CN107993714B - Human brain consciousness state predicting device - Google Patents
Human brain consciousness state predicting device Download PDFInfo
- Publication number
- CN107993714B CN107993714B CN201711002297.5A CN201711002297A CN107993714B CN 107993714 B CN107993714 B CN 107993714B CN 201711002297 A CN201711002297 A CN 201711002297A CN 107993714 B CN107993714 B CN 107993714B
- Authority
- CN
- China
- Prior art keywords
- brain
- subunit
- magnetic resonance
- imaging
- preset
- 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.)
- Active
Links
- 210000004556 brain Anatomy 0.000 title claims abstract description 164
- 238000003384 imaging method Methods 0.000 claims abstract description 83
- 239000013598 vector Substances 0.000 claims abstract description 43
- 238000012216 screening Methods 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 22
- 238000010238 partial least squares regression Methods 0.000 claims abstract description 6
- 230000003925 brain function Effects 0.000 claims description 61
- 230000006870 function Effects 0.000 claims description 58
- 230000000284 resting effect Effects 0.000 claims description 49
- 238000004393 prognosis Methods 0.000 claims description 38
- 206010010071 Coma Diseases 0.000 claims description 26
- 238000007781 pre-processing Methods 0.000 claims description 24
- 238000005070 sampling Methods 0.000 claims description 21
- 238000001914 filtration Methods 0.000 claims description 19
- 230000036961 partial effect Effects 0.000 claims description 9
- 230000002360 prefrontal effect Effects 0.000 claims description 9
- 210000003926 auditory cortex Anatomy 0.000 claims description 6
- 210000000977 primary visual cortex Anatomy 0.000 claims description 6
- 210000000857 visual cortex Anatomy 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000012952 Resampling Methods 0.000 claims description 5
- 210000001152 parietal lobe Anatomy 0.000 claims description 4
- 210000000976 primary motor cortex Anatomy 0.000 claims description 4
- 206010021143 Hypoxia Diseases 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000002860 competitive effect Effects 0.000 claims description 3
- 210000001652 frontal lobe Anatomy 0.000 claims description 3
- 210000004326 gyrus cinguli Anatomy 0.000 claims description 3
- 230000001146 hypoxic effect Effects 0.000 claims description 3
- 230000003238 somatosensory effect Effects 0.000 claims description 3
- 230000000472 traumatic effect Effects 0.000 claims description 3
- 108010001267 Protein Subunits Proteins 0.000 claims 4
- 230000003750 conditioning effect Effects 0.000 claims 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000002059 diagnostic imaging Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 11
- 238000000034 method Methods 0.000 description 11
- 238000011084 recovery Methods 0.000 description 6
- 230000001936 parietal effect Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 208000035475 disorder Diseases 0.000 description 3
- 238000012800 visualization Methods 0.000 description 3
- 238000001276 Kolmogorov–Smirnov test Methods 0.000 description 2
- 206010024264 Lethargy Diseases 0.000 description 2
- 230000000763 evoking effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 208000037273 Pathologic Processes Diseases 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000006735 deficit Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000005714 functional activity Effects 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 230000009054 pathological process Effects 0.000 description 1
- -1 r (feature Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention relates to the technical field of medical imaging, in particular to a device for predicting the consciousness state of human brain, aiming at solving the technical problem of accurately predicting the consciousness state of human brain. For this purpose, the device for predicting the consciousness state of the human brain comprises a consciousness state prediction model which can predict the consciousness state of the human brain according to the imaging characteristics of the human brain and corresponding clinical information. Specifically, the imaging characteristic calculation module in the consciousness state prediction model can calculate imaging characteristics representing the brain connection state of the patient, the imaging characteristic screening module can screen imaging characteristics with high prognostic relevance, the imaging characteristic processing module can combine the imaging characteristics and clinical information into a new characteristic vector, and the model training module can train the partial least squares regression model according to the characteristic vector to obtain a final prediction model. The prediction device in the invention gives consideration to clinical information and image detection information, thereby comprehensively and effectively predicting the brain consciousness state.
Description
Technical Field
The invention relates to the technical field of medical images, in particular to a human brain consciousness state prediction device.
Background
Disturbance of consciousness refers to disturbance in the ability of a person to recognize and perceive the surrounding environment and his own state, which is mainly caused by impaired functional activity of the higher nerve center, and may be manifested as lethargy, confusion, lethargy and coma.
Currently, the brain consciousness state prediction of Coma may be based on Glasgow Coma Scale (GCS), behavioral characteristic change, electroencephalogram, evoked potential, brain CT or MRI, and brain and general physiological changes.
Specifically, the brain consciousness state of a Coma patient can be objectively predicted by using a Glasgow Coma Scale (GCS), but the method depends on clinical observation of the patient by a doctor and has a high misjudgment rate. Meanwhile, disturbance of consciousness is usually caused by various injuries, and different pathological processes and severity degrees exist, so that the brain consciousness of a coma patient is predicted by any one of a prediction method based on behavior characteristic change, a prediction method based on electroencephalogram, a prediction method based on evoked potential, a prediction method based on brain CT image or brain MRI image and a prediction method based on brain and general physiological change, and high false positive or false negative is generated.
Disclosure of Invention
In order to solve the above problems in the prior art, i.e. to determine how to accurately predict the consciousness state of the human brain, especially the consciousness state of a coma patient, the invention provides a human brain consciousness state prediction device.
The device for predicting the consciousness state of the human brain comprises a consciousness state prediction model, a brain state prediction model and a brain state prediction model, wherein the consciousness state prediction model is configured to predict the consciousness state of the human brain according to the imaging characteristics of the human brain and corresponding clinical information; the consciousness state prediction model comprises an iconography characteristic calculation module, an iconography characteristic screening module, an iconography characteristic processing module and a model training module;
the imaging characteristic calculation module is configured to calculate imaging characteristics according to the resting function magnetic resonance images in a preset training set; the imaging characteristics comprise a first characteristic and a second characteristic, the first characteristic is the spatial similarity between the whole brain function connection and the standard brain function connection of the patient, and the second characteristic is the function connection of a preset brain region seed point in the resting function magnetic resonance image of the patient;
the imaging characteristic screening module is configured to screen the imaging characteristics according to the consciousness disturbance prognostic indicators;
the imaging characteristic processing module is configured to combine the imaging characteristics obtained by screening by the imaging characteristic screening module and the clinical information in the preset training set into a characteristic vector;
and the model training module is configured to train a preset partial least square regression model according to the feature vector to obtain a final prediction model.
Further, a preferred technical solution provided by the present invention is:
the imaging characteristic calculation module comprises a first imaging characteristic calculation unit and a second imaging characteristic calculation unit;
the first imaging characteristic calculation unit is configured to calculate first pearson correlation coefficients of the patient whole brain function connection and the standard brain function connection, and take each first pearson correlation coefficient corresponding to each preset brain region seed point in the patient resting function magnetic resonance image as a first characteristic;
the second imaging characteristic calculating unit is configured to calculate second pearson correlation coefficients of any two preset brain region seed points according to an average functional magnetic resonance signal of each preset brain region seed point in the patient resting functional magnetic resonance image, and use each second pearson correlation coefficient corresponding to each preset brain region seed point as a second characteristic.
Further, a preferred technical solution provided by the present invention is:
the imaging characteristic screening module comprises a first screening unit and a second screening unit;
the first screening unit is configured to perform the following operations:
calculating a third Pearson correlation coefficient between the imaging characteristic and the consciousness disturbance prognostic index:
selecting the imaging characteristics of which the third Pearson correlation coefficient is greater than a preset first threshold;
the second screening unit is configured to adopt a competitive adaptive re-weighting sampling algorithm and a partial least square algorithm, and take the image characteristics with the characteristic weight larger than a preset second threshold value in the image characteristics selected and obtained by the first screening unit as final image characteristics.
Further, a preferred technical solution provided by the present invention is:
the preset partial least squares regression model is shown as follows:
Y＝XB+B0
wherein, X is a feature vector containing imaging features and clinical information, and Y is an consciousness disturbance prognostic indicator; b and B0Respectively a feature weight vector to be estimated and a bias vector to be estimated.
Further, a preferred technical solution provided by the present invention is:
the consciousness disturbance prognosis index is a corresponding Glassy coma scale score of the patient in the preset training set;
the clinical information includes the age, etiology, and course of the patient; the causes include traumatic brain disturbance, stroke brain disturbance and hypoxic brain disturbance.
Further, a preferred technical solution provided by the present invention is:
the imaging characteristic calculation module further comprises a first image processing unit; the first image processing unit comprises a first preprocessing subunit and a patient brain function connection calculating subunit;
the first preprocessing subunit is configured to preprocess the patient resting function magnetic resonance image in the preset training set;
the patient brain function connection calculation unit is configured to calculate the whole brain function connection of the preset brain region seed points according to the patient resting function magnetic resonance image preprocessed by the first preprocessing subunit, and further count the whole brain function connection of all the preset brain region seed points to obtain the patient brain function connection.
Further, a preferred technical solution provided by the present invention is:
the first preprocessing subunit comprises a first sampling subunit, a first adjusting subunit, a first denoising subunit and a first filtering subunit;
the first sampling subunit is configured to resample the patient resting function magnetic resonance image in the preset training set;
the first adjusting subunit is configured to move and/or rotate the patient rest function magnetic resonance images obtained by resampling through the first sampling subunit so as to eliminate position deviation among different patient rest function magnetic resonance images;
the first denoising subunit is configured to denoise the patient resting function magnetic resonance image adjusted by the first adjusting subunit;
the first filtering subunit is configured to perform filtering processing on the patient rest function magnetic resonance image denoised by the first denoising subunit.
Further, a preferred technical solution provided by the present invention is:
the imaging characteristic calculation module further comprises a second image processing unit; the second image processing unit comprises a second preprocessing subunit and a standard brain function connection calculating subunit;
the second preprocessing subunit is configured to preprocess the normal person resting function magnetic resonance image in the preset training set;
and the standard brain function connection counting subunit is configured to calculate the whole brain function connection of each preset brain region seed point according to the normal person resting function magnetic resonance image preprocessed by the second preprocessing subunit, and further count the whole brain function connection of all the preset brain region seed points to obtain the standard brain function connection.
Further, a preferred technical solution provided by the present invention is:
the second preprocessing subunit comprises a second sampling subunit, a second adjusting subunit, a third adjusting subunit, a second denoising subunit and a second filtering subunit;
the second sampling subunit is configured to resample the magnetic resonance image of the rest function of the normal person in the preset training set;
the second adjusting subunit is configured to move and/or rotate the normal person resting function magnetic resonance images obtained by resampling through the second sampling subunit so as to eliminate the position deviation between different normal person resting function magnetic resonance images;
the third adjusting subunit is configured to register the normal resting function magnetic resonance image of the person adjusted by the second adjusting subunit to an MNI space;
the second denoising subunit is configured to denoise the normal person resting function magnetic resonance image registered by the third adjusting subunit;
the second filtering subunit is configured to perform filtering processing on the normal person resting function magnetic resonance image denoised by the second denoising subunit.
Further, a preferred technical solution provided by the present invention is:
the preset brain region seed points comprise a first type brain region seed point, a second type brain region seed point, a third type brain region seed point, a fourth type brain region seed point, a fifth type brain region seed point and a sixth type brain region seed point;
the first type brain region seed points include medial frontal lobe, posterior cingulate gyrus and cuneiform lobe, left parietal lobe lateral cortex and right parietal lobe lateral cortex within a default network;
the second type of brain region seed point comprises an dorsolateral prefrontal lobe, a left prefrontal lobe, a right prefrontal lobe, a left apical cortex, and a right apical cortex within an executive control network;
the third type brain region seed points comprise a left front brain island, a right front brain island and a front back in front of the button in the highlight network;
the fourth type brain region seed point comprises a left primary motor cortex, a right primary motor cortex and an auxiliary motor region in the somatosensory motor network;
the fifth type brain region seed point comprises a left primary auditory cortex, a right primary auditory cortex and a cingulum retrocenter in an auditory network;
the sixth type of brain region seed point includes a left primary visual cortex, a right primary visual cortex, a left contact visual cortex, and a right contact visual cortex within the visual network.
Compared with the closest prior art, the technical scheme at least has the following beneficial effects:
1. the imaging characteristic calculation module can calculate the brain network function connection according to the resting magnetic resonance image, so as to obtain the spatial similarity between the patient whole brain function connection and the standard brain function connection and the function connection of the preset brain region seed point in the resting magnetic resonance image of the patient.
2. The imaging characteristic screening module can screen the imaging characteristics according to the consciousness disturbance prognosis indexes to obtain the imaging characteristics with higher correlation with prognosis analysis.
3. The imaging characteristic processing module can fuse imaging characteristics and clinical information together to serve as a characteristic vector, so that the brain consciousness state prediction result has clinical information and image detection information, and the accuracy of the brain consciousness state prediction result is further improved.
4. The model training module can train the partial least square regression model according to the feature vectors and optimize each weight value in the partial least square regression model.
Drawings
FIG. 1 is a schematic structural diagram of a device for predicting a state of consciousness of a human brain according to an embodiment of the present invention;
FIG. 2 is a T-value plot of a default network in a brain function network;
FIG. 3 is a T-value diagram of an executive control network in a brain function network;
FIG. 4 is a graph of T values of a highlight network in a brain function network;
FIG. 5 is a T-value plot of a motor network in a brain function network;
FIG. 6 is a graph of T values for auditory networks in a brain functional network;
FIG. 7 is a graph of T values for a visual network in a brain function network;
FIG. 8 is a diagram illustrating the predicted result of the brain consciousness state of a patient with disturbance of consciousness in the pre-set training set according to an embodiment of the present invention;
FIG. 9 is a graphical representation of the correlation of the predicted outcome of the brain consciousness state with the prognostic score of FIG. 8;
FIG. 10 is a graphical representation of the correspondence between the predicted brain consciousness state and the prognostic score of FIG. 8;
FIG. 11 is a graphical representation of the number of persons having actual prognosis for regaining consciousness for the patient for the different predicted outcomes of brain consciousness in FIG. 8;
FIG. 12 is a graph showing the rate of recovery of consciousness in actual prognosis of a patient for the results of prediction of different states of brain consciousness shown in FIG. 8;
FIG. 13 is a schematic diagram of ROC curves for the brain consciousness state prediction result shown in FIG. 8;
FIG. 14 is a diagram illustrating a predicted result of a state of brain consciousness of a patient with central disturbance of consciousness in accordance with an embodiment of the present invention;
FIG. 15 is a graphical representation of the correlation of the brain consciousness state prediction result of FIG. 14 with a prognostic score;
FIG. 16 is a graphical representation of the correspondence between the predicted brain consciousness state and the prognostic score of FIG. 14.
Detailed Description
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and are not intended to limit the scope of the present invention.
The following describes a device for predicting a state of consciousness of a human brain according to an embodiment of the present invention with reference to the drawings.
Referring to fig. 1, fig. 1 schematically shows the structure of a human brain consciousness state predicting apparatus according to the present embodiment. As shown in fig. 1, the device for predicting the consciousness state of the human brain in the present embodiment includes a consciousness state prediction model, which may be configured to predict the consciousness state of the human brain according to the imaging characteristics of the human brain and the corresponding clinical information. The clinical information may include, among other things, the age, etiology, and course of the patient, and the etiology may include traumatic brain awareness disorder, stroke brain awareness disorder, and hypoxic brain awareness disorder.
Specifically, the awareness state prediction model in this embodiment may include a visualization feature calculation module 11, a visualization feature screening module 12, a visualization feature processing module 13, and a model training module 14.
In this embodiment, the imaging characteristic calculating module 11 may be configured to calculate the imaging characteristic according to the resting function magnetic resonance image in the preset training set. The imaging characteristics comprise a first characteristic and a second characteristic, the first characteristic is the spatial similarity between the whole brain function connection and the standard brain function connection of the patient, and the second characteristic is the function connection of preset brain region seed points in the resting function magnetic resonance image of the patient. The imaging characteristic screening module 12 may be configured to screen the imaging characteristics according to the consciousness impairment prognosis index. The imaging feature processing module 13 may be configured to merge the imaging features filtered by the imaging feature filtering module 12 and the clinical information in the preset training set into a feature vector. The model training module 14 may be configured to train a preset partial least squares regression model according to the feature vectors to obtain a final prediction model.
In a preferred embodiment of this embodiment, the preset brain region seed points may include a first type brain region seed point, a second type brain region seed point, a third type brain region seed point, a fourth type brain region seed point, a fifth type brain region seed point, and a sixth type brain region seed point. The first type brain region seed point may include an inner frontal lobe (anmain Anterior core, aMPFC), Posterior cingulate Posterior and Anterior wedge lobe (PCC), Left lateral parietal core (l.latp), and Right lateral parietal core (r.latp) within a Default network (Default mode). The second type of brain region seed point may include a dorsolateral prefrontal leaf (DMPFC), a Left prefrontal leaf (l.pfc), a Right prefrontal leaf (r.pfc), a Left superior partial cortex (l.parietal), and a Right superior partial cortex (r.parietal) within an executive control network (executive control). The third type brain region seed points may include a Left anterior brain island (l.ains), a Right anterior brain island (r.ains), and a anterior Left anterior brain island (dACC) within a highlight network (Salience). The fourth type of brain region seed point may include a Left primary motor core (l.m 1), a Right primary motor core (r.m 1), and an auxiliary motor region (SMA) within a somatosensory motor network (sensor). The fifth type brain region seed point may include a Left primary Auditory cortex (l.a 1), a Right primary Auditory cortex (r.a 1), and a Middle Cingulate Cortex (MCC) within an Auditory network (audioy). The sixth type of brain region seed point may include a Left primary Visual cortex (l.v 1), a Right primary Visual cortex (r.v 1), a Left associated Visual cortex (l.v 4), and a Right associated Visual cortex (r.v 4) within a Visual network (Visual).
Further, the imaging characteristic calculating module 11 in this embodiment may include a first image processing unit, a second image processing unit, a first imaging characteristic calculating unit, and a second imaging characteristic calculating unit.
The first image processing unit in this embodiment may include a first preprocessing subunit and a patient brain function connection calculating subunit. The first preprocessing subunit may be configured to preprocess the magnetic resonance image of the rest function of the patient in the preset training set. The patient brain function connection calculation unit can be configured to calculate the whole brain function connection of each preset brain region seed point according to the patient resting function magnetic resonance image preprocessed by the first preprocessing subunit, and further count the whole brain function connection of all preset brain region seed points to obtain the patient brain function connection.
In a preferred embodiment of this embodiment, the first preprocessing subunit may include a first sampling subunit, a first adjusting subunit, a first denoising subunit, and a first filtering subunit. The first sampling subunit may be configured to resample the resting function magnetic resonance image of the patient in the preset training set. The first adjustment subunit may be configured to move and/or rotate the patient rest function magnetic resonance images resampled by the first sampling subunit to eliminate a position deviation between different patient rest function magnetic resonance images. The first denoising subunit can be configured to denoise the patient resting function magnetic resonance image adjusted by the first adjusting subunit. The first filtering subunit may be configured to perform filtering processing on the patient resting function magnetic resonance image denoised by the first denoising subunit.
The first imaging characteristic calculating unit in this embodiment may be configured to calculate a first pearson correlation coefficient of the patient whole-brain functional link and the standard brain functional link according to the following formula (1), and use each first pearson correlation coefficient corresponding to each preset brain region seed point in the patient resting functional magnetic resonance image as the first characteristic. For example, when the imaging characteristics are calculated by using N predetermined brain region seed points in the present embodiment, N first characteristics can be obtained according to the following formula (1), where N is greater than or equal to 1.
The meaning of each parameter in the formula (1) is as follows:
pattern is a 1-dimensional vector corresponding to the patient whole brain function connection, template is a 1-dimensional vector corresponding to the standard brain function connection, r (pattern, template) is a Pearson correlation coefficient of the 1-dimensional vector pattern and the 1-dimensional vector template, Cov (pattern, template) is a covariance of the 1-dimensional vector pattern and the 1-dimensional vector template, Var (pattern) is a variance of the 1-dimensional vector pattern, and Var (template) is a variance of the 1-dimensional vector template.
The second image processing unit in this embodiment may include a second preprocessing subunit and a standard brain function connection calculating subunit. The second preprocessing subunit may be configured to preprocess the normal person resting function magnetic resonance image in the preset training set. The standard brain function connection statistics subunit may be configured to calculate the whole brain function connection of each of the preset brain region seed points according to the normal person resting function magnetic resonance image preprocessed by the second preprocessing subunit, and further to count the whole brain function connections of all the preset brain region seed points to obtain the standard brain function connection.
In a preferred embodiment of this embodiment, the standard brain function connection calculating unit may obtain the standard brain function connection according to the following steps: first, an average time sequence of all brain region seed points in each preset brain function network, such as a default network and an execution control network, is calculated. And then, calculating the whole brain function connection according to the calculated average time sequence, and obtaining a statistical T value graph of the calculated whole brain function connection by using a single sample T test method, wherein the statistical T value graph is the final standard brain function connection.
Referring to fig. 2 to 7, fig. 2 schematically shows a T-value graph of a default network in a brain function network in the present embodiment, fig. 3 schematically shows a T-value graph of an execution control network in the brain function network in the present embodiment, fig. 4 schematically shows a T-value graph of a highlight network in the brain function network in the present embodiment, fig. 5 schematically shows a T-value graph of a movement network in the brain function network in the present embodiment, fig. 6 schematically shows a T-value graph of an auditory network in the brain function network in the present embodiment, and fig. 7 schematically shows a T-value graph of a visual network in the brain function network in the present embodiment. Wherein the scales shown in fig. 2-7 are all T values.
In another preferred embodiment of this embodiment, the second preprocessing subunit may include a second sampling subunit, a second adjusting subunit, a third adjusting subunit, a second denoising subunit, and a second filtering subunit. The second sampling subunit may be configured to resample the normal human resting function magnetic resonance image in the preset training set. The second adjusting subunit may be configured to move and/or rotate the normal human resting function magnetic resonance image obtained by resampling by the second sampling subunit, so as to eliminate a position deviation between different normal human resting function magnetic resonance images. The third adjusting subunit may be configured to register the normal human resting function magnetic resonance image adjusted by the second adjusting subunit to the MNI space. The second denoising subunit can be configured to denoise the normal person resting function magnetic resonance image registered by the third adjusting subunit. The second filtering subunit is configured to perform filtering processing on the normal person resting function magnetic resonance image denoised by the second denoising subunit.
As can be seen from the foregoing, the preset brain region seed points in this embodiment may include a first type brain region seed point, a second type brain region seed point, a third type brain region seed point, a fourth type brain region seed point, a fifth type brain region seed point, and a sixth type brain region seed point. The coordinates of the brain region seed points configured to the MNI space are shown in table 1 below:
TABLE 1
In this embodiment, the second imaging characteristic calculating unit may be configured to calculate second pearson correlation coefficients of any two predetermined brain region seed points according to the following formula (2) and an average functional magnetic resonance signal of each predetermined brain region seed point in the patient resting functional magnetic resonance image, and use each second pearson correlation coefficient corresponding to each predetermined brain region seed point as the second characteristic. For example, when the imaging characteristics are calculated by using N predetermined brain region seed points in the present embodiment, N × (N-1)/2 second characteristics can be obtained according to the following formula (2), where N ≧ 1.
The meaning of each parameter in the formula (2) is as follows:
r(MP1,MP2) A second Pearson correlation coefficient, M, between the brain region seed point and the brain region seed point in the magnetic resonance image of the patient's resting functionP1And MP2Are respectively brainsMean functional magnetic resonance signal of region seed points and brain region seed points, Cov (M)P1,MP2) To average the functional magnetic resonance signal MP1And average functional magnetic resonance signal MP2Covariance of (1), Var (M)P1) And Var (M)P2) Respectively mean functional magnetic resonance signal MP1And average functional magnetic resonance signal MP2The variance of (c).
Further, the imaging characteristic screening module 12 in this embodiment may include a first screening unit and a second screening unit. The first screening unit can screen out the imaging characteristics which accord with the primary screening conditions. The second screening unit can further screen the result obtained by the first screening unit to obtain the final imaging characteristic.
The first screening unit in this embodiment may be configured to perform the following operations: first, a third pearson correlation coefficient between the imaging characteristic and the consciousness deterioration prognostic indicator is calculated according to the following equation (3). And then, selecting the imaging characteristics of which the third Pearson correlation coefficient is greater than a preset first threshold value to obtain the imaging characteristics which accord with the primary screening condition.
The meaning of each parameter in the formula (3) is as follows:
feature is a vision feature vector, CRS _ Score is a consciousness disturbance prognosis index vector, r (feature, CRS _ Score) is a pearson correlation coefficient between the vision feature vector feature and the consciousness disturbance prognosis index vector CRS _ Score, Cov (feature, CRS _ Score) is a covariance between the vision feature vector feature and the consciousness disturbance prognosis index vector CRS _ Score, Var (feature) is a variance between the vision feature vector feature, and Var (CRS _ Score) is a variance between the consciousness disturbance prognosis index vector CRS _ Score. In a preferred embodiment of this embodiment, the glasgow coma scale Score is used, i.e. the disturbance of consciousness prognosis indicator vector CRS _ Score is the Score vector of the glasgow coma scale Score.
In this embodiment, the second screening unit may be configured to use a Competitive Adaptive weighted Sampling algorithm (CARS) and a partial Least squares algorithm (PLS), and use, as the final imaging feature, the imaging feature with the feature weight larger than a preset second threshold among the imaging features selected by the first screening unit. For example, the second screening unit may obtain the final imaging characteristics by using the method disclosed in Li H. -D, Xu Q. -S, Liang Y. -Z (2014) LibPLS: An Integrated Library for Partial Least Squares Regression and characterization analysis.
Further, in this embodiment, the imaging characteristic processing module may be configured to merge the imaging characteristics obtained by the imaging characteristic screening module and the clinical information in the preset training set into a characteristic vector X.
Further, in this embodiment, the model training module 14 may be configured to train a partial least squares regression model shown in the following formula (4) according to the feature vector, so as to obtain a final prediction model.
Y＝XB+B0 (4)
The meaning of each parameter in the formula (4) is as follows:
x is a feature vector containing imaging features and clinical information, Y is an index for prognosis of disturbance of consciousness, B and B0Respectively a feature weight vector to be estimated and a bias vector to be estimated. When the feature weight vector B and the offset vector B0After the value of (3) is determined, the consciousness disturbance prognosis index Y corresponding to the characteristic vector X can be obtained according to any characteristic vector X.
Referring to fig. 8-13, the brain consciousness state prediction result obtained by the human brain consciousness state prediction apparatus according to the preset training set is analyzed. Wherein, the preset training set comprises 63 training samples.
Referring first to fig. 8, fig. 8 is a diagram illustrating a predicted brain consciousness state of a patient with disturbance of consciousness in a pre-set training set according to an exemplary embodiment. As shown in fig. 8, the abscissa is the clinical behavior score when scanning magnetic resonance and the ordinate is the glasgow coma prognosis score. The circles indicate patients with a glasgow coma prognosis score of 2 or less, the triangles indicate patients with a glasgow coma prognosis score of 3 or less, and the pentagons indicate patients with a glasgow coma prognosis score of 4 or more. Wherein a glasgow coma prognosis score of 2 or less indicates no recovery of consciousness, and a glasgow coma prognosis score of 3 or more indicates recovery of consciousness.
With continuing reference to fig. 9, fig. 9 is a diagram illustrating the correlation between the predicted brain awareness status and the prognosis score in the present embodiment. As shown in fig. 9, in the present embodiment, based on the result of predicting the state of brain consciousness shown in fig. 9, it can be obtained that the pearson correlation coefficient with the actual prognosis score is 0.81, and the determination coefficient R is 02＝0.65。
With continued reference to fig. 10, fig. 10 is a diagram illustrating the consistency between the predicted brain awareness state and the prognosis score in the present embodiment. As shown in fig. 10, in this embodiment, the data points are distributed on both sides of the direct current where y is 0, which means that the predicted result has no significant difference from the actual prognosis score, i.e., the deviation of the predicted result is small and most of the deviations are within ± 1.96 standard deviations, and the deviation of the predicted result is less. Wherein, the parameter test p is 1.0, and the nonparametric KS test p is 0.666.
With continued reference to fig. 11 and 12, fig. 11 illustrates the number of persons having actual prognosis of consciousness of patients with different brain consciousness state prediction results in the present embodiment, and fig. 12 illustrates the ratio of actual prognosis of consciousness of patients with different brain consciousness state prediction results in the present embodiment. As shown in fig. 11 and 12, in the present embodiment, as the score of the predicted result increases, the number of patients who recover consciousness and the ratio thereof also increase, indicating that the score of the predicted result can characterize the recovery level of the patients.
With continued reference to fig. 13, fig. 13 illustrates an ROC curve of the brain consciousness state prediction result in the present embodiment. As shown in fig. 13, in this embodiment, based on the brain consciousness state prediction result shown in fig. 9, an AUC area corresponding to the ROC curve is 0.96, which indicates that the prediction result has better prediction on the consciousness restoration level of the patient.
Referring to fig. 14-16, the brain consciousness state prediction result obtained by the human brain consciousness state prediction apparatus according to the preset test set in this embodiment is analyzed.
Referring first to fig. 14, fig. 14 is a diagram illustrating a predicted brain consciousness state of a patient with central disturbance of consciousness through a predetermined test in accordance with the present embodiment. As shown in fig. 14, the abscissa is the clinical behavior score when scanning magnetic resonance and the ordinate is the glasgow coma prognosis score. The circles indicate patients with a glasgow coma prognosis score of 2 or less, the triangles indicate patients with a glasgow coma prognosis score of 3 or less, and the pentagons indicate patients with a glasgow coma prognosis score of 4 or more. Wherein a glasgow coma prognosis score of 2 or less indicates no recovery of consciousness, and a glasgow coma prognosis score of 3 or more indicates recovery of consciousness.
With continuing reference to fig. 15, fig. 15 is a diagram illustrating the correlation between the predicted brain awareness status and the prognosis score in the present embodiment. As shown in fig. 15, in the present embodiment, based on the result of predicting the state of brain consciousness shown in fig. 14, it can be obtained that the pearson correlation coefficient with the actual prognosis score is 0.61, and the determination coefficient R is 02＝0.35。
With continued reference to fig. 16, fig. 16 is a diagram illustrating the consistency between the predicted brain awareness state and the prognosis score in the present embodiment. As shown in fig. 16, the data points in this embodiment are distributed on both sides of the case where y is 0.14 dc, which means that the predicted result has no significant difference from the actual prognosis score, i.e., the deviation of the predicted result is small and most of the deviation is within ± 1.96 standard deviation, and the predicted deviation outlier is small. Wherein, the parameter test p is 0.89, and the nonparametric KS test p is 0.506.
In summary, based on the analysis of the brain consciousness state prediction results shown in fig. 8 to 16, it can be shown that the human brain consciousness state prediction apparatus in the present embodiment can well predict the prognosis state of the patient after coma.
It will be appreciated by those skilled in the art that the aforementioned human brain awareness state prediction apparatus may also include other known structures, such as a processor, a controller, a memory, etc., wherein the memory includes, but is not limited to, a random access memory, a flash memory, a read only memory, a programmable read only memory, a volatile memory, a non-volatile memory, a serial memory, a parallel memory or a register, etc., and the processor includes, but is not limited to, a CPLD/FPGA, a DSP, an ARM processor, a MIPS processor, etc., and these known structures are not shown in fig. 1 in order to unnecessarily obscure the embodiments of the present disclosure.
It should be understood that the number of individual modules in fig. 1 is merely illustrative. The number of modules may be any according to actual needs.
Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the claims of the present invention, any of the claimed embodiments may be used in any combination.
The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components in a server, client, or the like, according to embodiments of the present invention. The present invention may also be embodied as an apparatus or device program (e.g., PC program and PC program product) for carrying out a portion or all of the methods described herein. Such a program implementing the invention may be stored on a PC readable medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed PC. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (10)
1. A human brain consciousness state prediction device is characterized by comprising a consciousness state prediction model which is configured to predict the consciousness state of the human brain according to the imaging characteristics of the human brain and corresponding clinical information; the consciousness state prediction model comprises an iconography characteristic calculation module, an iconography characteristic screening module, an iconography characteristic processing module and a model training module;
the imaging characteristic calculation module is configured to calculate imaging characteristics according to the resting function magnetic resonance images in a preset training set; the imaging characteristics comprise a first characteristic and a second characteristic, the first characteristic is the spatial similarity between the whole brain function connection and the standard brain function connection of the patient, and the second characteristic is the function connection of a preset brain region seed point in the resting function magnetic resonance image of the patient;
the imaging characteristic screening module is configured to screen the imaging characteristics according to the consciousness disturbance prognostic indicators; the consciousness disturbance prognosis index is a consciousness disturbance coma scale score corresponding to the prognosis of the patient in the preset training set;
the imaging characteristic processing module is configured to combine the imaging characteristics obtained by screening by the imaging characteristic screening module and the clinical information in the preset training set into a characteristic vector;
and the model training module is configured to train a preset partial least square regression model according to the feature vector to obtain a final prediction model.
2. The apparatus of claim 1, wherein the imaging feature calculation module comprises a first imaging feature calculation unit and a second imaging feature calculation unit;
the first imaging characteristic calculation unit is configured to calculate first pearson correlation coefficients of the patient whole brain function connection and the standard brain function connection, and take each first pearson correlation coefficient corresponding to each preset brain region seed point in the patient resting function magnetic resonance image as a first characteristic;
the second imaging characteristic calculating unit is configured to calculate second pearson correlation coefficients of any two preset brain region seed points according to an average functional magnetic resonance signal of each preset brain region seed point in the patient resting functional magnetic resonance image, and use each second pearson correlation coefficient corresponding to each preset brain region seed point as a second characteristic.
3. The apparatus of claim 1, wherein the imaging feature screening module comprises a first screening unit and a second screening unit;
the first screening unit is configured to perform the following operations:
calculating a third Pearson correlation coefficient between the imaging characteristic and the consciousness disturbance prognostic index:
selecting the imaging characteristics of which the third Pearson correlation coefficient is greater than a preset first threshold;
the second screening unit is configured to adopt a competitive adaptive re-weighting sampling algorithm and a partial least square algorithm, and take the image characteristics with the characteristic weight larger than a preset second threshold value in the image characteristics selected and obtained by the first screening unit as final image characteristics.
4. The apparatus of claim 3,
the preset partial least squares regression model is shown as follows:
Y＝XB+B0
wherein, X is a feature vector containing imaging features and clinical information, and Y is an consciousness disturbance prognostic indicator; b and B0Respectively a feature weight vector to be estimated and a bias vector to be estimated.
5. The apparatus according to any one of claims 1 to 4,
the disturbance of consciousness coma scale score comprises a glasgow coma scale score;
the clinical information includes the age, etiology, and course of the patient; the causes include traumatic brain disturbance, stroke brain disturbance and hypoxic brain disturbance.
6. The apparatus of claim 2, wherein the imaging characteristic calculation module further comprises a first image processing unit; the first image processing unit comprises a first preprocessing subunit and a patient brain function connection calculating subunit;
the first preprocessing subunit is configured to preprocess the patient resting function magnetic resonance image in the preset training set;
the patient brain function connection calculation unit is configured to calculate the whole brain function connection of the preset brain region seed points according to the patient resting function magnetic resonance image preprocessed by the first preprocessing subunit, and further count the whole brain function connection of all the preset brain region seed points to obtain the patient brain function connection.
7. The apparatus of claim 6,
the first preprocessing subunit comprises a first sampling subunit, a first adjusting subunit, a first denoising subunit and a first filtering subunit;
the first sampling subunit is configured to resample the patient resting function magnetic resonance image in the preset training set;
the first adjusting subunit is configured to move and/or rotate the patient rest function magnetic resonance images obtained by resampling through the first sampling subunit so as to eliminate position deviation among different patient rest function magnetic resonance images;
the first denoising subunit is configured to denoise the patient resting function magnetic resonance image adjusted by the first adjusting subunit;
the first filtering subunit is configured to perform filtering processing on the patient rest function magnetic resonance image denoised by the first denoising subunit.
8. The apparatus of claim 2, wherein the imaging characteristic calculation module further comprises a second image processing unit; the second image processing unit comprises a second preprocessing subunit and a standard brain function connection calculating subunit;
the second preprocessing subunit is configured to preprocess the normal person resting function magnetic resonance image in the preset training set;
and the standard brain function connection counting subunit is configured to calculate the whole brain function connection of each preset brain region seed point according to the normal person resting function magnetic resonance image preprocessed by the second preprocessing subunit, and further count the whole brain function connection of all the preset brain region seed points to obtain the standard brain function connection.
9. The apparatus of claim 8, wherein the second pre-processing sub-unit comprises a second sampling sub-unit, a second conditioning sub-unit, a third conditioning sub-unit, a second denoising sub-unit, and a second filtering sub-unit;
the second sampling subunit is configured to resample the magnetic resonance image of the rest function of the normal person in the preset training set;
the second adjusting subunit is configured to move and/or rotate the normal person resting function magnetic resonance images obtained by resampling through the second sampling subunit so as to eliminate the position deviation between different normal person resting function magnetic resonance images;
the third adjusting subunit is configured to register the normal resting function magnetic resonance image of the person adjusted by the second adjusting subunit to an MNI space;
the second denoising subunit is configured to denoise the normal person resting function magnetic resonance image registered by the third adjusting subunit;
the second filtering subunit is configured to perform filtering processing on the normal person resting function magnetic resonance image denoised by the second denoising subunit.
10. The apparatus of claim 1, 2, 6 or 8,
the preset brain region seed points comprise a first type brain region seed point, a second type brain region seed point, a third type brain region seed point, a fourth type brain region seed point, a fifth type brain region seed point and a sixth type brain region seed point;
the first type brain region seed points include medial frontal lobe, posterior cingulate gyrus and cuneiform lobe, left parietal lobe lateral cortex and right parietal lobe lateral cortex within a default network;
the second type of brain region seed point comprises an dorsolateral prefrontal lobe, a left prefrontal lobe, a right prefrontal lobe, a left apical cortex, and a right apical cortex within an executive control network;
the third type brain region seed points comprise a left front brain island, a right front brain island and a front back in front of the button in the highlight network;
the fourth type brain region seed point comprises a left primary motor cortex, a right primary motor cortex and an auxiliary motor region in the somatosensory motor network;
the fifth type brain region seed point comprises a left primary auditory cortex, a right primary auditory cortex and a cingulum retrocenter in an auditory network;
the sixth type of brain region seed point includes a left primary visual cortex, a right primary visual cortex, a left contact visual cortex, and a right contact visual cortex within the visual network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711002297.5A CN107993714B (en) | 2017-10-24 | 2017-10-24 | Human brain consciousness state predicting device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711002297.5A CN107993714B (en) | 2017-10-24 | 2017-10-24 | Human brain consciousness state predicting device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107993714A CN107993714A (en) | 2018-05-04 |
| CN107993714B true CN107993714B (en) | 2020-11-13 |
Family
ID=62031066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711002297.5A Active CN107993714B (en) | 2017-10-24 | 2017-10-24 | Human brain consciousness state predicting device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107993714B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN113082447B (en) * | 2021-04-02 | 2021-12-07 | 电子科技大学 | Prediction method for music modulation brain plasticity effect of fMRI brain loop |
| CN113077483B (en) * | 2021-04-27 | 2022-03-25 | 中国科学院自动化研究所 | Method and system for predicting distribution of induced electric field based on brain tissue and depth regression |
| CN117011657A (en) * | 2023-07-05 | 2023-11-07 | 深圳市龙岗中心医院 | Brain CT image processing method and related equipment based on self-attention mechanism fusion |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101739681A (en) * | 2009-12-14 | 2010-06-16 | 西北工业大学 | Related prediction model-based method for detecting structural deformation in magnetic resonance image |
| CN102293656A (en) * | 2011-05-25 | 2011-12-28 | 四川大学华西医院 | Emotional stability evaluation system based on magnetic resonance imaging and evaluation method thereof |
| CN105534534A (en) * | 2015-12-09 | 2016-05-04 | 浙江大学 | Emotion recognition method, device and system based on real-time functional magnetic resonance |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9510756B2 (en) * | 2012-03-05 | 2016-12-06 | Siemens Healthcare Gmbh | Method and system for diagnosis of attention deficit hyperactivity disorder from magnetic resonance images |
| US20140107521A1 (en) * | 2012-10-12 | 2014-04-17 | Case Western Reserve University | Functional brain connectivity and background noise as biomarkers for cognitive impairment and epilepsy |
-
2017
- 2017-10-24 CN CN201711002297.5A patent/CN107993714B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101739681A (en) * | 2009-12-14 | 2010-06-16 | 西北工业大学 | Related prediction model-based method for detecting structural deformation in magnetic resonance image |
| CN102293656A (en) * | 2011-05-25 | 2011-12-28 | 四川大学华西医院 | Emotional stability evaluation system based on magnetic resonance imaging and evaluation method thereof |
| CN105534534A (en) * | 2015-12-09 | 2016-05-04 | 浙江大学 | Emotion recognition method, device and system based on real-time functional magnetic resonance |
Non-Patent Citations (4)
| Title |
|---|
| DPABI: Data Processing & Analysis for (Resting-State)Brain Imaging;Chao-Gan Yan等;《Neuroinformatics》;20160731;第14卷(第3期);第339-351页 * |
| 基于静息态功能连接的杏仁核谱聚类研究;田洪君等;《磁共振成像》;20161031;第7卷(第10期);第769-774页 * |
| 脑影像与脑认知发展研究;蒋田仔等;《2012-2013控制科学与工程学科发展报告》;20140430;第141-151页 * |
| 静息态磁共振脑影像的动态功能连接模式分析;李圳峰;《中国优秀硕士学位论文全文数据库 医药卫生科技辑》;20170315(第3期);第E059-90页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107993714A (en) | 2018-05-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107993714B (en) | Human brain consciousness state predicting device | |
| Dai et al. | Identifying and mapping connectivity patterns of brain network hubs in Alzheimer's disease | |
| US10810512B1 (en) | Validating a machine learning model prior to deployment | |
| Alberts et al. | Dissociating vestibular and somatosensory contributions to spatial orientation | |
| KR102346696B1 (en) | Brain function test system and its device | |
| CA2750347C (en) | Methods and apparatuses for quantitatively determining the likelihood of a disease | |
| Schmithorst et al. | Left ear advantage in speech-related dichotic listening is not specific to auditory processing disorder in children: A machine-learning fMRI and DTI study | |
| WO2020218460A1 (en) | Diagnostic assistance device, speculation device, diagnostic assistance system, diagnostic assistance method, diagnostic assistance program, and learned model | |
| WO2013076211A1 (en) | Method and device improving visual performance | |
| Zhao et al. | Neural correlates of cognitive dysfunctions in cervical spondylotic myelopathy patients: a resting-state fMRI study | |
| US20140296733A1 (en) | Inter-rater and intra-rater reliability of physiological scan interpretation | |
| KR102172195B1 (en) | Method and apparatus for diagnosing lumbar spinal stenosis | |
| Huber et al. | Network changes in patients with phobic postural vertigo | |
| US20120296194A1 (en) | Method and system for determining a condition of a brain | |
| Turner et al. | Behavioral measures of saccade latency and inhibition in manifest and premanifest Huntington's disease | |
| Zhu et al. | Cortical atrophy is associated with cognitive impairment in Parkinson’s disease: A combined analysis of cortical thickness and functional connectivity | |
| WO2019235367A1 (en) | Tomographic image prediction device and tomographic image prediction method | |
| WO2013188408A1 (en) | Method of obtaining and analyzing data from an upright mri from the spinal region of a subject | |
| US11670423B2 (en) | Method and system for early detection of neurodegeneration using progressive tracking of eye-markers | |
| CN112911988A (en) | Method for adapting an ophthalmic device according to a visual exploration strategy of a wearer | |
| Presson et al. | Quantifying white matter structural integrity with high-definition fiber tracking in traumatic brain injury | |
| Ankuda et al. | Assessing health care use trajectories after the onset of functional disability: Application of a group-based trajectory model | |
| Presson et al. | An exploratory analysis linking neuropsychological testing to quantification of tractography using High Definition Fiber Tracking (HDFT) in military TBI | |
| Talian et al. | Normative ranges for, and interrater reliability of, rotational vestibular and balance tests in US Military Service Members and Veterans | |
| de Best et al. | Convergence along the visual hierarchy is altered in posterior cortical atrophy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |