CN102266223B - Pain evaluation system based on magnetic resonance resting state functional imaging - Google Patents
Pain evaluation system based on magnetic resonance resting state functional imaging Download PDFInfo
- Publication number
- CN102266223B CN102266223B CN 201010189516 CN201010189516A CN102266223B CN 102266223 B CN102266223 B CN 102266223B CN 201010189516 CN201010189516 CN 201010189516 CN 201010189516 A CN201010189516 A CN 201010189516A CN 102266223 B CN102266223 B CN 102266223B
- Authority
- CN
- China
- Prior art keywords
- pain
- attribute
- formation
- origin cause
- brain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses a pain evaluation system based on magnetic resonance resting state functional imaging, and the pain evaluation system comprises a pain characteristic specimen library for storing history resting state functional imaging attributes of specimens of patients with different pain levels and pain causes and control normal persons; a functional imaging unit for obtaining resting state functional imaging data from brains of new individual patients suffering from chronic pain and transmitting to an attribute extraction unit; the attribute extraction unit for extracting resting state functional image attributes of new individuals from the resting state functional imaging data; and a pain evaluation unit for comparing the resting state functional image attributes of new individuals and the history resting state functional image attributes in the pain characteristic specimen library, and evaluating pain levels and/or pain causes of new individuals. Hereby, the pain evaluation system disclosed by the invention can objectively and accurately evaluate pain levels and/or pain causes through objective image data of human brains.
Description
Technical field
The present invention relates to a kind ofly evaluate the pain grade of chronic pain patient and/or the iconography system of the pain origin cause of formation based on magnetic resonance tranquillization attitude functional imaging data.
Background technology
Pain is one of human modal misery, also is one of the most insufferable symptom of patient.IASP (IASP) is defined as pain " discomfort that necessary being or potential tissue injury or analogue are brought and emotional experience ".This definition has emphasized that pain is not only a kind of sentience, and is a kind of various dimensions phenomenon, comprises sensation, emotion, motivation, environment and cognitive component.IASP is defined as chronic pain " pain of paranormal organization healing time (being generally 3 months) ".
Chronic pain has not only brought misery to the individual, and because the occurrence rate of chronic pain among the crowd is higher, entire society has also been caused serious financial burden.Data shows, have 35% to suffer from chronic pain among the U.S. common people, wherein surpass 500,000 Americans because chronic pain becomes partially or completely disabled and can't work, annual because the total output value loss that chronic pain causes is 65,000,000,000 dollars, the medical treatment cost is 75,000,000,000 dollars; China expert in July, 2003 China's chronic pain survey result is shown: the chronic pain patient of going to see a doctor in six cities in month reaches more than 130,000 people; The pain distribution of grades for slightly account for 9%, moderate accounts for 68.1%, severe accounts for 22.9% (data from Tianjin medical association pain credit meeting website).Because evaluation criteria after the objective diagnosis of chronic pain shortage and the treatment, the selection of its therapeutic scheme is mainly based on patient's subjectivity statement and doctor's individual clinical experience, under the situation of chronic pain, the pathological changes that continues: such as arthritis or nerve injury, nonvolatil damage: such as amputation, the capital causes lasting pain, and may not use usually feasible method to go to treat this pain.More difficultly, under many circumstances, can not determine to cause the pathology of constant pain, in this case, the side effect that analgesic brings may surpass its curative effect.
For the evaluation of chronic pain, be subjected to its character and existing diagnostic means (in addition the information processing technology is not auxiliary with diagnostic techniques merely) limitation all can't its pain essence of objective quantification and the related neural basis change.
The pathomechanism of pain comprises mental mechanism and physiological mechanism.The persistent period of chronic pain may with the nocuity perception that continues and the nervous system that brings out thereof change relevant, but most of pain researcheres unanimously think psychological factor or Nervous and Mental Factors generation, the development of chronic pain, continue or increase the weight of in play key effect.In the research of pain, find already between noxious stimulation and the pain sensation to be not simple response relation, stimulation degree and pain grade are also not the same, and pain still can come from non-noxious stimulation, and these phenomenons show that pain and psychological process have substantial connection.Psychological factor all exerts an influence to perception, resolution and the extent of reaction in characteristic of pain, degree, time and space, and is reflected on the links of pain." gate control theory " that Melzack proposes helps further to understand the effect of psychological factor in pain.The physiological mechanism of chronic pain is very complicated, relates to each nervous system, neurotransmitter and biochemical substances.Through years of researches, chronic pain generally is considered to be made of three distinguishing but overlapped origin causes of formation now: nociception, neuropathic and psychogenic, these all factors can be on people's pain perception and performance generation impact in various degree.Nociceptive pain is the most general types of pain, and it is considered to the signal by the bodily tissue generation of pathological changes.Therefore as arthritis, calculus and tumor all can cause the signal input from the increase of damaged tissues, and this is perceived as pain.Although these input signals may be revised by the physiological process of brain and spinal cord (raising or reduction), pain still arrives the central nervous system owing to the signal input that increases basically.Be different from nociceptive pain, the second pain origin cause of formation, neuropathic pain be impaired by periphery or central nervous system due to.In a narrow sense, this pain occurrence is when nerve feature in disorder occurs, and is impaired such as sense organ or nervus motorius, this class factor takes on a different character, and it is described to a prickling sensation, electric-shock feeling usually, burn feeling or tingling, present imaging technique is difficult to detect its pathological changes.The third pain origin cause of formation, the psychological origin cause of formation, it acts in people's the pain experience widely, and psychological process can affect the impression of pain and the expression of pain (afunction) in the mode that increases or alleviate.In football match, many sportsmen can note less than or ignore unexpected seriously injuredly, and in this case, although damaged tissues has a large amount of signals inputs, pain and afunction all are minimum.Cause the process of pain although people have been familiar with these widely, still be difficult to especially distinguish at present each role in three kinds of pain origin causes of formation.And the treatment means that the dissimilar pain origin cause of formation is taked and mode also are not quite similar, and some means and medicine do not use in correct situation then can produce curative effect hardly.Therefore, the best therapeutic scheme of pain is depended on the different origins of accurate differentiation pain.Present subjective diagnostic techniques (for example estimating scale, pain questionnaire table) can only solely be evaluated the grade of pain, can not draw the character of pain; And this method is restricted for the subjective patient who expresses of forfeiture.Present objective evaluation method (for example behavior determination method, physiology's algoscopy) all is from different angles pain indirectly to be evaluated, and is reliable not and can only provide limited information, and the quantification of pain also is restricted.
Therefore, a kind of energy is evaluated the pain grade objective and accurately and is detected the system of various origin cause of formation effects in the pain, and very large social need is arranged.Present pain measurement equipment depends on patient's subjectivity statement to a great extent, and a kind of system that can reliably evaluate patient's pain grade and can distinguish the different origins in the pain will bring strong impetus to evaluation and the treatment of pain.
The in recent years research of cerebral function imaging has made people that the understanding of pain central mechanism has been produced revolutionary variation.In fact, as if people have been in by FMRI (Functional Magnetic ResonanceImaging, fMRI) and have obtained the especially edge of the brand-new clinical information of chronic pain of pain.It is a kind of degenerative disease that functional imaging has redefined chronic pain, and has brought hope for some complex diseases such as the research of these diseases of fibromyalgia.Tranquillization attitude functional mri be a kind of to patient and imaging operation personnel require less, repeatable high, can reflect a kind of new functional imaging means of brain spontaneous activity.Because the reaction of brain is the final co-channel of behavior reaction (consciously or be not intended to spontaneous) to pain, the utilization of functional imaging will make people be familiar with pain with a kind of objective method, and understand better potential loop, distinguish the different origins of pain.
Yet, the subjective assessment method of existing evaluation pain easily is subjected to many impacts that adds factor, thereby the impact evaluation is objective and accurate, existing objective evaluation method only relies on single index to judge, the grade that can not accurately reflect objectively pain, and existing assessment method mainly pays close attention to pain impression, can not distinguish each origin cause of formation in the pain.In summary, existing evaluation pain technology obviously exists inconvenience and defective, in actual use so be necessary to be improved.
Summary of the invention
For above-mentioned defective, the object of the present invention is to provide a kind of pain measurement system based on magnetic resonance tranquillization attitude functional imaging, it can evaluate pain grade and/or the pain origin cause of formation of chronic pain patient objective and accurately.
To achieve these goals, the invention provides a kind of pain measurement system based on magnetic resonance tranquillization attitude functional imaging, comprising:
Pain feature samples storehouse, be used for storing different pain grades and the pain origin cause of formation the patient and with the historical tranquillization attitude functional image attribute of its contrast normal person sample;
The functional imaging unit is used for obtaining tranquillization attitude functional imaging data from the new individual patient brain that bears chronic pain, and is transferred to the attributes extraction unit;
The attributes extraction unit is used for from the new individual tranquillization attitude functional image attribute of described tranquillization attitude functional imaging extracting data;
The pain measurement unit is used for the corresponding historical tranquillization attitude functional image attribute of tranquillization attitude functional image attribute and described pain feature samples storehouse of described new individuality is compared, and assesses pain grade and/or the pain origin cause of formation of described new individual patient.
According to pain measurement of the present invention system, described tranquillization attitude functional image attribute comprises: the subnet of nodal community, reflection brain district's modularity or the different component of the function connection attribute of cooperative ability, reflection brain district power of influence/component attribute between the full brain network attribute of reflection brain overall situation integration ability, the local attribute of the local integration ability of reflection brain, each brain district of reflection brain.
According to pain measurement of the present invention system, described pain measurement unit comprises a pain ranking subelement, be used for tranquillization attitude functional image attribute and the historical tranquillization attitude functional image attribute of described new individuality are compared, to utilize must the make new advances fuzzy membership of the historical tranquillization attitude functional image attribute of different pain grade colony in individual tranquillization attitude functional image attribute and the described pain feature samples storehouse of fuzzy cluster analysis with generic attribute, give full brain network attribute, subnet/component attribute, the different weights of the drawn fuzzy membership of function connection attribute are as final degree of membership; In described local attribute and the described pain feature samples storehouse with the similarity of the generic attribute confidence level as this degree of membership; Determine personalized pain added value according to the variation of Different Individual key node attribute; The pain ranking value of described new individual patient=a certain pain membership function+confidence level+personalized pain added value.
According to pain measurement of the present invention system, described pain measurement unit also comprises pain origin cause of formation evaluation subelement, be used for between the new individual subnet of described new individual patient/component attribute self and from described pain feature samples storehouse different pain origin cause of formation colony historical subnet/component attribute compare, by carrying out must the make new advances pain origin cause of formation of individual patient of lateral comparison between new individual subnet/component attribute self; Must the make new advances weight of the various pain origin causes of formation of individual patient of longitudinal comparison between same subnet by new individual subnet/component attribute and different historical genesis samples/component attribute.
According to pain measurement of the present invention system, subnet/the component of the different pain origin causes of formation of reflection is isolated in described attributes extraction unit from full brain network by the mode identification technology with priori, calculate respectively isolated subnet/component attribute, function connection attribute and nodal community;
The described different sub-network of described pain origin cause of formation evaluation subelement lateral comparison self/component attribute is determined the pain origin cause of formation of new individual patient, and longitudinal comparison should represent subnet/component attribute, function connection attribute and the nodal community of different historical genesis samples between the subnet/component attribute of this kind pain origin cause of formation, draw the weight of the various pain origin causes of formation.
According to pain measurement of the present invention system, the described pain origin cause of formation comprises the nociception origin cause of formation, the nerve origin cause of formation and the psychological origin cause of formation.
According to pain measurement of the present invention system, described attributes extraction unit at first is divided into several different brain districts to full brain according to dissecting template, then utilize described dissection template to cut apart the resulting described tranquillization attitude functional imaging data in described functional imaging unit, extract the time series of reflection tranquillization attitude cerebration take described brain district as unit, make up full brain network and subnet net with described brain district and time series.
According to pain measurement of the present invention system, the brain district of described subnet/component comprises: thalamus, corpus amygdaloideum, island leaf, assisted movement cortex, posterior parietal cortex, prefrontal cortex, cingulate cortex, periaqueductal gray, Basal ganglia, cerebellum, primary and secondary sensory cortex.
According to pain measurement of the present invention system, described pain measurement unit also is used for obtaining the Clinical Pain information of new individual patient, analyzes to get pain grade and/or the pain origin cause of formation of the individual patient that makes new advances in conjunction with the tranquillization attitude functional image attribute of described new individuality and the optimization of described Clinical Pain informix; And/or
Described pain measurement unit also is used for finely tuning by the Self-adaptive feedback learning method in described pain feature samples storehouse pain grade and/or the pain origin cause of formation of described new individual patient.
According to pain measurement of the present invention system, described pain feature samples storehouse is used for also that storage is used for making up full brain network, subnet net, function connects and crosses intermediate value, pain grade/origin cause of formation evaluation result, the Clinical Pain information that calculates, the trim values after the study of self adaptation feedback algorithm by described tranquillization attitude functional imaging data communication device during node.
The present invention is based on the objective image data of magnetic resonance tranquillization attitude functional imaging gained, adopt the information processing technology therefrom to extract the tranquillization attitude functional image attribute of new individuality and comparing of the historical tranquillization attitude functional image attribute in the pain feature samples storehouse, preferably by fuzzy clustering, the method such as pattern recognition and Self-adaptive feedback learning is classified to image data eigenvalue and Clinical Pain information, extract, optimize, the comprehensive assessment system that becomes to reflect objective pain grade and the origin cause of formation, and it is objective finally to obtain quantizating index, evaluate exactly pain grade and/or the pain origin cause of formation of chronic pain patient, this pain measurement result can be used for instructing medical personnel that the patient is carried out the science diagnosis, management and treatment.
Description of drawings
Fig. 1 is the structure chart that the present invention is based on the pain measurement system of magnetic resonance tranquillization attitude functional imaging;
Fig. 2 is the flow chart that pain measurement of the present invention system carries out the pain ranking;
Fig. 3 is the flow chart that pain measurement of the present invention system carries out the evaluation of the pain origin cause of formation; And
Fig. 4 is the component identification process figure of pain measurement of the present invention system.
The specific embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, is not intended to limit the present invention.
By fMRI technology studies have shown that human brain: many brain area (being called for short brain district or node) are relevant with Pain Process, and these brain districts comprise the contained brain of pain network district, the contained brain of endogenous pain control loop district, somatic cortex, front and back cingule gyrus, prefrontal cortex, cerveau isole cortex and thalamus.Chronic pain may be relevant with function connection, structure and the biochemical change in these brain districts.Human brain tranquillization attitude functional imaging data and can detect the patient based on the subsequent analysis of dissecting template and whether how standing pain and pain grade; Point out to increase pain grade and the function connection variation that is connected pain, possible structural change and biochemical change; Central nervous system's mechanism that identification is relevant with the pain origin cause of formation with pain and sense pain sensation mechanism.
The invention provides a kind of pain measurement system based on magnetic resonance tranquillization attitude functional imaging, as shown in Figure 1, this pain measurement system 100 comprises functional imaging unit 10, attributes extraction unit 20, pain measurement unit 30 and pain feature samples storehouse 40, wherein:
Pain feature samples storehouse 40, be used for storing different pain grades and the pain origin cause of formation the patient and with the historical tranquillization attitude functional image attribute of its contrast normal person sample.The tranquillization attitude functional image attribute of indication of the present invention comprises: between the local attribute of the full brain network attribute of reflection brain overall situation integration ability, the local integration ability of reflection brain, each brain district of reflection brain the function connection attribute of cooperative ability, reflection brain district power of influence determine grade nodal community, reflect that brain district's modularity or different component determine the subnet of the pain origin cause of formation/component attribute.Described function connection attribute comprises: function connection degree, dependency, mutual information etc.Described nodal community comprises: the degree (degree) in some brains district (node), efficient (Efficiency) and connectedness (Betweenness), the amplitude index of physiology frequency range between the brain district voxel.Described subnet/component attribute comprises: the importance ranking of the origin cause of formation (ranking value), each stability of dividing, the significance level of the origin cause of formation etc.This pain feature samples storehouse 40 also is used for carrying out when pain measurement with reference to comparison and necessary information feedback.Described subnet refers to the combination in several brains district, but its function is indefinite; And described component also is the combination in several brains district, but has clear and definite function.
Preferably, pain measurement unit 30 comprises a pain ranking subelement 31, this pain ranking subelement 31 is used for new individual tranquillization attitude functional image attribute and historical tranquillization attitude functional image attribute are compared, to utilize must the make new advances fuzzy membership of the historical tranquillization attitude functional image attribute of different pain grade colony in individual tranquillization attitude functional image attribute and the pain feature samples storehouse 40 of fuzzy cluster analysis with generic attribute, give respectively full brain network attribute according to priori, subnet/component attribute, heavy (the full brain network attribute 50% for example of the different weights of the drawn fuzzy membership of function connection attribute, subnet/component attribute 30%, function connection attribute 20%), as final degree of membership D (KP); In local attribute and the pain feature samples storehouse 40 with the similarity of the generic attribute confidence level C (C%) as this degree of membership D (KP); Determine personalized pain added value according to the variation of Different Individual key node attribute.Finally, the pain ranking value of new individual patient=a certain pain membership function D (KP)+confidence level (C%)+personalized pain added value.
Studies show that: the different pain origin cause of formation may be corresponding different subnet/components, detects different subnets/component activation degree and may determine the pain origin cause of formation.For the tranquillization attitude functional imaging data of pain patients, adopt component analysis isotype recognition technology to detect different sub-network/component in conjunction with priori and activate degree; Difference between more same individual different sub-network/component can be determined the pain origin cause of formation; And by comparing the difference between respective subnet/component in different sub-network/component and the pain feature samples storehouse 40, can determine the shared proportion of the various pain origin causes of formation.
Accordingly, pain measurement unit 30 also comprises pain origin cause of formation evaluation subelement 32, this pain origin cause of formation evaluation subelement 32 be used for between the new individual subnet of new individual patient/component attribute self and the pain origin cause of formation different from pain feature samples storehouse 40 colony historical subnet/component attribute compare, by carrying out must the make new advances pain origin cause of formation of individual patient of lateral comparison between will be new individual self different sub-network/component attribute; Must the make new advances weight of the various pain origin causes of formation of individual patient of longitudinal comparison by different historical genesis samples between new individual subnet/component attribute and the same historical subnet/component attribute.Subnet/the component of the different pain origin causes of formation of reflection is isolated in attributes extraction unit 20 from full brain network by the mode identification technology with priori, calculate respectively isolated subnet/component attribute, function connection attribute and nodal community; The described different sub-network of pain origin cause of formation evaluation subelement 32 lateral comparisons self/component attribute is determined the pain origin cause of formation of new individual patient, and longitudinal comparison should represent subnet/component attribute, function connection attribute and the nodal community of different historical genesis samples between the subnet/component attribute of this kind pain origin cause of formation, draw the weight of the various pain origin causes of formation.
Consideration and correction that pain measurement of the present invention system 100 also relates to for individual difference in the pain measurement, for the employed individual tranquillization attitude functional imaging data of pain measurement, at first give the benchmark weight of full brain network attribute maximum, reduce as far as possible the pain measurement result error that causes based on certain brain district image feature value that causes because of individual configurations difference; Simultaneously, individual variation is also admitted by pain measurement system 100, with reflecting that the nodal community of individual variation adjusts individual pain added value.
Concrete example, pain origin cause of formation evaluation subelement 32 also are used for obtaining patient's Clinical Pain information, the correction weight of the pain origin cause of formation that goes out the patient in conjunction with benchmark weight and the Clinical Pain information analysis of the described pain origin cause of formation; And pain origin cause of formation evaluation subelement 32 is also finely tuned the fine setting weight of the described pain origin cause of formation, the weight after being optimized by the Self-adaptive feedback learning method in pain feature samples storehouse 40.Correction and fine setting when below also being applicable to pain ranking subelement 31 evaluation pain grade.
Show the flow process that pain measurement system 100 carries out the pain ranking such as Fig. 2: the 10 pairs of chronic pain patient in functional imaging unit are carried out the data acquisition of tranquillization attitude functional imaging, the tranquillization attitude functional imaging data that gather are carried out pretreatment, and attributes extraction unit 20 extracts the full brain network attribute of patients etc. and the full brain network attribute of normal person etc.Choose pain network (and subnet) according to types of pain (being the origin cause of formation), be aided with Clinical Pain scale and other information, calculate pain network and node image feature, determining the pain calibration algorithm according to types of pain, Clinical Pain information, basic feature by algorithms library.Demarcate the pain grade point according to image feature.Pain feature samples storehouse 40 establishment stage pain determining values, pain feature samples storehouse 40 establishment stage evaluation values and expert determination value compare, and both match condition and pain ranking value all deposit pain feature samples storehouse 40 in.In addition, in sample process multi-level artificial neuron learning network correction and the fine setting to pain feature samples storehouse 40, the pain ranking value of finally being measured.
Show the flow process that pain measurement system 100 carries out pain origin cause of formation evaluation such as Fig. 3: at first carry out after the pretreatment obtaining tranquillization attitude functional imaging data, extract the image features such as full brain network attribute, extract respectively nociception component, nerve component, psychological component in conjunction with priori, calculating image feature is worth subnet/component attribute out to compare through after the normalization, be aided with Clinical Pain information, draw the component of the pain origin cause of formation (type) that can represent this individual patient.Consider difference of specific gravity between the different pain origin causes of formation, the difference of more a certain component and pain feature samples storehouse 40 similar component different brackets samples draws the weight of every kind of origin cause of formation at last.Carry out according to the pain origin cause of formation that subnet/component is divided and node calculates, same individuality is carried out lateral comparison between different sub-network/component attribute, carry out longitudinal comparison between same subnet/component attribute for Different Individual; Determine more respectively the leading origin cause of formation (nociception, nerve or psychological) and the different origins proportion of this individuality pain twice.
The instantiation of a pain grade classification is below described:
The pain grade classification is at first determined several classification standard groups, calculates each criterion group fuzzy relation matrix numerical value, then calculates and each criterion group each individual degree of membership relatively by membership function, judges thereby make grade.The border that is divided in of pain grade shows stronger ambiguity, directivity index measure value shows strong range attenuation rule, more stable the closer to the center, its membership function can be expressed as the Gauss distribution form, following (foundation of membership function is determined in Clinical Pain information and the pain ranking value conduct of reserving before):
C
Ij, σ
IjRepresent respectively center and the width of membership function.Connection is connected many index with function for full brain network, subnet/component, and its comprehensive membership function can draw by the linear weighted function method:
U＝∑ai*μ(x) (12)
In the formula, ai is the weight of i item image index.
Can represent the image feature vector U of individual pain feature according to formula (12)
U＝(μ(1)，μ(2)，μ(3)) (13)
Because each origin cause of formation has different stressing among the U, need to give different weights to each origin cause of formation, wherein the weight of full brain network attribute a1 is 50%, the weight of subnet/component attribute a2 is 30%, the weight of function connection attribute a3 is 20%.Characteristic vector U has reflected the degree of membership of tested individual to pain grade samples at different levels.In order to judge the pain grade of individual specimen, characteristic vector U formation fuzzy relationship matrix r and fuzzy subset W are carried out the Fuzzy Compound computing.This paper adopts " " and "+" fuzzy operator, is designated as model M (,+).If the result of compound operation is E, then the element among the E is:
Ab=ab in the formula is Product Operator (algebraic product); A+b=(a+b) ∧ 1 is closed addition operator (algebraical sum); ∑ represents the k number in "+" lower summation.
Finally obtain the degree of membership that this individuality pain characteristic vector belongs to certain pain sample criterion group, equally, carry out the degree of membership of component efficiency attribute and this attribute of pain sample criterion group again and calculate, the degree of membership that draws belongs to the confidence level of certain pain sample criterion group as this individuality.And nodal community is as personalized pain value, when nodal community forward during greater than this criterion group meansigma methods, and this individuality pain grade+1 minute; And when nodal community negative sense during less than this criterion group meansigma methods, this individuality pain grade-1 minute.
Be input as example with two dimension, mode input is respectively " the activation degree of feature brain district K " and the pain grade of VAS evaluation after the obfuscation, and model is output as the pain grade.Pain grade obfuscation class set is [severe pain high, normal, basic zero], and each fuzzy subset's membership function is Gauss distribution, and according to the situation that the brain district activates, its pain grade can be from " zero pain " until change between " having an intense pain ".Clinical and image information adjustment function makes the sensitivity between the pain grade pitch different by study pain feature samples storehouse, and is the highest to its sensitivity between the low pain in zero pain, makes early diagnosis and intervenes more pregnancy.Through the Anti-fuzzy process, can obtain at last the pain grade of a non-zero, this moment, the pain grade mapping was output as: pain grade D (KP)+confidence level (C%)+individual character pain added value.
The instantiation of a pattern recognition is below described:
Suppose to exist in the full brain network attribute n separate component x (t)=[x
1(t), x
2(t) ..., x
n(t)]
T, the component identification of full brain in fact also is categorizing process, purpose is labeled as a new individual D predefined N kind component type C={C exactly
1, C
2.。。C
NIn a kind of, training algorithm is exactly to form in part set in one of the good type of labelling to obtain an effective classification function F:D → C, and new individuality is mapped to type of coding.
We use d and c to represent respectively individual and two variablees of component type, and what the application Bayesian formula calculated individual d is that the probability of c is as follows in component code classification:
The posterior probability of individual d depends on prior probability and class conditional probability, and prior probability and class conditional probability all can obtain in training before.
If in the full brain network attribute N kind component is arranged, then obtain the distribution matrix of a N*P (the brain district number of P for determining according to the dissection masterplate).Choose activation degree (spontaneous activity level) as recognition feature, according to independence assumption in the model-naive Bayesian, each characteristic quantity is independently to the contribution of classification, does not have to each other impact.So when characteristic point was M, formula (21) can be reduced to:
Can obtain the coding classification of probability maximum according to the posterior probability maximization principle:
Each pain origin cause of formation individual amount is roughly the same in training, and the principle of identification is exactly to calculate according to formula (24) the probability size of every kind of component in the full brain network attribute, alternative maximum probability and satisfy the component classification of threshold range.
At first in full brain network attribute, find M the characteristic point that represents the pain general character according to priori in the identifying, (following the example of of M can exert an influence to recognition effect).Certainly, the larger selected characteristic of M value is more, and recognition accuracy is also higher, but also corresponding increase of computation complexity.In actual identifying, M comes value according to priori (the brain district number of pain network) and empirical equation, (from 10 to 30 incremental variations are observed the M value to the impact of recognition result stability, employed M value scope when selected stability is the highest).
Can obtaining by the random function that structure is dissected in the template brain district number at random of M characteristic point, construct relatively difficulty of a good random function, the more difficult true random of accomplishing when especially being confined to dissect the masterplate number, so the recognition strategy that M is ordered in the actual moving process is to choose non-conterminous M point as characteristic point in full brain scope and subnet/component, such M point is dispersed in each position, all the probability of the same feature of representative is very little, and because the distribution probability of different characteristic in the various pain origin causes of formation is different, therefore have the difference degree for different pain components.
Fig. 4 shows the flow process of identifying whole process from training to: choose M characteristic point from individuality to be identified; The characteristic frequency table of concentrating according to feature database component training sample calculates the different component product of probability; Select component of candidate; According to each component threshold values that feature database component training sample set provides, judge that component whether in the threshold values scope, if it is exports component, otherwise be considered to unknown component.
By repeatedly component identification, finally obtain the importance ranking of (1) origin cause of formation: utilize the dependency repeatedly move between the same origin cause of formation that component analysis obtains, try to achieve the variance of every kind of origin cause of formation with the stability indicator of tolerance different origins, utilize this index and the distribution on Different Individual, calculate the general stability index of every kind of origin cause of formation, draw the ordering curve chart of all origin causes of formation on Different Individual and overall ordering curve chart; The stability of (2) at every turn dividing: take turns component identification more and obtain a plurality of origin causes of formation, utilize the correlation coefficient between the origin cause of formation and carry out Multidimensional Scaling also to show the similarity that repeatedly moves between the same origin cause of formation that obtains; (3) significance level of the origin cause of formation: obtain all origin causes of formation by cartogram (mean chart, T check figure, standard deviation figure) or individual cause of formation distribution figure, calculate the significant indexes in the population sample composition of certain individual origin cause of formation after repeatedly dividing.
The present invention has carried out two researchs, first research is about judging whether the tranquillization attitude can identify the pain grade, research contents comprises that the cerebral cortex fMRI to suffering from chronic pain patient and pairing healthy volunteer after the amputation compares, and finds that there is the cortex hormone function recombination zone in the phantom pain patient; The phantom pain patient further again by occuring and phantom pain brain in patients cortex fMRI not occuring relatively in us after the amputation, the generation of finding cerebral cortex function Reorganization and phantom pain is closely related, thereby confirms: the genesis of cerebral cortex function Reorganization and phantom pain has substantial connection.This studies and previous research conclusion: degree relevant (Flor, etal., 1995 of the restructuring of the grade of chronic pain and function of cortex after the amputation; Saadahet al., 1994) result is consistent, and result of study is to have set up to utilize functional image to evaluate the pattern of different pain grades.In conjunction with former studies, can draw: the generation of chronic pain and brain in patients cortex hormone function restructuring relation are close after the amputation, the degree of cerebral cortex function restructuring may with phantom limb pain grade positive correlation.An other research is for morphology algoscopy (the Voxel-based morphometry of high resolution structures picture use based on voxel, VBM) find: grey matter volume relative normal person in the outside has significantly and reduces behind the thalamus in patients with amputation amputation offside brain district, time correlation after the degree that the grey matter volume reduces and the amputation, but there is no obviously relevant with the occurrence frequency of phantom pain with degree.In addition, research is proof also: although phantom pain and thalamus grey matter change in volume without relevant, phantom pain is positively correlated with the volume minimizing in brain pain management zone.
Have the function MR investigation to show: the different pain origin cause of formation may be corresponding different function subnets, detects different function subnet state of activation (specifically threshold value) and may determine the shared proportion of this kind origin cause of formation in the origin cause of formation of pain or the pain.These researchs are about judging that whether the imaging of magnetic resonance tranquillization attitude can be used for identifying the different pain origin causes of formation, and research contents comprises patient's pain loop/network (Painmatrix), endogenous pain control loop, anterior cingutate, posterior cingutate, prefrontal cortex, front colour band cortex and the thalamus of patient's phantom pain after the amputation.Result of study is to have set up the pattern that can distinguish the different pain origin causes of formation.
Second research is about judging each image index with fuzzy comprehensive evaluation method to the image degree of disease, three image index: 1-Cost, 2-Eglobal as shown in the table, 3-Elocal.
Divide group image 1-Cost image 2-Eglobal image 3-Elocal
Organize 1 329.5359 158.1113 264.7445
Organize 1 326.4346 158.3553 265.6859
Organize 1 326.5555 158.3569 265.5657
Organize 2 322.1001 158.4251 266.0616
Organize 2 326.9431 158.4602 265.5657
Organize 2 326.4387 158.3185 265.4383
Organize 3 330.8502 158.1233 264.1331
Organize 3 331.2514 158.1714 264.5345
Organize 3 327.2779 158.2419 265.4953
2947.3874 1424.5639 2387.2246
(2) calculate every group mean Ridit value
Result of calculation,
Be worth less, larger to the influence degree of disease; Otherwise,
Value is larger, and is less on the impact of disease.Research conclusion:
Illustrate that image index 3-Elocal is larger to the influence degree of disease.
In addition, there is function MR investigation (Chicago Northwest University Feenberg medical college) to carry out the comparative study discovery by the patient that the patient that experiences the lower back portion chronic pain and one group of pain are controlled fully: not have a group of pain, the brain position presents poised state: when enlivened in a zone, other zones just quieted down.In the patient of experience chronic pain, mainly follow the relevant cerebral cortex proparea of emotion but " forever peace and quiet are not got off ".This zone keeps the state that highly enlivens always, neurocyte is strained, and changed the connection line between the neurocyte.This studies show that low back pain patient emotion and psychological factor may account for mastery reaction, and the poised state of whole brain is destroyed.
In sum, the present invention is based on the objective image data of magnetic resonance tranquillization attitude functional imaging gained, adopt the information processing technology therefrom to extract the tranquillization attitude functional image attribute of new individuality and comparing of the historical tranquillization attitude functional image attribute in the pain feature samples storehouse, preferably by fuzzy clustering, the method such as pattern recognition and Self-adaptive feedback learning is classified to image data eigenvalue and Clinical Pain information, extract, optimize, the comprehensive assessment system that becomes to reflect objective pain grade and the origin cause of formation, and it is objective finally to obtain quantizating index, evaluate exactly pain grade and/or the pain origin cause of formation of chronic pain patient, this pain measurement result can be used for instructing medical personnel that the patient is carried out the science diagnosis, management and treatment.The present invention compared with the prior art has the noinvasive of pain information, dynamic and instantaneous collection, by the sample size characteristic optimization, can accurate, objective, stably evaluate pain grade and the origin cause of formation situation of patient within a certain period.
Certainly; the present invention also can have other various embodiments; in the situation that do not deviate from spirit of the present invention and essence thereof; those of ordinary skill in the art work as can make according to the present invention various corresponding changes and distortion, but these corresponding changes and distortion all should belong to the protection domain of the appended claim of the present invention.
Claims (9)
1. the pain measurement system based on magnetic resonance tranquillization attitude functional imaging is characterized in that, comprising:
Pain feature samples storehouse, be used for storing different pain grades and the pain origin cause of formation the patient and with the historical tranquillization attitude functional image attribute of its contrast normal person sample;
The functional imaging unit is used for obtaining tranquillization attitude functional imaging data from the new individual patient brain that bears chronic pain, and is transferred to the attributes extraction unit;
The attributes extraction unit is used for from the new individual tranquillization attitude functional image attribute of described tranquillization attitude functional imaging extracting data;
The pain measurement unit is used for the corresponding historical tranquillization attitude functional image attribute of tranquillization attitude functional image attribute and described pain feature samples storehouse of described new individuality is compared, and assesses pain grade and/or the pain origin cause of formation of described new individual patient;
Described tranquillization attitude functional image attribute comprises: the function connection attribute of cooperative ability, reflection brain district power of influence are determined the nodal community of grade or are reflected that brain district's modularity or different component determine the subnet of the pain origin cause of formation/component attribute between the local attribute of the full brain network attribute of reflection brain overall situation integration ability, the local integration ability of reflection brain, each brain district of reflection brain.
2. pain measurement according to claim 1 system, it is characterized in that, described pain measurement unit comprises a pain ranking subelement, be used for tranquillization attitude functional image attribute and the historical tranquillization attitude functional image attribute of described new individuality are compared, to utilize must the make new advances fuzzy membership of the historical tranquillization attitude functional image attribute of different pain grade colony in individual tranquillization attitude functional image attribute and the described pain feature samples storehouse of fuzzy cluster analysis with generic attribute, give full brain network attribute, subnet/component attribute, the different weights of the drawn fuzzy membership of function connection attribute are as final degree of membership; In described local attribute and the described pain feature samples storehouse with the similarity of the generic attribute confidence level as this degree of membership; Determine personalized pain added value according to the variation of Different Individual key node attribute; The pain ranking value of described new individual patient=a certain pain membership function+confidence level+personalized pain added value.
3. pain measurement according to claim 1 system, it is characterized in that, described pain measurement unit also comprises pain origin cause of formation evaluation subelement, be used for between the new individual subnet of described new individual patient/component attribute self and from described pain feature samples storehouse different pain origin cause of formation colony historical subnet/component attribute compare, by carrying out must the make new advances pain origin cause of formation of individual patient of lateral comparison between new individual subnet/component attribute self; Must the make new advances weight of the various pain origin causes of formation of individual patient of longitudinal comparison between same subnet by new individual subnet/component attribute and different historical genesis samples/component attribute.
4. pain measurement according to claim 3 system, it is characterized in that, subnet/the component of the different pain origin causes of formation of reflection is isolated in described attributes extraction unit from full brain network by the mode identification technology with priori, calculate respectively isolated subnet/component attribute, function connection attribute and nodal community;
Described pain origin cause of formation evaluation subelement lateral comparison new individual patient self different sub-network/component attribute is determined the pain origin cause of formation of new individual patient, and longitudinal comparison represents the subnet of different historical genesis samples between the subnet of this kind pain origin cause of formation/component attribute/component attribute, function connection attribute and nodal community, draws the weight of the various pain origin causes of formation.
5. pain measurement according to claim 3 system is characterized in that, the described pain origin cause of formation comprises the nociception origin cause of formation, the nerve origin cause of formation and the psychological origin cause of formation.
6. pain measurement according to claim 1 system, it is characterized in that, described attributes extraction unit at first is divided into several different brain districts to full brain according to dissecting template, then utilize described dissection template to cut apart the resulting described tranquillization attitude functional imaging data in described functional imaging unit, extract the time series of reflection tranquillization attitude cerebration take described brain district as unit, make up full brain network and subnet net with described brain district and time series.
7. pain measurement according to claim 1 system, it is characterized in that, described brain district comprises: thalamus, corpus amygdaloideum, island leaf, assisted movement cortex, posterior parietal cortex, prefrontal cortex, cingulate cortex, periaqueductal gray, Basal ganglia, cerebellum, primary and secondary sensory cortex.
8. each described pain measurement system according to claim 1～7, it is characterized in that, described pain measurement unit also is used for obtaining the Clinical Pain information of new individual patient, analyzes to get pain grade and/or the pain origin cause of formation of the individual patient that makes new advances in conjunction with the tranquillization attitude functional image attribute of described new individuality and the optimization of described Clinical Pain informix; And/or
Described pain measurement unit also is used for finely tuning by the Self-adaptive feedback learning method in described pain feature samples storehouse pain grade and/or the pain origin cause of formation of described new individual patient.
9. pain measurement according to claim 8 system, it is characterized in that, described pain feature samples storehouse is used for also that storage is used for making up full brain network, subnet net, function connects and crosses intermediate value, pain grade/origin cause of formation evaluation result, the Clinical Pain information that calculates, the trim values after the study of self adaptation feedback algorithm by described tranquillization attitude functional imaging data communication device during node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010189516 CN102266223B (en) | 2010-06-01 | 2010-06-01 | Pain evaluation system based on magnetic resonance resting state functional imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010189516 CN102266223B (en) | 2010-06-01 | 2010-06-01 | Pain evaluation system based on magnetic resonance resting state functional imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102266223A CN102266223A (en) | 2011-12-07 |
| CN102266223B true CN102266223B (en) | 2013-01-30 |
Family
ID=45048823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010189516 Expired - Fee Related CN102266223B (en) | 2010-06-01 | 2010-06-01 | Pain evaluation system based on magnetic resonance resting state functional imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102266223B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016124482A1 (en) * | 2015-02-02 | 2016-08-11 | Koninklijke Philips N.V. | Pain management wearable device |
| CN104715150A (en) * | 2015-03-19 | 2015-06-17 | 上海海事大学 | Migraineur cerebral cortex assistant classification analyzing method based on complex network |
| CN104921727B (en) * | 2015-06-24 | 2017-11-07 | 上海海事大学 | The brain function detection of connectivity system and method instructed based on adaptive prior information |
| CN106473758A (en) * | 2015-08-24 | 2017-03-08 | 上海联影医疗科技有限公司 | Breast imaging equipment and its control method |
| CN106251379B (en) * | 2016-07-25 | 2017-11-07 | 太原理工大学 | A kind of brain structural network connection optimization method based on random sectional pattern |
| CN107610467B (en) * | 2017-09-26 | 2019-12-20 | 青岛海信网络科技股份有限公司 | Traffic state discrimination method and device |
| CN109192304B (en) * | 2018-08-31 | 2021-07-27 | 重庆高铂瑞骐科技开发有限公司 | Multi-mode information fusion system for esophagus functional disease diagnosis system |
| CN110652307B (en) * | 2019-09-11 | 2021-01-12 | 中国科学院自动化研究所 | Functional nuclear magnetic image-based striatum function detection system for schizophrenia patient |
| CN114266321A (en) * | 2021-12-31 | 2022-04-01 | 广东泰迪智能科技股份有限公司 | Weak supervision fuzzy clustering algorithm based on unconstrained prior information mode |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101677775A (en) * | 2007-04-05 | 2010-03-24 | 纽约大学 | System and method for pain detection and computation of a pain quantification index |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7462155B2 (en) * | 2004-10-27 | 2008-12-09 | England Robert L | Objective determination of chronic pain in patients |
| WO2006071891A2 (en) * | 2004-12-23 | 2006-07-06 | The General Hospital Corporation | Evaluating central nervous system |
-
2010
- 2010-06-01 CN CN 201010189516 patent/CN102266223B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101677775A (en) * | 2007-04-05 | 2010-03-24 | 纽约大学 | System and method for pain detection and computation of a pain quantification index |
Non-Patent Citations (2)
| Title |
|---|
| 杨汉丰等.fMRI在疼痛研究中的运用.《川北医学院学报》.2008,第23卷(第2期),189-192. * |
| 谈高等.fMRI在疼痛、镇痛和镇痛药研究领域中的新进展.《解剖学研究》.2007,第29卷(第2期),141-144. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102266223A (en) | 2011-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102266223B (en) | Pain evaluation system based on magnetic resonance resting state functional imaging | |
| CN102293656B (en) | Emotional stability evaluation system based on magnetic resonance imaging and evaluation method thereof | |
| Parker et al. | Classifying depression by mental state signs | |
| Noonan et al. | Measuring fatigue in persons with multiple sclerosis: creating a crosswalk between the Modified Fatigue Impact Scale and the PROMIS Fatigue Short Form | |
| Ehrenbrusthoff et al. | A systematic review and meta-analysis of the reliability and validity of sensorimotor measurement instruments in people with chronic low back pain | |
| CN102178514A (en) | Coma degree evaluating method based on multiple indexes of non-linearity and complexity | |
| Bordescu et al. | Fractal analysis of Neuroimagistic. Lacunarity degree, a precious indicator in the detection of Alzheimer’s disease | |
| Sutker et al. | MMPI response patterns and alcohol consumption in DUI offenders. | |
| Engin et al. | The classification of human tremor signals using artificial neural network | |
| Hart et al. | Computerized adaptive test for patients with foot or ankle impairments produced valid and responsive measures of function | |
| CN106096636A (en) | A kind of Advancement Type mild cognition impairment recognition methods based on neuroimaging | |
| Sugavaneswaran et al. | Ambiguity domain-based identification of altered gait pattern in ALS disorder | |
| Desai et al. | Decision support system for arrhythmia beats using ECG signals with DCT, DWT and EMD methods: A comparative study | |
| CN106355033A (en) | Life risk assessment system | |
| Rasheed et al. | Anomaly detection of moderate traumatic brain injury using auto-regularized multi-instance one-class SVM | |
| Lowe et al. | Greater precision when measuring dementia severity: establishing item parameters for the Clinical Dementia Rating Scale | |
| Hussein et al. | Sex estimation of femur using simulated metapopulation database: A preliminary investigation | |
| Yang et al. | Research on rehabilitation effect prediction for patients with SCI based on machine learning | |
| Baum et al. | Comparing PROMIS computer-adaptive tests to the Brief Symptom Inventory in patients with prostate cancer | |
| de Jong et al. | A more meaningful parameterization of the Lee–Carter model | |
| Tabassum et al. | Gender classification from anthropometric measurement by boosting decision tree: A novel machine learning approach | |
| CN110427367B (en) | Damage assessment method, device and equipment based on residue assessment parameter and storage medium | |
| Izquierdo et al. | Robust prediction of cognitive test scores in Alzheimer's patients | |
| Chen et al. | Alcoholism detection by wavelet energy entropy and linear regression classifier | |
| Khoma et al. | Artificial neural network capability for human being identification based on ECG |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130130 Termination date: 20170601 |