ないために、それは不可能である。そこで、前記ゲーティング機能によって、焦点位置からの反射信号を取り込む。そして、この反射信号から音速を求める。ここで求められた音速は、探触子面と計測点との間の平均音速を示すことと成る。
以上の二つの計測から組織領域の音速を求めることができる。最初に測定した脂肪層の平均音速をｃ_１、次に求めた脂肪層と組織領域の双方を含む平均音速をｃ_ａと符号を付ける。そして、脂肪層の厚さ、つまり焦点位置１８０の深度をｌ_ｆ、後者の計測の焦点深度をｌ_ａとすると、組織領域における音速ｃ_２は式１３により求められる。

これにより求まった脂肪層の音速ｃ_１と組織中の音速ｃ_２とを前述の式１以下へ適用することで、超音波のレンズ層、脂肪層及び組織中の音速を加味した屈折補正が可能となる。以上の計測値はモニタ画面へ数値等で表示されるようにする。図２１における表示例は、ｖ：１４５０／１５４０ｍ／ｓは脂肪層の音速が１４５０ｍ／ｓ，組織内の音速が１５４０ｍ／ｓを示し、ｔ：２／８ｃｍは脂肪層の厚みが２ｃｍ，組織内の焦点位置深さが８ｃｍであることを示している。
次に、他の実施例を述べる。この実施例は、上記音速計測に用いたキャリパの距離計測機能の補正を考慮するものである。すなわち、超音波撮像装置に組み込まれたキャリパの距離計測機能は、装置に初期設定された音速に基いた演算により成されるようになっている。したがって、式１３に用いたｌ_ｆ，ｌ_ａのキャリパによる距離測定値は実際の音速によるものに補正して使用することが望ましい。本実施例はこれに対応するものである。本実施例では、まず、超音波撮像装置を駆動して、関心領域（ＲＯＩ）を含む断面の超音波断層像を取得し、その断層像をモニタへ表示する。この状態で、図１９に示す超音波撮像装置の操作盤上に配置された屈折補正実行用スイッチ３１０をオンにする。この屈折補正実行スイッチ３１０は、オンすると屈折補正が実行され、オフ状態にしておけば屈折補正を行わない通常の撮像を可能とするものである。このスイッチ操作を行うとモニタ１２０の画面に２点のカーソルから成るキャリパ３００が表示される。そして、トラックボール叉はジョイスティック等の入力操作器を操作してキャリパ３００の一つのカーソルをを被検体の体表面へ、もう一方のカーソルを脂肪層の末端に移動し、入力情報固定キー（Ｅｎｔｅｒ ｋｅｙ）３２０を操作して計測される脂肪層を特定する。、次に，キャリパ３００における二つのカーソルのうち、前記脂肪層の末端に位置させたカーソルをそれより深い地点で関心領域内に移動し，前記キー３２０を操作して組織中における音速計測のための測定点を特定する。以上の２ステップの操作で入力した計測点のデータを制御部１１に読み込ませ、前述した前述の音速計測手法の計算が実行される。
装置設定音速がｖ_０，その音速ｖ_０でのキャリパ深度ｘ_０，計測された脂肪層の平均音速ｃ_ｆｍとすると、脂肪層の真の厚みｌ_ｒｔは式１４から求まる。

これらの値ｃ_ｆｍ、ｌ_ｆｔから組織中の真の平均音速ｃ_ｏｍ、組織中の真の音波伝播距離ｌ_ｏｔを求め式１３に適用することより屈折補正のための遅延時間データを演算することができ、そのフォーカスデータで超音波の送受信を行うと、良好な画像が得られる。
層構造を有する生体の音速と距離を求める方法としては上記の他に種々な方法を挙げることができる。一例として、超音波ビームのある特定のビーム、例えば中心のビーム着目し、そのビーム中の受信信号の強度が体表面近傍において非常に大きいところを検出し，それを時間として求め、前述の計算を自動的に行うようにしても良い。この計測方法は，層構造の境界では反射信号が大きいことから、検出のための閾値を決めておけば可能である。
以上述べた屈折補正法を装置へ自動シーケンスとしてプログラミングして組み込み、撮像中に数フレーム毎に繰り返して実行し、設定値を更新するようにしておけば，複数の検査部位がある場合のように探触子を移動させて検査していても屈折補正が自動的に行われるので、画像は常に良好となる。
屈折補正を可能とした装置では、表示画像に通常画像と屈折補正画像とを識別するマーキングを施す態様や、複数画像の同時表示法を用い、画面の左に通常の画像、右に屈折補正画像を表示する態様を採ることができる。着目点の画像が良好になっても，複雑な生体構成の場合，他の領域で画像が乱れる可能性がある。したがって，同時表示することにより実時間で屈折補正有り無しの画像を見比べることができるため、ユーザにとって有効となる。また，診断部位をＲＯＩとして指定し，その部位にのみ屈折補正を施した画像を取得し、それを通常の画像の中に嵌め込み合成して表示することも有用である。
本発明は、上記の特定の実施形態に限定されるものでなく、その技術思想の範囲を逸脱しない範囲で様々な変形が可能である。
以上述べたように、本発明によれば、レンズ層と脂肪層（あるいは筋肉層など）による屈折の影響を低減することができるので、超音波画像の画質向上が図れる。さらに、本発明によれば、層構造を有する被検体の各層の音速を計測して、その値を加味して超音波の各層における屈折の影響を低減することができるので、画質をさらに向上することができる。
【図面の簡単な説明】
図１は本発明の一実施形態による超音波撮像装置の概略構成例を示すブロック図、図２は図１に示す超音波撮像装置ないの超音波ビームフォーマの一実施形態を示すブロック図、図３は振動子アレイと超音波画像の関係を示す図、図４は超音波送受波時の振動子アレイと遅延時間の関係を示す図、図５は本発明の屈折補正の一例を適用した断層像撮像の流れを示すフローチャート、図６は画像表示部に表示された画像により層の厚みを計測する方法の一例を示す図、図７は超音波パルスの屈折状態を示す説明図、図８は本発明による屈折補正用の遅延時間計算の手順の第１の例を説明するフローチャート、図９は本発明による屈折補正用の遅延時間計算の手順の第２の例を説明するフローチャート、図１０はフェーズドアレイ型探触子における焦点位置と送波口径位置の関係を説明する図、図１１は超音波パルスの屈折状態の説明図、図１２は離散ニュートン法の説明図、図１３は本発明による屈折補正用の遅延時間計算の手順の第３の例を説明するフローチャート、図１４は本発明による屈折補正用の遅延時間計算の手順の第４の例を説明するフローチャート、図１５はコンベックス型探触子における超音波パルスの屈折状態の説明図、図１６は本発明による超音波撮像装置の他の実施形態の概略構成を示すブロック図、図１７は本発明による超音波撮像装置のさらなる他の実施形態の概略構成を示すブロック図、図１８は各振動子から焦点まで到達するのに要する時間を表すグラフ、図１９は屈折補正機能のためのスイッチを備えた装置概観図、図２０は屈折補正のための音速を求めるための装置構成を示すブロック図、図２１は音速測定の画面表示例を示す図である。Technical field
The present invention relates to an ultrasonic imaging apparatus that extracts an intra-subject tissue as an image using ultrasonic waves, and particularly relates to an ultrasonic beam former used therein.
Background art
An ultrasonic device, for example, an ultrasonic imaging device used for medical image diagnosis, obtains a tomographic image of a soft tissue of a living body or a blood flow image flowing in the living body using an ultrasonic pulse reflection method in almost real time. It can be displayed on a monitor for observation, and it is considered to be highly safe because it does not give radiation exposure to the subject as in the case of diagnostic imaging equipment that uses radiation. Applied.
In an ultrasound imaging apparatus, an ultrasound probe is used for transmitting ultrasound into the subject and receiving echo signals from the subject. One scanning method of the ultrasonic imaging apparatus is an electronic scanning method. The electronic scanning ultrasonic probe is arranged by arranging elongated rod-like transducers in a one-dimensional array, and driving each transducer with a predetermined delay time. Thereby, an ultrasonic beam that converges in a predetermined direction and a predetermined depth in the subject is transmitted from the ultrasonic probe. The received wave is synthesized by giving a predetermined delay time for each transducer to the received signal from each transducer of the ultrasonic probe, thereby capturing a received signal from a predetermined depth and direction. A processing portion in which a delay time is set for each transducer and the delay time is given to each transducer is called an ultrasonic beam former.
In order to obtain a good ultrasonic image by electronic scanning, it is necessary to transmit an accurate delay time for each transducer over the entire scanning range of the ultrasonic beam. This is because only by this method, it is possible to collect ultrasonic waves at a planned focal position without distorting or spreading. The reception echo signal is beam-formed by performing the same operation for reception. The ultrasonic probe is provided with a lens layer for converging ultrasonic waves generated from a large number of transducers arranged in a one-dimensional array in a direction perpendicular to the arrangement direction of the transducers. .
In particular, since the phasing circuit of an ultrasonic imaging apparatus has been digitized in recent years, it has become possible to easily and accurately control delay time and phasing in ultrasonic transmission / reception. However, the ultrasonic waves transmitted from each transducer do not collect at the focal point only through a medium having a constant sound velocity. The transmission pulse is converted from an electric signal to an ultrasonic wave by a piezoelectric vibrator, and then reaches a desired focal position through the lens layer and the biological tissue on the probe surface. At this time, since the sound velocity is different between the lens layer and the living tissue, sound refraction occurs in the direction of arrangement of the piezoelectric vibrators at the interface between the two. It is necessary to incorporate this refraction effect in the calculation of the delay time. Since the refraction angle is obtained by Snell's law, the delay time can be calculated by obtaining the refraction path of sound.
However, the ultrasonic imaging apparatus captures a tomographic image of about 30 frames per second, but the focal position is 50 in the horizontal direction (azimuth direction) and about 20 in the depth direction (distance direction) during the imaging of one frame. Need to change. The number of transducers in one aperture is about 32 to 192, depending on the probe. For these reasons, it is necessary to set the delay time for one vibrator at a considerably high speed even if it is calculated. However, there is a limit to the amount of calculation that can be performed in real time even in light of arithmetic processing performance such as MPU (Micro Processor Unit) or DSP (Digital Signal Processor) installed in an actual ultrasonic imaging apparatus. Therefore, in the current ultrasonic imaging apparatus, a method of storing data obtained by calculating the delay time for each focus position in advance for each probe in a storage medium such as a hard disk, or a calculation formula in time for real-time processing is used. One of the methods used to perform the approximate calculation is handled.
Currently, probes specialized for diagnosis target organs, parts, symptoms and the like are used as probes of ultrasonic imaging apparatuses, and the types of probes are increasing. Therefore, a considerably large capacity storage device is required to store the data of the delay time described above. However, it is not realistic for an ultrasonic imaging apparatus to be equipped with a storage device with an excessively large capacity due to price restrictions. In particular, difficulties arise when the ultrasonic imaging apparatus is provided with extensibility such that a probe developed later can be used. Further, in the two-dimensional probe that will become mainstream in the future, the number of transducers further increases, so that the amount of data to be stored is increasing.
On the other hand, the method for obtaining the delay time by the approximate calculation may be insufficient with the conventional calculation accuracy depending on conditions. Since the accuracy is determined by the ratio of the period and error time at the center frequency of the probe, a probe that transmits a pulse with a high center frequency, that is, a short period, like a high-definition probe is developed. As the delay time becomes more demanding, the accuracy becomes stricter. Therefore, an approximation method with higher accuracy is required. Also, considering application to a two-dimensional probe, it is necessary to shorten the calculation time for one transducer, so that going in a more severe direction is the same as in the former case.
The image used for the ultrasonic diagnosis, combined with the adoption of the digital phasing technique, has achieved a marked improvement in image quality compared to the conventional one, but the image obtained by the ultrasonic imaging apparatus is an X-ray apparatus, When compared with images obtained by other modalities such as an X-ray CT apparatus and an MRI apparatus, further improvement in image quality is desired.
As a measure for improving the image quality of the ultrasonic image, a technique for correcting the sound velocity of the ultrasonic wave in the subject has been recently developed. In the conventional ultrasonic imaging apparatus, delay control data for converging the ultrasonic wave to the focal position in the subject is set based on the average sound speed of the living body. Because there are individual differences such as those with a lot of fat, when the control data is used for all subjects, an image that can be said to be a good image on average is obtained, but the best image is obtained for each subject However, the improvement was desired. The above-mentioned sound speed correction technology is a countermeasure for this, and by measuring or estimating the ultrasonic propagation velocity in the body for each subject to be examined, the delay time control data for beam formation with that value. Is what you want. This sound speed correction technique is described in, for example, Japanese Patent Application Laid-Open No. 8-317923 and Japanese Patent Application Laid-Open No. 10-66694 filed by the present applicant.
However, it is considered that there is still room for improvement in the image quality of the ultrasonic image even if this sound speed correction technique is adopted. The inventors of the present invention believe that this can be improved by forming a beam by incorporating refraction in the ultrasonic wave propagation path. As refraction in the propagation path of ultrasonic waves, refraction by the lens layer is first cited as a problem, but the refraction of ultrasonic waves is not limited to the boundary surface between the lens layer and the subject. A living body that is a diagnostic target of the ultrasonic imaging apparatus is composed of various tissues such as fat, muscle fiber, various organs, and blood. Since the speed of sound is different in each of these tissues, the ultrasonic wave is slightly refracted at the boundary surface. Among these refractions, it is easiest to incorporate the refraction effect by the fat layer into the beam formation.
There are mainly three reasons for this. First, since the fat layer always exists under the skin, when an ultrasonic probe is applied transcutaneously to see any organ, the ultrasonic pulse penetrates the fat layer. That is, refraction by the fat layer is inevitable as long as a transcutaneous probe is used. Second, since the fat layer is on the outermost side in the living tissue, the influence of the shift to the focal position caused by refraction is the greatest. Third, subcutaneous fat can be assumed to have a certain thickness within the aperture of the probe, so it is easier to incorporate into the calculation than other tissues such as blood vessels. .
Disclosure of the invention
The present invention has been made in view of the above, and a first object thereof is to further improve the image quality of an ultrasonic image as compared with the current situation.
A second object of the present invention is to provide an ultrasonic imaging apparatus capable of setting a delay time in consideration of the influence of ultrasonic refraction by the lens layer and / or fat layer of an ultrasonic probe. is there.
Furthermore, a third object of the present invention is to set a delay time that can be set in consideration of the influence of refraction by the lens layer and / or the fat layer and the ultrasonic propagation velocity of each subject. An object is to provide an acoustic imaging apparatus.
When taking into account the refraction of ultrasonic waves by the fat layer, it is necessary to calculate the delay time in advance and store it in the storage device. Storage device is required. Moreover, when the sound speed of the fat layer is slightly different due to individual differences, it is advantageous to be able to calculate on the spot. Therefore, the present invention employs a technique for obtaining a delay time in consideration of the influence of refraction of ultrasonic waves by the lens layer and the fat layer, using an algorithm with sufficient approximation accuracy and quick calculation.
That is, an ultrasonic imaging apparatus according to a representative example of the present invention includes an ultrasonic probe including an array transducer, and transmission focusing or reception when transmitting or receiving ultrasonic waves to a subject. Delay control means for controlling a delay time for each transducer for performing focusing, and the transmission or transmission of the ultrasonic wave by incorporating an ultrasonic refraction effect by an ultrasonic propagation medium between the array transducer and a set focal position A refraction correction delay data generation unit that generates a delay time for performing focusing of received waves and supplies the delay time to the delay control unit, and a display unit that displays an ultrasonic image are provided. The delay control means stores delay time data obtained in advance based on the average sound speed of the living body, and ultrasonic transmission / reception for obtaining refraction correction data is performed in advance using the stored delay time data. . The refraction correction delay data generation means uses parameters related to the ultrasonic probe including the lens layer thickness of the probe, the sound speed of the lens layer, and the arrangement pitch of the transducers, and uses the parameters of the transducer and the designated focus. In consideration of the ultrasonic refraction effect in the ultrasonic propagation path, the delay time given to each transducer is obtained by calculation. The refraction correction delay data generation means includes a lens layer thickness, a sound speed of the lens layer, parameters relating to the ultrasonic probe including a pitch between transducers, a fat layer thickness of the subject and a sound speed of the fat layer, and a biological tissue. The delay time given to each transducer is obtained by calculation in consideration of the ultrasonic refraction effect in the ultrasonic propagation path between the transducer and the designated focal point. Further, the refraction correction delay data generation control means calculates a delay time using a parameter recursively solved from a parameter relating to a sound path from a transducer adjacent to a transducer to be calculated to the focal point. Calculate by calculation.
Further, the ultrasonic imaging apparatus of the present invention has a means for detecting data relating to the layer structure of the subject for refraction correction from the screen of the display unit on which the ultrasonic image is displayed, and an output of the layer structure data detecting means. And means for generating delay control data taking into account the influence of ultrasonic refraction due to the layer structure of the subject. The layer structure data detection means includes a caliper that displays two cursors movable on the screen and measures the distance between the cursors on the screen.
The ultrasonic imaging apparatus of the present invention is for performing an ultrasonic probe provided with an array transducer and a transmission focusing or a receiving focusing when transmitting or receiving ultrasonic waves to a subject. A delay control means for controlling a delay time for each vibrator; a means for measuring the thickness of the layered structure in the subject; a means for measuring the sound velocity of the layer structure portion; and the layer thickness measuring means. Using the measured layer thickness and the sound velocity in the layer structure measured by the sound velocity measuring means, each of the ultrasonic refraction effects in the ultrasonic propagation path between the transducer and the designated focal point is considered. A refraction correction delay control means for obtaining a delay time to be given to the vibrator and supplying the delay control means to the delay control means, and a display unit for displaying an ultrasonic image are provided. Then, the sound velocity measuring means obtains a delay time error of each reception channel by using an output of a delay circuit that performs phasing processing on echo signals received by a plurality of transducers, and the sound velocity in the subject is calculated from the delay error. Sound speed measuring means for obtaining The sound velocity measuring means includes means for specifying the sound velocity measurement region to the fat layer of the subject and the tissue part inside the fat layer and measuring the sound velocity for each.
Furthermore, the ultrasonic imaging apparatus of the present invention uses parameters related to the ultrasonic probe including the lens layer thickness of the ultrasonic probe, the sound velocity of the lens layer, and the transducer arrangement pitch, and the ultrasonic wave is emitted from the transducer. A program for causing a computer to execute a method of calculating a delay time given to each transducer in consideration of the effect of refraction until reaching a specified focal point, or the lens layer thickness, the sound velocity of the lens layer, Using the parameters related to the ultrasound probe, including the pitch between the transducers, and the fat layer thickness of the subject, the sound velocity of the fat layer, and the sound velocity data of the living tissue excluding the fat layer, the ultrasonic waves output the transducer. It is characterized by a built-in program for causing a computer to execute a method of calculating a delay time given to each transducer in consideration of the effect of refraction until reaching a specified focal point.
BEST MODE FOR CARRYING OUT THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a schematic configuration diagram showing an example of an ultrasonic imaging apparatus according to the present invention. The ultrasonic imaging apparatus 100 includes a main body 10 and an ultrasonic probe 20. The main body 10 of the ultrasonic imaging apparatus includes a control unit 11, a memory 12, an image display unit 13, and an ultrasonic beam former 14. An input unit 15 such as a keyboard or a pointing device is connected to the control unit 11. The ultrasonic probe 20 is detachably attached to the main body 10 of the ultrasonic imaging apparatus, and an appropriate one is selected and attached according to the diagnosis target of the subject 30. The ultrasonic probe 20 includes a transducer array 21 and a memory 22. The memory 12 of the main body 10 stores a table of data specific to the probe corresponding to the ID number of the probe, data such as the speed of sound in the subcutaneous fat layer and the speed of sound in the living tissue. The probe-specific data here is, for example, the thickness of the lens layer, the speed of sound in the lens layer, the transducer pitch, the radius in the case of convex probes, and the oblique angle in the case of oblique probes. It is a parameter etc.
The ultrasonic beam former 14 controls the operation of the transducer array 21 in the ultrasonic probe 20 under the control of the control unit 11. By using this ultrasonic beam former, ultrasonic image data is obtained by scanning the ultrasonic beam within the subject. The control unit 11 displays an ultrasonic tomographic image of the subject 30 on the image display unit 13 based on the obtained ultrasonic image data.
FIG. 2 is a block diagram showing an example of an ultrasonic beam former. The ultrasonic beam former includes an arithmetic / control circuit 41, a pulsar 42, a preamplifier and AD converter 43, a delay circuit 44, and an adder circuit 45. In the ultrasonic beam former 14, the delay time is calculated and controlled by the calculation / control circuit 41. The arithmetic / control circuit 41 may be realized by a computer and software. In that case, a program for delay time calculation is recorded in a recording medium such as a ROM, and the calculation / control circuit 41 is realized by a configuration in which the program is read by a computer.
The beamforming mechanism for both transmission and reception is as follows. First, each transducer 25 of the transducer array 21 included in the ultrasonic probe 20 is calculated from the arithmetic / control circuit 41. ₀ , 25 ₁ , 25 ₂ , ..., 25 _n A pulse signal is sent to the pulsar group 42 connected to at a timing shifted by the delay time of each transducer. Each pulsar 42 receiving this pulse signal immediately transmits the pulse signal to the vibrator 25 connected to itself. ₀ , 25 ₁ , 25 ₂ , ..., 25 _n Send to. Vibrator 25 ₀ , 25 ₁ , 25 ₂ , ..., 25 _n Generates an ultrasonic wave corresponding to the voltage due to its piezoelectricity, and transmits the ultrasonic wave into the subject 30. The ultrasonic waves reflected in the subject 30 return to the transducer array 21 of the ultrasonic probe 20 again, and the individual transducers 25. ₀ , 25 ₁ , 25 ₂ , ..., 25 _n Are converted into electric signals by piezoelectricity, amplified by a preamplifier and A / D converter 43 and converted into digital signals. This digital signal is phased and added by being added to the addition circuit 45 after being shifted by the delay circuit 44 adjusted by the signal from the arithmetic / control circuit 41.
Next, an ultrasonic beam scanning method by the transducer array 21 provided in the ultrasonic probe 20 will be described. In scanning the ultrasonic beam, as can be seen from the example of the ultrasonic diagnostic image shown in FIG. 3, it is necessary to move the focal point of the ultrasonic beam both in the depth direction and in the lateral direction. Here, in FIG. 3, when I is an ultrasonic image, the ultrasonic image I is a point F. _n1 , F _n2 , F _n3 , ..., F _nN (Where n is an integer from 1 to N). This point F _n1 , F _n2 , F _n3 , ..., F _nN The data connection is measured by one ultrasonic beam.
FIG. 4 shows the focus F ₁₁ Send an ultrasonic pulse toward the focus F ₁₁ It is a figure which shows typically the ultrasonic probe which detects the reflected signal from. FIG. 4 shows the state of wavefront movement at regular time intervals Ts. As shown in FIG. 4A, the focus of the ultrasonic probe is F. ₁₁ In order to match with this, the inner diameter transducer 25 of the ultrasonic probe is used. ₀ ~ 25 _n The ultrasonic pulse transmitted from the focal point F ₁₁ The ultrasonic waves are transmitted by giving a delay time to the operation of each transducer. In the case of FIG. 4A, the vibrators 25 located at both ends of the aperture. ₀ , 25 _n The delay time given to is j, and the j-th vibrator 25 _j Delay time T _j As shown in FIG. ₀ , 25 _n A virtual ultrasonic wavefront WA at the position of _j It is the time to reach the vibration surface. As described above, when an ultrasonic wave is transmitted by giving a delay time to the operation of each transducer, the wavefront WA has a focal point F as shown in FIG. ₁₁ Converge towards. This focus F ₁₁ The wave front WB of the reflected wave propagates as shown in FIG. ₀ , 25 _n Delay time T from time to reach _j Deviated by the amount of the vibrator 25 _j Reaches the surface of the vibrator. Therefore, the inner diameter transducer 25 of the ultrasonic probe ₀ ~ 25 _n Delay time T in each of _j By adding the phasing signal by shifting the amount of ₁₁ The reflected signal from can be picked up.
This transmission / reception operation is indicated by point F. _nn An image is acquired by performing for (n is an integer from 1 to N). In an actual apparatus, this transmission / reception operation is a method known as a dynamic focus method, that is, the transmission is performed by setting a focus on a specific depth located in a region of interest (ROI) within the subject. A single received beam signal is transmitted by continuously changing the focal position of the received wave from a shallow area to a deep area as the transmitted ultrasonic pulse travels through the subject. Is used. When the measurement of one beam is completed, the beam position is sequentially moved in the adjacent direction (azimuth direction), and similar transmission / reception operations are repeated to obtain N received beam signals to form an image.
There are two methods for moving the beam in the azimuth direction: a fixed transmission aperture and a variable transmission aperture. The former is a scanning method used for a sector type probe. In this method, all transducers of the probe are used for transmission and reception each time, and the beam direction is changed radially. In this case, the beam direction with respect to the transmission aperture is insignificant when the focal point of transmission / reception comes in front of the aperture. This has significant implications in later delay time calculations. The latter is a method used for a linear probe or a convex probe. Unlike the former, this method does not use all the transducers included in the probe for one transmission, for example, a probe with a total of 192 channels and a caliber of 64 channels for one transmission. Is used. In this case, when a beam at a certain position is obtained, transmission is performed with the aperture positioned in front of the beam, and the beam position is moved by moving the aperture. In contrast to the former case, transmission is performed so that the focal point is always in front of the aperture. How to wave.
Next, an example of the flow of tomographic imaging using the ultrasonic imaging apparatus of the present invention will be described with reference to the flowchart of FIG. Since the ultrasonic probe is detachable and exchangeable in the ultrasonic imaging apparatus, prior to imaging, in step 11, the ultrasonic probe 20 mounted on the main body 10 of the ultrasonic imaging apparatus is recognized. Do. This recognition is performed by reading parameters such as a probe ID number stored in the memory 22 of the ultrasonic probe 20.
Next, in step 12, a tomographic image of the subject is captured in a state in which the refractive index correction described later is not applied, that is, based on the delay control data obtained based on the average sound speed of the living body initially set in the apparatus. FIG. 6 is a schematic diagram of a diagnostic screen displayed on the image display unit 13 at this time, and an image of a somewhat blurred object is displayed because no correction is applied. The image shown in FIG. 6 is an image captured by a convex probe, and a subcutaneous fat layer is shown on the side in contact with the probe.
In step 13, the diagnostician measures and inputs this fat layer thickness. Specifically, it is measured using a caliper conventionally provided in an ultrasonic imaging apparatus. In FIG. 6, by positioning the caliper cursor at the place where the fat layer indicated by A starts and at the end where the fat layer indicated by B ends, the caliper measurement function allows the distance between A and B on the screen. Find the distance. (In the example shown, 13 mm) This value can be used as the fat layer thickness. After the fat layer thickness is input by the diagnostician, all processing is executed in the diagnostic apparatus.
Thereafter, the imaging mode is entered, and the diagnostician instructs the beam former 14 from the input unit 15 via the control unit 11 of the ultrasonic imaging apparatus in step 14 about the focal depth of transmission. In step 15, the beam former 14 calculates a delay time in which an arithmetic / control circuit 41 performs refractive index correction described later. The delay time calculated here is the delay time given to each transducer for transmission corresponding to the focal depth of the transmission, and the delay given to each transducer to form one ultrasonic reception beam. This is the time, that is, the delay time for continuously changing the focal position of the received wave as described above.
Subsequently, in step 16, when the transducer array of the ultrasonic probe is excited using the obtained transmission delay time, it converges to the focal point designated on the first ultrasonic beam line. An ultrasonic beam is transmitted. Next, in step 17, the received signal from the transducer array of the ultrasound probe is phased using the receiving delay time calculated in step 15, and added to obtain the first received beam signal. It is done.
In step 18, it is determined whether or not signal acquisition from all points in the imaging screen has been completed. That is, when an ultrasonic image is formed by N ultrasonic beams, it is determined whether or not an Nth address ultrasonic beam is obtained. If it is not completed, the process returns to step 14 to perform transmission / reception by changing the beam position. This is repeated, and when the entire imaging range has been scanned with respect to scanning in the depth direction and scanning in the horizontal direction, one diagnostic image is taken. When the diagnostic image has been taken, the process proceeds to step 19 where it is displayed on the image display unit 13 and the process ends. In addition, instead of displaying the data for all the points of one diagnostic image at a time after displaying all the points of the diagnostic image, every time the data about one received beam is acquired in step 17, the image is displayed on the image display unit. 13 may be sequentially displayed. This is possible by utilizing the function of the digital scan converter provided in the ultrasonic imaging apparatus.
Next, the processing content of “calculate the delay time after refractive index correction” in step 15 of the flowchart of FIG. 5 will be described in detail. Here, as an example, a case will be described in which the transmission surface of the probe has no curvature in the array direction, such as a linear or sector type probe. Other types of probes such as convex type can be handled in the same way just by changing the form of the expression.
Referring to FIG. 7, a vibrator 25 having a diameter is provided. _j After sending an electric pulse signal to the vibrator 25 _j Consider until the ultrasonic wave emitted from reaches the focal position. The ultrasonic probe has a transducer array arranged one-dimensionally at a transducer pitch p, and a thickness d arranged in front of the transducer array. ₁ This lens layer is provided. The lens layer is for converging the ultrasonic waves generated from the transducer array in a direction orthogonal to the transducer array (direction orthogonal to the paper surface of FIG. 7). Further, it is assumed that the speed of sound in the living tissue existing inside the fat layer of the living body is uniform on average, and the fat layer has a constant thickness d within the transmission aperture of the ultrasonic probe. ₂ Think of it as having In this way, the ultrasonic wave transmitted from the transducer propagates through the three layers of the lens layer, the fat layer, and the living tissue and reaches the focal point, and refraction becomes a three-layer problem.
Now, the j-th vibrator 25 located at a distance X from the center of the vibrator row. _j Let us consider a path through which the ultrasonic wave emitted from the laser beam reaches the focal point located at the depth F in front of the center of the transducer array. The electric pulse signal is generated by a piezoelectric vibrator 25 having a matching layer on the front surface. _j Is converted to ultrasound. Vibrator 25 _j The ultrasonic wave generated from the thickness d ₁ Distance x in the array direction of the transducer array ₁ The angle of incidence θ on the subcutaneous fat layer ₁ And refraction at the boundary between the lens layer and the fat layer (refraction angle θ ₂ ) The ultrasonic wave has a thickness d ₂ Distance x in the direction of the array of transducers ₂ The angle of incidence θ ₂ It enters the living tissue under the fat layer. Refraction at the interface between fat layer and living tissue (refraction angle θ ₃ ) Ultrasonic wave is a distance x in the living tissue in the arrangement direction of the transducer array. ₃ D in depth direction ₃ In the diagonal direction and reach the focal point in the living tissue.
The refraction of ultrasonic waves at the interface between each layer follows Snell's law. Therefore, the speed of sound in the lens layer is c ₁ C is the velocity of sound in the fat layer ₂ C represents the speed of sound in the body tissue. ₃ Then, the refraction angle θ ₁ , Θ ₂ , Θ ₃ Equation 1 holds between

Where the speed of sound c ₁ Is determined by the material of the lens and is known because it can be measured. Also, the speed of sound c ₂ , C ₃ However, in the following description, the following description is advanced under the condition that it is empirically known as an approximate value, and a method for obtaining it by measurement instead of the known value will be described later as another embodiment. Note that the speed of sound c ₁ Is stored in the memory 22 of the ultrasonic probe 20 and the speed of sound c ₂ , C ₃ Are held in the memory 12 of the ultrasonic imaging apparatus main body 10.
In the present invention, two methods are proposed as a numerical solution to this problem. First, the first method will be described below. When sin is deleted from Equation 1, Equation 2 is obtained.

However, Equation 2 cannot be solved analytically as it is. Here, Equation 2 is solved by three simultaneous equations in order to obtain an analytical solution. Here, as a variable to be left when the variable is dropped, the lateral movement amount x in the lens layer ₁ , Lateral displacement x in fat layer ₂ , Lateral movement amount x in the living body ₃ There is an option on which to leave, but the most advantageous under many conditions when considering the magnitude of the error is x ₃ Is the way to leave. This is because the speed of sound increases gradually in the lens layer, fat layer, and living tissue, so that the path of the sound wave that passes through each layer lies down side by side, and under many imaging conditions, Thickness d ₃ Depends on the largest. Therefore, in an actual system, the sound path has the largest amount of lateral movement in the living tissue. Since obtaining the largest amount directly by calculation is the method for minimizing the relative error, x under normal imaging conditions. ₃ If you ask for. However, although the lens layer is the thinnest regardless of the conditions, the thickness of the fat layer and the thickness of the living tissue may differ depending on the imaging conditions and the depth of the imaging target region. Therefore, the relationship between the fat layer thickness and the focal length ₂ Leave x ₃ There may be an algorithm that considers solving a univariate polynomial after leaving or determining the condition. However, the explanation of the method is omitted here and x ₃ Only the method of obtaining is described.
x ₁ , X ₂ Is eliminated from the simultaneous equations to obtain Equation 3. In Equation 3, x ₃ In the following, this is simply abbreviated as x and r ₂ = (C ₃ × c ₃ ) / (C ₂ × c ₂ ).

The calculation unit of the ultrasonic beam former in an actual ultrasonic imaging apparatus uses an MPU or DSP as described above. Of these, when a DSP is used, all calculations are converted to sums of products. In a typical DSP, the square root corresponds to 32 product sums. Since division depends on the value, the number of calculations is not known before the calculation. Therefore, transforming Equation 3 into a form without square root and division is important when using a DSP, and Equation 4 is the result of the operation.

According to Galois theory, it is known that there is no solution formula when a polynomial whose degree cannot be reduced by factorization is 5th or higher. From this, it is impossible to solve Equation 4 suddenly. However, although Equation 4 is a twelfth-order equation, it is a rational polynomial and can be differentiated over the entire range, so that it can be sufficiently approximated as a straight line in the vicinity of the solution. It is sufficient if there is x that satisfies even the condition that it is always in the vicinity of the solution.
Analyzing the behavior of Equation 4 in the vicinity of the solution, it is the square of both sides to eliminate the square root in the transformation from Equation 3 to Equation 4. ₁ , G ₂ Then g ₁ (X) -g ₂ G was looking for a solution of (x) ₁ (X) × g ₁ (X) -g ₂ (X) × g ₂ (X) = (g ₁ (X) -g ₂ (X)) x (g ₁ (X) + g ₂ Since the solution of (x)) is being searched, the solution of Equation 4 includes the original g ₁ (X) -g ₂ In addition to the solution of (x), g ₁ (X) + g ₂ The solution of (x) is also included. Since the operation of squares on both sides is performed twice, there are two extra unrelated solutions, but the solution newly included in Equation 4 by the second square is sufficiently original. This is not a problem here because it is far from the solution. g ₁ (X) + g ₂ Since Equation 4 has an inflection point in the vicinity of the solution by also having a solution of (x), the position for obtaining an approximate value needs to be closer to the true solution than the inflection point. As appropriate for the conditions, x + p × d using the notation in the figure ₃ / F is used as the starting point of the approximate solution. Then x which is x at the transducer position N _N Is obtained by the following equation (5). From here, the left side of Equation 4 is denoted as f (x).

Hereafter, a method for calculating the delay time for each transducer in real time using this recurrence formula will be described with reference to a flowchart. First, using the flowchart of FIG. 8, the delay time of the probe in which the line segment connecting the focal point and the center transducer of the aperture is orthogonal to the transmission surface, such as a linear probe or a convex probe, is calculated. The case of calculating will be described. Under this condition, since the focal point is in front of the aperture, the delay time of each transducer is symmetrical within the aperture. That is, it is sufficient to calculate only one half.
The parameters for the probe are already given in step 11 of FIG. 5, and the depth of focus value is already given in step 14 of FIG. Here, the delay time is calculated in order starting from the vibrator at the center of the aperture and moving outward one by one. First, in step 21, the calculation object is set as a central vibrator. The ultrasonic wave from the center transducer toward the focal point is incident perpendicularly to the interface of the layer and is not refracted at the interface. ₀ = 0. Also, the vibrator number N is 1 (the vibrator next to the center vibrator).
Next, in step 22, x ₀ = 0, so x for vibrator number N = 1 ₁ Is calculated by the recurrence formula 5. Next, the process proceeds to step 23, where x determined in step 22 ₁ Based on the above, the delay time τ for the vibrator having the vibrator number N = 1 is obtained by the following series of equations, that is, Equation 6.

Next, the process proceeds to step 24, where the vibrator number is incremented by one, the next vibrator is moved to, the process returns from step 25 to step 22, and the recurrence formula (formula 5) is used again. ₂ And the x ₂ Based on the above, the delay time τ for the vibrator of vibrator number 2 is calculated using Equation 6. This operation is performed for all the resonators in the aperture. If the determination in step 25 is “YES”, this means that the delay time for all the transducers has been calculated for this depth of focus, so the process proceeds to step 26 and the calculation result is output from the arithmetic / control circuit to the delay circuit. . When the next depth of focus to be calculated is given to the ultrasonic beam former, the calculation is performed in the same manner for the full-aperture transducer. By repeating this for all the depths of focus, imaging is performed in which deviation due to refraction is corrected.
Next, using the flowchart of FIG. 9, the line segment connecting the focal point and the center transducer of the aperture and the transmission surface are not orthogonal to each other as in the case of the oblique type probe, the sector type probe, or the phased array type probe. A method of calculating the delay time for a type of probe, or a delay time for a type of probe in which the angle between the line segment connecting the focal point and the center transducer of the aperture and the transmission surface changes A method of calculating the will be described. At this time, the relationship between the focal position, the aperture, and the transducer array is schematically shown in FIG.
(1) When the foot A of the perpendicular line dropped from the focal point F to the surface where the transducers are lined enters the transmission aperture
(2) When A does not enter the transmission aperture
There are two ways. In the latter case, an imaginary transducer array is considered by ticking from A to the transducer at the end of the aperture with an element pitch. Based on this idea, before entering the delay time calculation, that is, in FIG. 5, in the next stage of step 13, the virtual oscillator at the position A is defined as the first term of the recurrence formula by Formula 5. X is sequentially calculated to obtain x at the vibrator at the end of the aperture. Parameters relating to the probe such as the oblique angle are already given in step 11 of FIG. 5, and values such as the depth of focus and the focus direction are already given in step 14 of FIG.
In step 31 of FIG. 9, with respect to each focal depth and direction within the range to be imaged, it is calculated whether or not the perpendicular foot (point A in FIG. 10) dropped from the focal point to the transmission surface is within the aperture. To do. If it is within the caliber, the process proceeds to step 39 where the number of vibrators from the perpendicular foot to the right and left ends of the caliber is calculated. In step 40, the vibrator closest to the foot is determined as N = 1. , X ₀ = 0. Subsequently, in step 33, x is obtained from the recurrence formula (formula 5), and in step 34, the delay time τ is obtained from x by formula 6 until the position of the end of the aperture is increased while increasing the transducer number (S26). , S27). In this case, since symmetry cannot be used, it is necessary to calculate from the transducer of N = 0 to the end of the caliber on both sides as in Step 37, Step 41, and Step 40. This is because Fcos θ is not always an integral multiple of the transducer pitch.
If the perpendicular foot dropped from the focal point to the transmission surface is not in the aperture, the process proceeds from step 31 to step 32, where x at the transducer at the end of the aperture is read. Since this point A is the starting point for calculating the delay time, the focal length F is Fcos θ, d ₃ Uses the original F and d ₃ = Fcosθ-d ₁ -D ₂ Needs to be replaced.
Next, x at the end of the aperture is x ₀ And N = 1. Thereafter, in step 33, x is obtained from the recurrence formula (formula 5), and in step 34, the delay time τ is obtained from x by formula 6. Next, the process proceeds to the next vibrator in step 35, and the processing from step 33 to step 35 is repeated until it is determined in step 36 that the vibrator at the end of the aperture has been reached. Since the delay time for all the transducers has been calculated for the depth of focus and the focus direction, the calculation result is output from the arithmetic / control circuit to the delay circuit in step 38. When the depth of focus and the direction of focus to be calculated next are given to the ultrasonic beam former, the calculation is performed in the same manner for the full-aperture transducer. By repeating this for all the focal depths and the focal directions, imaging is performed with correction for deviation due to refraction.
By using the recurrence formula of the above formula 5, the approximate value of the delay time can be obtained with sufficient accuracy (1 / accuracy of 1/10 of the center frequency). This method has sufficient accuracy as compared with the conventional approximate expression, and the calculation speed is considerably faster because the algorithm does not have a loop as compared with the method for obtaining the accuracy by focusing on the iteration. Combined with this algorithm and the recent speedup of arithmetic processing devices, a real-time high-speed delay time calculation algorithm including refraction by a fat layer as in the present invention has been realized. Furthermore, compared with the calculation of a general recurrence formula, it is extremely advantageous in that errors do not accumulate. This is because between x and f (x), f ′ (x) because f (x) is a function that does not include approximation even if x for obtaining an approximate solution includes an error every time. Since there is no error in Equation 5, x in Equation 5 _N-1 The error at x is _N It is because it is not reflected in.
This way of thinking is a DSP-specific solution, but if there are no restrictions, the calculation is easier. Since Equation 3 can be regarded as a straight line in the vicinity of the solution, there is no need to obtain the slope from the differential coefficient, and x + p × d at the adjacent transducer as two points clearly sandwiching the solution. ₃ Using / F, x + p, a solution is obtained as an internal dividing point as in the following equation. Therefore, the processing speed can be improved by using the recurrence formula (formula 7) instead of the recurrence formula (formula 5).

Next, another embodiment of the ultrasonic beam former according to the present invention will be described with reference to FIG. In FIG. 11, the transducer position X _N Angle θ from the vibrator N in the lens layer _N It is assumed that the sound emitted at the point reaches the focal point F. This angle θ _N Since the refraction angle in each layer is obtained, the sound path is determined and the delay time can be determined. In order to obtain this delay time, conventionally, a sound is first emitted at an appropriate angle, and the side through which the sound passes is calculated. If the sound passes farther than the focus, θ _N Is increased and θ passes through the focal point. _N Make it smaller. By repeating this process until the sound passes through the focus, θ _N Seeking. However, with this method, the loop must be repeated a considerable number of times before convergence, and the calculation is not in time for the required speed. To achieve fast convergence, θ _N It is necessary to increase the way that changes are made.
The algorithm used in the ultrasonic beamformer according to the present embodiment is an improvement of the Newton-Raphson method so that it can be applied to a system having no explicit function. That is, it thinks as follows. First, assuming that an ultrasonic wave is emitted at an appropriate angle θ, the magnitude of the deviation is ΔX (θ). Since θ at which ΔX becomes 0 is a solution to be obtained, a deviation ΔX (θ + dθ) when radiated at an angle of θ + dθ is obtained in order to obtain a differential coefficient of ΔX. As shown in FIG. 12, in the space of the horizontal axis θ and the vertical axis ΔX, θ when a straight line passing through these two points crosses the x axis is a better approximate solution than the original θ. When this is expressed by an equation, it becomes an equation 8.

As the amount of deviation, in addition to using ΔX, a method using ΔY and a method using a square sum of ΔX and ΔY can be considered. Any of them can be used in the present invention, but in this embodiment, a case where ΔX is used will be described.
By repeating this operation, θ approaches the true solution. At this time, the smaller the difference between θ used first and the true solution, the smaller the number of repetitions of calculation, which is advantageous in terms of calculation time. In the present invention, when the delay time of the Nth vibrator is obtained as the initial value of θ here, θ obtained by the (N−1) th vibrator is obtained. _N-1 Is used. In this method, it was found that the practical calculation accuracy converges in two loops. Therefore, the number of loops is determined as 2 times for calculation. Since the number of loops is significantly smaller than that of the conventional method, it is significantly superior to the conventional method. When θ is obtained for the Nth vibrator, θ is obtained as an initial value for θ for the (N + 1) th vibrator. In the case of this method as well, when the focal point is in front of the vibrator, there is no refraction and θ = 0, which can be used as a starting point for the calculation.
A method for calculating the delay time of a probe in which the line segment connecting the focal point and the center transducer of the aperture is orthogonal to the transmission surface, such as a linear probe or a convex probe, using the flowchart of FIG. Will be described. Under this condition, since the focal point is in front of the aperture, the delay time of each transducer is symmetrical within the aperture. That is, it is sufficient to calculate only one half. The parameters for the probe are already given in step 11 of FIG. 5, and the depth of focus value is already given in step 14 of FIG.
In step 51 of FIG. 13, the ultrasonic wave emitted from the vibrator at the center of the aperture is perpendicularly incident on the boundary surface between the lens layer and the fat layer and the boundary surface between the fat layer and the living tissue and is focused without being refracted. Therefore, θ = 0 and N = 1 for the vibrator of vibrator number 0. Next, the process proceeds to step 52, where a deviation ΔX (θ) between the path when the ultrasonic wave is radiated from the transducer of number N = 1 at an angle θ and the target focal position is obtained. Here, as the angle θ, the angle θ with respect to the adjacent vibrator (vibrator of number 0) is used. In step 53, a deviation ΔX (θ + dθ) between the ultrasonic path and the target focal position when the ultrasonic wave is emitted from the vibrator of number 1 at an angle θ + dθ is obtained.
Here, in the case of a convex probe, unlike the linear probe, calculation is performed as follows. When observing an object with a convex probe, the probe is pressed against the object and the fat layer bends along the probe. Assuming that the fat layer is concentric with the structure of the convex probe, the sound path is as shown in FIG. The two methods of calculating the deviation Δ when passing through the vicinity of the element starting from the element and the method of obtaining the deviation Δ when passing through the vicinity of the element starting from the focus are the same in principle, but the total reflection condition Also, since the sound path cannot go to the left half of FIG. 15, this system is not symmetric, and the former method is not stable, so it is desirable to solve by the latter method. If the sound path is divided into each layer and analytically solved, the position and angle of the point where the sound enters the layer and the point that exits the layer from the angle are obtained, and the point that reaches the end by calculating in order of each layer And the deviation Δ can also be obtained. For example, when the sound path in the living body is obtained outside the fat layer, the position of the starting point of the sound is expressed as F = (x ₄ , Y ₄ ) P when exiting ₃ = (X ₃ , Y ₃ ), Let R be the curvature of the layer interface on the side where the sound comes out, and θ be the angle when the sound comes in. Then x ₄ , Y ₄ Is x from the simultaneous equation of Equation 9. ₄ It can be obtained by solving a solution which is always positive among ± 2 types of solutions of the quadratic equation in which is eliminated.

At this time θ ₄ For the triangle OFP ₃ It is obtained by thinking. In other words, now (x ₃ , Y ₃ ) To θ _p3 Is obtained, so θ ₄ = Θ + θ _p3 It can be obtained from the following formula. After θ ₄ From Snell's law ₃ In the same manner, a route when passing through a certain layer in order is obtained. In this example, the three-layer problem has been described as an example, but the present invention can also be applied when the number of layers is larger by the same method. Even if the interface between the fat layer and the living body outside the fat layer cannot be represented by a concentric circle of the probe, f (x ₃ ) Can be expressed in the form of Equation 10 using Equation 10 modified from Equation 9. ₄ Is θ + arctan (−f ′ (x ₃ )) Can be used instead.

Next, in step 54, a new θ is obtained by extrapolation from equation (8). Subsequently, the process returns from step 55 to step 52, and the deviation ΔX (θ) is obtained using the new θ obtained in step 54. In step 53, a deviation ΔX (θ + dθ) between the ultrasonic path and the target focal position when the ultrasonic wave is emitted at a new angle θ + dθ is obtained. Next, in step 54, a new θ is obtained by extrapolation from Equation 8.
Next, the process proceeds to step 56, where the delay time τ is calculated by the following equation 11 from the obtained θ. Next, the vibrator number is incremented by 1, and the processing from step 42 is repeated. If it is determined in step 58 that the calculation has been completed for all transducers within the used aperture, the process proceeds to step 59 and the calculation result is output.

The flowchart shown in FIG. 13 performs the recurrence formula calculation process in the flowchart shown in FIG. 8, the deviation calculation process from the focus position, the new θ calculation process by extrapolation, and the loop process of this part. This corresponds to the replacement with the determination processing of the number of loops.
Also in the case of this second method, when the focal point is not in front of the transmission aperture, θ at the end transducer is obtained in advance at the transmission aperture before imaging and stored. A plurality of θs for the transducers at the end of the transmission aperture are obtained for different subcutaneous fat layer thicknesses. When this condition is met during imaging, this θ is read and used to solve the recurrence formula as the first term of the recurrence formula. The calculation flow is shown in the flowchart of FIG. This detail is easily explained because the relationship between the calculation process of FIG. 8 and the calculation process of FIG. 9 and the relationship of the calculation process of FIG. 13 and the calculation process of FIG.
The method of calculating the parameters of the vibrator at the end of the caliber in advance is a measure for the DSP or MPU mounted in the current intermediate ultrasonic imaging apparatus having no margin, and is a high-class capable of mounting a high-speed arithmetic processing unit. In the machine, even when calculating in real time, it is also possible to perform all imaging and real time calculations including calculation that is wasted from the center of the virtual aperture at the position of the foot of the perpendicular line dropped from the focal point to the aperture surface. In that case, the calculation before imaging becomes unnecessary. Of course, in this case, it is not necessary to calculate the virtual transducers side by side with the same transducer spacing as the actual transducer, so it is effective to speed up the calculation by increasing the transducer spacing only in the virtual portion. It is. Since the calculation before imaging needs to be calculated every time the thickness of the fat layer changes, imaging stops every time. There are the following two methods for solving the problem.
The first method is a method in which all calculations for each fat layer thickness are performed in advance within a conceivable range, and all the calculation results before imaging are stored in the memory 22 of the ultrasonic probe 20 as data. is there. Then, when the ultrasonic probe 20 is connected to the ultrasonic imaging apparatus main body 10, the contents are transferred to the memory 12 of the ultrasonic imaging apparatus main body 10. During imaging, a delay time is calculated by the ultrasonic beam former 14 with respect to this parameter and the focal depth given by the control unit 11, the delay time is given to the transducer array 21, and transmitted to the subject 30. To do. The received wave is also phased by this delay time, and the control unit 11 calculates a diagnostic image and outputs it to the image display unit 13.
On the other hand, in the second method, data calculated in advance is put in a medium such as a CD-ROM and attached to the ultrasonic probe. Then, the data is installed in the memory 12 of the ultrasonic imaging apparatus body 10 from the CD-ROM or the like in advance or when using the probe. The operation after the installation is exactly the same as the first method.
FIG. 16 is a schematic configuration diagram illustrating another example of the ultrasonic imaging apparatus according to the present invention. 16, the same functional parts as those in FIG. 1 are denoted by the same reference numerals as those in FIG. In the explanation so far, the thickness of the subcutaneous fat layer has been input by the diagnostician. That is, as shown in FIG. 6, the distance is displayed on the screen when the diagnostician indicates the beginning and end of the fat layer on the diagnosis screen of the diagnostic apparatus. This value was entered by the diagnostician.
The ultrasonic imaging apparatus shown in FIG. 16 includes a fat layer thickness calculation unit 16 in the control unit 11 so that a screen output related to the subcutaneous fat layer can be directly input to the ultrasonic beam former 14. In the fat layer thickness calculation mode, as shown in FIG. 6, when the diagnosis person designates a location A and a location B where the subcutaneous fat layer starts with a pointing device or the like, the fat layer thickness calculation unit 16 calculates the distance between them, The calculation result is transmitted to the beam former 14.
Next, a method for adaptively improving the image quality during imaging using the ultrasonic beamformer mounted algorithm of the present invention will be described. The calculation algorithm described so far has the premise that the fat layer is uniform and has a constant thickness, and that the speed of sound in the fat layer does not vary depending on the individual or target site. This assumption was buried in the error when the delay time calculation had a certain error, and the effect was not understood. However, when high-accuracy imaging is possible according to the present invention, it is clear that the image quality is improved by removing this premise. In order to remove the premise, it is only necessary to change the sound speed of the fat layer while imaging.
FIG. 17 is a schematic configuration diagram illustrating another example of the ultrasonic imaging apparatus according to the present invention. 17, the same functional parts as those in FIG. 1 are denoted by the same reference numerals as those in FIG. This ultrasonic imaging apparatus makes it possible to finely adjust parameters relating to the fat layer, and to deal with a fat layer that is not uniform and has a constant thickness. Specifically, the ultrasonic imaging apparatus having the configuration shown in FIG. 1 includes an input device and a parameter calculation unit 17 so that the sound velocity of the fat layer held by the ultrasonic beam former can be varied. In this apparatus, the input device and parameter calculation unit 17 calculates a parameter corresponding to the input value and inputs the value to the ultrasonic beam former 14. The effect of changing this parameter is explained as follows. Considering the case where there is no refraction, the time required to reach the focal point from each transducer is calculated as shown in Equation 12. This is shown in a graph in FIG.

At this time, if one or more of the parameters in Equation 12, that is, the speed of sound, the pitches p and F, or the like is changed, the curvature of the graph changes as shown in FIG. This is the same in principle as the current problem is complicated even in the problem that should be taken into account of the lens layer, fat layer, biological tissue other than the fat layer and refraction made of three layers, as in the present problem. Since the delay time is determined by the time required to reach the focal point from each transducer, “vertical axis: the time required to reach the focal point from each transducer” vs. “horizontal axis: the position of each transducer” The delay time is expressed by the relationship curve. Therefore, as described above, when the sound velocity of the fat layer is not constant depending on the location, this curve is distorted, so it is possible to improve defocus by changing various parameters to curve fit to it. It becomes possible. The curve fitting is most preferably adjusted so that the imaging screen is optimized, and the operability is good when the parameters are controlled by an external knob or the like. At this time, it is also effective to adjust the thickness of the fat layer or the pitch of the vibrator instead of the sound speed of the fat layer and appropriately change these plural parameters in conjunction with each other. In FIG. 18, (a) shows the case where the vibrator pitch and the sound speed of the fat layer are not changed, and (b) and (c) show the case where the vibrator pitch and the fat layer sound speed are adjusted.
From here, an example of an example in which a layer structure exists on the surface of the subject will be described. First, a specific example for obtaining the thickness and sound speed of the layer structure on the surface of the subject during the examination will be described.
FIG. 19 shows an embodiment in which the examiner can select them. A thickness table and a sound velocity table shown in FIG. 19B are prepared in advance and incorporated into the apparatus, and the examiner observes the image 170 displayed on the monitor 120 and obtains it by referring to the depth scale 160. The thickness of the layer is read from the result, and a close thickness is selected by the switch SW2 (140) or the like, and the sound speed is the same as that of the switch SW1 (150). In this case, the switch SW1 and the switch SW2 can be operated independently, and values that the examiner considers appropriate can be selected as the respective values. Regarding the speed of sound, a numerical value may be used, or a fat layer or a muscle layer may be selected. For example, when the fat layer 170 can be read as approximately 2 cm from the scale 160 in the depth direction indicated on the screen in the screen of FIG. 19, 2 cm is selected by the layer thickness setting switch 140 provided on the console 130. Then, the thickness (Thickness) is displayed as 2 cm on the screen. Next, the sound speed setting switch 150 is operated in a predetermined direction to select the sound speed. In order to confirm the selected values, the input values may be displayed on the screen. The display may be a numerical value such as 1450 m / s, but may also be an expression such as hard muscular, muscular, normal, fat, or high fat. Also, the switch may be rotary or anything. Of course, a touch panel may be used.
In the above description, the embodiment of removing the influence of refraction of ultrasonic waves generated from the layer structure of the living body and improving the image quality of the ultrasonic image has been described. In order to further improve the image quality of the ultrasonic image, It is necessary to accurately grasp the ultrasonic wave propagation speed in the living body and reflect the data in the refraction correction control in the ultrasonic beam former.
Thus, next, another embodiment of the present invention is shown in FIG. In this method, the sound speed of a living body having a layer structure is obtained, and the refraction correction processing of the above-described embodiment is performed based on the sound speed. FIG. 20 is a block diagram showing an embodiment of a portion for obtaining the sound speed in the ultrasonic imaging apparatus. In the figure, reference numeral 200 denotes a digitally controllable digital delay unit that delay-controls a plurality of ultrasonic signals received by a probe and outputs a received beam signal. The circuit has a number of channels corresponding to the number of transducers used for reception. have. A delay data error estimation unit 210 receives a plurality of signals delayed by the digital delay unit 200 and estimates an error of the digital delay data subjected to delay control by calculation with respect to the true delay data. It is a digital delay control unit that controls the operation of each channel.
In addition to the above configuration, the ultrasonic imaging apparatus of the present embodiment further includes a delay error recording unit 230 that stores delay times due to a plurality of medium sound velocities in advance, and a delay error obtained by the delay error estimation unit 210. A new delay time is calculated from the delay time, compared with a value stored in the sound speed corresponding delay time recording unit 230, and a delay time comparing unit 240 that outputs data closest to the stored value, and a sound speed corresponding delay time recording The sound speed data recording unit 250 that records what medium sound speed the delay time data stored in the unit 230 records, and the delay time recording unit from the recording location of the delay time that matches the output of the delay time comparing unit 240 And a medium sound speed selector 260 for selecting the medium sound speed. The output line of the medium sound speed selection unit 260 is connected to the calculation / control circuit 41 shown in FIG. 2, and the digital delay control unit 220 is connected to be controlled by the calculation / control circuit 41.
As described above, in the configuration shown in FIG. 2, the delay circuit can be digitally controlled, the delay error estimation unit 210 is provided at the output thereof, the digital delay control unit 220 is added to the arithmetic / control circuit 41, and As described above, the ultrasonic imaging apparatus according to the present embodiment is almost capable of achieving the true medium sound speed by newly providing the sound speed corresponding delay time recording section 230, the delay time comparing section 240, the sound speed data recording section 250, and the medium sound speed selecting section 260. Refraction correction with equal sound speed is possible.
Next, the operation of the portion configured as described above will be described. First, delay time data obtained by assuming that the sound speed in the living body is, for example, a constant sound speed of an average value of the living body is output from the arithmetic / control circuit 42 to the digital delay unit 2 to transmit ultrasonic waves into the subject. . At this time, the focus of transmission is set to an appropriate depth. Then, delay time data D based on the same sound speed as that of the transmitted wave is supplied from the arithmetic / control circuit 41 to the digital delay unit 200 via the digital delay unit 220 to delay control the received signal.
Then, a delay-controlled signal of each channel of the digital delay unit 200 and a signal before addition processing for phasing is input to the delay error estimation unit 210, and an error ΔD with respect to the used delay time data D is input. ₁ Is calculated as an estimated value, and D _c1 = D + ΔD ₁ Is output. As this calculation method, the correlation processing method disclosed in the above-mentioned JP-A-8-317923 can be used. D output from the delay error estimation unit 210 _c1 Is input to the delay time comparison unit 240. Then, the delay time comparison unit 240 receives the data recorded in the sound speed corresponding delay time recording unit 230 and the input data D. _c1 And D _c1 The delay time data closest to is selected and output to the medium sound speed selection unit 260. When data is input, the medium sound speed selection unit 260 selects how much the sound speed of the input data is. This selection can be made by associating the storage addresses of the data of the sound speed corresponding delay time recording unit 230 and the sound speed data of the sound speed recording unit 250. Therefore, the sound speed corresponding delay time recording unit 230 and the sound speed recording unit 250 can be combined into one.
The sound speed data selected by the medium sound speed selector 260 is fed back to the digital delay controller via the arithmetic / control circuit 41. This feedback circuit is for repeating the above operation in order to obtain an accurate sound speed, and issues a repeat execution command for the above operation from the arithmetic / control circuit 41 as necessary. With the above operation, the sound speed in the living body can be measured as an estimated value.
In order to simplify the comparison operation in the delay time comparison unit 240, the data D output from the delay error estimation unit 210. _c1 And the data recorded in the sound velocity corresponding delay time recording unit 230 are two-dimensional distribution data, and it is useful because the amount of information to be handled can be reduced. Furthermore, it is useful to obtain a difference in delay time distribution and fit a linear line to the difference delay time sequence because the amount of information to be handled is further reduced.
Next, a method of obtaining the sound speed of each layer of a living body having a layer structure with the configuration shown in FIG. FIG. 21 is an ultrasonic tomogram of the subject displayed on the monitor of the ultrasonic imaging apparatus. In the image, 170 is a fat layer, and 190 is a tissue region deeper than the fat layer 170. First, the sound speed of the fat layer 170 is obtained. For this purpose, a focal position is set at an appropriate depth position inside the fat layer 170, for example, a measurement position close to the boundary between the fat layer 170 and the tissue region by using the caliper cursor, and the value is measured. A signal for giving a focal position to the arithmetic / control circuit 41 is supplied. Thereafter, an ultrasonic pulse is transmitted from the probe, and the reflected signal is received. In order to obtain the sound speed of the fat layer 170, only the reflected signal from the fat layer needs to be taken into the sound speed measuring unit shown in FIG. For this purpose, the arithmetic / control circuit 41 is provided with a gating function. This gating function is preferably linked to the caliper function. Then, the reflected signal from the fat layer 170 is taken into the digital delay unit 200 by this gating function, and the sound speed is obtained according to the operation description of the configuration shown in FIG. Although the sound speed to be obtained may be one point, it is desirable to obtain the sound speed at a plurality of points and obtain the average sound speed from them.
Next, the sound speed of the tissue region deeper than the fat layer 170 is obtained. At this time, the focus position for measurement is set to a position 190 shown in FIG. The caliper function described above is also used for this setting. Then, a signal for giving a focal position is supplied to the arithmetic / control circuit 41, and then an ultrasonic pulse is transmitted from the probe and its reflected signal is received. In order to obtain the sound speed of the tissue region, it is only necessary to capture the reflected signal from the tissue region, but this is impossible because the reflected wave in this case must pass through the fat layer 170. Therefore, the reflected signal from the focal position is captured by the gating function. Then, the speed of sound is obtained from the reflected signal. The sound speed obtained here indicates an average sound speed between the probe surface and the measurement point.
The sound speed of the tissue region can be obtained from the above two measurements. The average sound velocity of the fat layer measured first is c ₁ Next, the average velocity of sound including both the fat layer and the tissue region obtained is c _a And sign. Then, the thickness of the fat layer, that is, the depth of the focal position 180 is set to l _f , The depth of focus for the latter measurement _a Then, the speed of sound c in the tissue region ₂ Is obtained by Equation 13.

The speed of sound c of the fat layer determined by this ₁ And sound velocity c in the organization ₂ Is applied to the above formula 1 and below, it is possible to perform refraction correction in consideration of the speed of sound in the ultrasonic lens layer, fat layer and tissue. The above measured values are displayed as numerical values on the monitor screen. In the display example in FIG. 21, v: 1450/1540 m / s indicates the sound velocity of the fat layer is 1450 m / s, and the sound velocity in the tissue is 1540 m / s, and t: 2/8 cm indicates the thickness of the fat layer is 2 cm. It shows that the focal position depth is 8 cm.
Next, another embodiment will be described. This embodiment considers correction of the caliper distance measurement function used for the sound velocity measurement. That is, the caliper distance measurement function incorporated in the ultrasonic imaging apparatus is performed by a calculation based on the sound velocity that is initially set in the apparatus. Therefore, l used in Equation 13 _f , L _a It is desirable to use the measured distance of the caliper by correcting it to the actual sound speed. This embodiment corresponds to this. In this embodiment, first, the ultrasonic imaging apparatus is driven, an ultrasonic tomographic image of a cross section including the region of interest (ROI) is acquired, and the tomographic image is displayed on the monitor. In this state, the refraction correction execution switch 310 disposed on the operation panel of the ultrasonic imaging apparatus shown in FIG. 19 is turned on. The refraction correction execution switch 310 executes refraction correction when it is turned on, and enables normal imaging without refraction correction when it is turned off. When this switch operation is performed, a caliper 300 composed of two cursors is displayed on the screen of the monitor 120. Then, by operating an input operation device such as a trackball or joystick, one cursor of the caliper 300 is moved to the body surface of the subject and the other cursor is moved to the end of the fat layer, and an input information fixing key (Enter) is moved. key) The fat layer to be measured is specified by operating 320. Next, of the two cursors in the caliper 300, the cursor positioned at the end of the fat layer is moved into the region of interest at a deeper point, and the key 320 is operated to measure the speed of sound in the tissue. Specify the measurement point. The data of the measurement point input by the above two-step operation is read by the control unit 11, and the above-described calculation of the sound speed measurement method is executed.
Device set sound speed is v ₀ , The speed of sound v ₀ Caliper depth x ₀ , Average sound velocity of fat layer c measured _fm Then, the true thickness of the fat layer l _rt Is obtained from Equation 14.

These values c _fm , L _ft To true average speed of sound c in the organization _om , The true acoustic propagation distance in tissue _ot By applying the above to Equation 13, delay time data for refraction correction can be calculated, and when ultrasound is transmitted / received using the focus data, a good image can be obtained.
In addition to the above, there are various methods for obtaining the sound speed and distance of a living body having a layer structure. As an example, paying attention to a specific beam of the ultrasonic beam, for example, the central beam, detecting a place where the intensity of the received signal in the beam is very large near the body surface, obtaining it as time, and calculating the above-mentioned calculation You may make it carry out automatically. This measurement method is possible if the threshold value for detection is determined because the reflection signal is large at the boundary of the layer structure.
If the refraction correction method described above is programmed and incorporated into the apparatus as an automatic sequence and is repeatedly executed every several frames during imaging, and the setting value is updated, as in the case where there are a plurality of examination sites. Even if the probe is moved and inspected, refraction correction is automatically performed, so the image is always good.
In a device that enables refraction correction, the display image is marked with a marking that distinguishes between a normal image and a refraction correction image, and a multiple image simultaneous display method is used. The mode which displays can be taken. Even if the image of the point of interest becomes good, the image may be distorted in other areas in the case of a complex biological structure. Accordingly, simultaneous display enables comparison of images with and without refraction correction in real time, which is effective for the user. It is also useful to designate a diagnostic site as an ROI, acquire an image with refraction correction applied only to that site, fit it into a normal image, synthesize and display it.
The present invention is not limited to the specific embodiment described above, and various modifications can be made without departing from the scope of the technical idea thereof.
As described above, according to the present invention, the influence of refraction by the lens layer and the fat layer (or muscle layer, etc.) can be reduced, so that the image quality of the ultrasonic image can be improved. Furthermore, according to the present invention, the sound velocity of each layer of the subject having a layer structure can be measured, and the influence of refraction in each layer of ultrasonic waves can be reduced by taking the value into account, thereby further improving the image quality. be able to.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration example of an ultrasonic imaging apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram showing an embodiment of an ultrasonic beamformer without the ultrasonic imaging apparatus shown in FIG. 3 is a diagram showing the relationship between the transducer array and the ultrasonic image, FIG. 4 is a diagram showing the relationship between the transducer array and the delay time during ultrasonic transmission / reception, and FIG. 5 is a tomographic image to which an example of refraction correction of the present invention is applied. FIG. 6 is a diagram illustrating an example of a method for measuring the thickness of a layer from an image displayed on the image display unit, FIG. 7 is an explanatory diagram illustrating a refraction state of an ultrasonic pulse, and FIG. FIG. 9 is a flowchart for explaining a first example of a procedure for calculating a delay time for refraction correction according to the present invention, FIG. 9 is a flowchart for explaining a second example of a procedure for calculating a delay time for refraction correction according to the present invention, and FIG. In phased array transducers FIG. 11 is an explanatory diagram of the refraction state of an ultrasonic pulse, FIG. 12 is an explanatory diagram of the discrete Newton method, and FIG. 13 is a delay time calculation for refraction correction according to the present invention. FIG. 14 is a flowchart for explaining a fourth example of the procedure for calculating the delay time for refraction correction according to the present invention, and FIG. 15 is a flowchart for explaining the ultrasonic pulse in the convex probe. FIG. 16 is a block diagram showing a schematic configuration of another embodiment of the ultrasonic imaging apparatus according to the present invention, and FIG. 17 shows a schematic configuration of still another embodiment of the ultrasonic imaging apparatus according to the present invention. FIG. 18 is a graph showing the time required to reach the focal point from each transducer, FIG. 19 is an overview of the apparatus having a switch for the refraction correction function, and FIG. 20 is the sound speed for refraction correction. Block diagram showing an apparatus configuration for determining, FIG. 21 is a diagram showing a screen display example of sound speed measurement.

Claims (8)

Ultrasound probe with array transducers and delay control to control the delay time for each transducer to transmit or receive ultrasound when transmitting or receiving ultrasound to the subject And a delay time for focusing the transmitted or received wave by incorporating an ultrasonic refraction effect by an ultrasonic wave propagation medium between the array transducer and the set focal position and generating the delay time to the delay control unit In an ultrasonic imaging apparatus comprising a refraction correction delay data generating means to be supplied and a display unit for displaying an ultrasonic image,
The refraction-corrected delay data generation control means calculates a delay time by using a parameter recursively solved from a parameter related to a sound path from a transducer adjacent to a calculation target transducer to the focal point. What is claimed is: 1. An ultrasonic imaging apparatus comprising:   The delay control means stores delay time data obtained in advance by the average sound speed of the living body, and ultrasonic transmission / reception for obtaining refraction correction data is performed in advance using the stored delay time data. The ultrasonic imaging apparatus according to claim 1, wherein the ultrasonic imaging apparatus is characterized.   The refraction correction delay data generation means uses parameters relating to the ultrasonic probe including a lens layer thickness, a sound velocity of the lens layer, and an arrangement pitch of the transducers, and an ultrasonic wave between the transducer and a designated focal point. The ultrasonic imaging apparatus according to claim 1, wherein a delay time given to each transducer is obtained by calculation in consideration of an ultrasonic refraction effect in a propagation path.   The refraction correction delay data generation means includes a lens layer thickness, a sound speed of the lens layer, parameters relating to the ultrasonic probe including a pitch between the transducers, a fat layer thickness of the subject, a sound speed of the fat layer, and a sound speed of the living tissue The delay time given to each transducer is obtained by calculation in consideration of the ultrasonic refraction effect in the ultrasonic propagation path between the transducer and the designated focal point. 3 is an ultrasonic imaging apparatus according to 3.   A layer thickness measuring means for measuring the thickness of the layered structure in the subject; and a sound speed measuring means for measuring the sound speed of the layer structure portion, wherein the refractive correction delay control means is the layer thickness measuring means. Each vibration based on the ultrasonic refraction effect in the ultrasonic propagation path between the transducer and the designated focal point using the layer thickness measured by the sound velocity and the sound velocity in the layer structure measured by the sound velocity measuring means. 2. The ultrasonic imaging apparatus according to claim 1, wherein a delay time given to the child is obtained and supplied to the delay control means. 6. The ultrasonic imaging apparatus according to claim 5, wherein the layer thickness measuring unit includes a caliper that displays two cursors movable on the screen and measures a distance between the cursors on the screen. .   The sound velocity measuring means obtains a delay time error of each reception channel by using an output of a delay circuit that performs phasing processing on echo signals received by a plurality of transducers, and obtains a sound velocity in the subject from the delay error. The ultrasonic imaging apparatus according to claim 5, further comprising a sound speed measuring unit.   8. The sound speed measuring means includes means for measuring a sound speed for each of the sound speed measurement regions specified to a fat layer of a subject and a tissue part inside the fat layer. Ultrasonic imaging device.

Images