CN111739647B - Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device - Google Patents
Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device Download PDFInfo
- Publication number
- CN111739647B CN111739647B CN202010793478.XA CN202010793478A CN111739647B CN 111739647 B CN111739647 B CN 111739647B CN 202010793478 A CN202010793478 A CN 202010793478A CN 111739647 B CN111739647 B CN 111739647B
- Authority
- CN
- China
- Prior art keywords
- brain
- image data
- tester
- biological characteristic
- brain image
- 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
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
Abstract
The invention provides a bipolar disorder suicide risk prediction method based on a magnetic resonance image and a related device, wherein the method comprises the following steps: acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state; inputting the image data into a trained suicide risk prediction model, and determining a risk assessment value corresponding to the biological characteristic index; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested. The invention can avoid the uncertainty and misdiagnosis rate caused by depending on the professional knowledge and subjective judgment of doctors, and can provide theoretical support for the diagnosis of patients.
Description
Technical Field
The invention relates to the technical field of functional magnetic resonance images, in particular to a bipolar disorder suicide risk prediction method based on a magnetic resonance image and a related device.
Background
With the development of economic, sanitary and medical levels, the average life of people in all countries in the world is prolonged. At the same time, the incidence of mental illness has increased year by year and is one of the leading causes of death worldwide due to many factors such as increased competitive pressure.
In the case of recurrent mental diseases with high suicide risk, such as bipolar disorder, improper treatment and poor predictability, adverse events such as suicide are frequently caused.
Currently, many clinical scoring tables are used to identify symptoms of bipolar disorder disease to help identify bipolar disorder early, which, while having some utility, still lack sufficient specificity to reduce diagnostic uncertainty, rely heavily on the expertise and empirical judgment of physicians, and fail to predict suicidal propensity and early intervention in patients with bipolar disorder.
Disclosure of Invention
In view of the above, the invention provides a magnetic resonance image-based bipolar disorder suicide risk prediction method and a related device, so as to avoid uncertainty and misdiagnosis rate caused by depending on professional knowledge and subjective judgment of a doctor and provide theoretical support for diagnosis of a patient.
In a first aspect, the present invention provides a method for predicting suicidal risk of bipolar disorder based on magnetic resonance imaging, the method comprising: acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state; inputting the image data into a trained suicide risk prediction model, and determining a risk assessment value corresponding to the biological characteristic index; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested.
In a second aspect, the present invention provides a training method for a bipolar disorder suicide risk prediction model based on magnetic resonance imaging, the method comprising: acquiring brain image data corresponding to a plurality of testers; the image data comprises functional brain image data and structural brain image data; the plurality of testers are divided into a first tester and a second tester; the first tester is at risk of suicide; the second tester is not at risk of suicide; preprocessing the functional brain image data according to the structural image data of each tester; calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester; determining the characteristic brain area corresponding to the first tester according to the comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject; constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and using the regression model as an initial suicide risk prediction model; and training the initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained.
In a third aspect, the present invention provides a magnetic resonance image-based device for predicting suicidal risk of bipolar disorder, comprising: the device comprises an acquisition module and a determination module; the acquisition module is used for acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state; the determining module is used for inputting the image data into a trained suicide risk prediction model and determining a risk assessment value corresponding to the biological characteristic index; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested.
In a fourth aspect, the present invention provides a training apparatus for a bipolar disorder suicide risk prediction model based on magnetic resonance imaging, including: the device comprises an acquisition module, a processing module, a calculation module, a determination module, a construction module and a training module; the acquisition module is used for acquiring brain image data corresponding to a plurality of testers; the image data comprises functional brain image data and structural brain image data; the plurality of testers are divided into a first tester and a second tester; the first tester is at risk of suicide; the second tester is not at risk of suicide; the processing module is used for preprocessing the functional brain image data according to the structural image data of each tester; the calculation module is used for calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester; the determining module is used for determining the characteristic brain area corresponding to the first tester according to the comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject; the construction module is used for constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and the regression model is used as an initial suicide risk prediction model; and the training module is used for training the initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained.
In a fifth aspect, the present invention provides an electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor to implement the suicide risk prediction method of the first aspect or to implement the training method of the suicide risk prediction model of the second aspect.
In a sixth aspect, the present invention provides a storage medium having stored thereon a computer program which, when executed by a processor, implements a suicide risk prediction method as described in the first aspect or implements a training method of a suicide risk prediction model as described in the second aspect.
The invention provides a bipolar disorder suicide risk prediction method based on a magnetic resonance image and a related device, wherein the method comprises the following steps: acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state; inputting the image data into a trained suicide risk prediction model, and determining a risk assessment value corresponding to the biological characteristic index; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested. According to the suicide risk prediction method, the suicide risk prediction model is obtained by establishing the relation between the biological characteristic indexes of the characteristic brain area of the patient and the risk assessment, and for each person to be tested, the risk assessment value of the person to be tested can be obtained by inputting the biological characteristic indexes into the model, so that uncertainty and misdiagnosis rate caused by depending on professional knowledge and subjective judgment of doctors are avoided, and theoretical support can be provided for diagnosis of the patient.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flowchart of a training method of a bipolar disorder suicide risk prediction model based on magnetic resonance imaging according to an embodiment of the present invention;
FIG. 2 illustrates an implementation of step S102 provided in the present invention;
FIG. 3 illustrates an implementation of step S103 provided in the present invention;
FIG. 4 shows an implementation of step S105 provided in the present invention;
fig. 5 is a schematic flowchart of a method for predicting suicidal risk of bipolar disorder based on magnetic resonance imaging according to an embodiment of the present invention;
fig. 6 is a functional block diagram of a magnetic resonance image-based bipolar disorder suicide risk prediction apparatus according to an embodiment of the present invention;
FIG. 7 is a functional block diagram of a training apparatus for a magnetic resonance imaging-based bipolar disorder suicide risk prediction model according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", etc. indicate an orientation or a positional relationship based on that shown in the drawings or that the product of the present invention is used as it is, this is only for convenience of description and simplification of the description, and it does not indicate or imply that the device or the element referred to must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present invention.
Furthermore, the appearances of the terms "first," "second," and the like, if any, are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.
It should be noted that the features of the embodiments of the present invention may be combined with each other without conflict.
Currently, bipolar disorder is a recurrent mental disease, accompanied by cognitive and functional disorders, fluctuation of emotional state and energy, and behavioral changes, misdiagnosis of bipolar disorder as depression may lead to inappropriate treatment, increased suicidal tendency, and other adverse events, and many clinical scoring components are used to help identify bipolar disorder at an early stage, such as beck suicide scale scoring, although the above clinical measures have some utility, but lack sufficient specificity to reduce the uncertainty of diagnosis, and are heavily dependent on the professional knowledge and subjective judgment of doctors, which cannot provide theoretical support for diagnosis of patients with bipolar disorder.
In order to solve the above technical problems, an embodiment of the present invention provides a training method for a dual-phase obstacle suicide risk prediction model based on a magnetic resonance image and a suicide risk prediction method implemented based on the suicide risk prediction model, and the core of the method is as follows: the characteristic information is extracted from the brain function image, the suicide risk prediction model is obtained by establishing the relation between the characteristic information and suicide symptoms, and the obtained suicide risk prediction model can be used for conducting suicide risk prediction on any tester.
To facilitate understanding of the principle of suicide risk prediction implemented by the embodiment of the present invention, a method for training a suicide risk prediction model provided by the embodiment of the present invention is described first, referring to fig. 1, where fig. 1 is a schematic flowchart of a method for training a magnetic resonance image-based bipolar disorder suicide risk prediction model provided by the embodiment of the present invention, and the method includes the following steps:
s101, acquiring brain image data corresponding to a plurality of testers; the brain image data comprises functional brain image data and structural brain image data; the multiple testers are divided into a first tester and a second tester; the first tester has a suicide risk; the second tester had no suicide risk.
In an embodiment of the present invention, the brain image data may be obtained by Magnetic Resonance Imaging (MRI), can include structural magnetic resonance imaging (structural MRI) brain image data and resting-state functional MRI brain image data (Rs-fMRI for short), wherein, in order to establish the relationship between the brain region characteristic information and the suicide symptom, among a plurality of testers performing data sample collection, a tester having suicide risk, namely a bipolar disorder mental disease patient, is taken as a first tester, a tester not having suicide risk, namely a healthy person, is taken as a second tester, in the present example, considering that patients with suicidal untreated bipolar disorder and patients with suicidal idemic bipolar disorder are at risk of suicide, thus, the first test subject may also include patients with suicidal untreated bipolar disorder and patients with suicidal idemic bipolar disorder.
S102, preprocessing functional brain image data according to the structural brain image data of each tester.
S103, calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester.
In an embodiment of the present invention, the above-mentioned biometric indicator may be dynamic low-frequency oscillation amplitude (referred to as dALFF) (for convenience of description, the following biometric indicators are all expressed by dALFF values), and represents fluctuation intensity of brain region functional activity in a resting state, and related technical studies indicate that internal brain functional activity is time-varying and dynamically-varying and is related to behaviors currently ongoing, and feature information is used to represent high-time-varying and fluctuating internal brain activity.
S104, determining a characteristic brain area corresponding to the first tester according to a comparison result between biological characteristic indexes corresponding to all brain areas of the first tester and biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject.
In the present embodiment, according to the above embodiment, the first test subject may be a patient with suicidal ideation bipolar disorder and a patient with suicidal non-adoptive bipolar disorder, and the second test subject is a healthy subject, and in order to obtain the relationship between the dynamic low-frequency oscillation amplitude and the suicidal behavior (mainly suicidal ideation and non-adoptive), two control groups may be obtained: the first group of control group is the brain area dALFF value of the patient with the suicidal ideographic disorder and the brain area dALFF value of the healthy patient, the second group of control group is the brain area dALFF value of the patient with the suicidal non-adoptive bipolar disorder and the brain area dALFF value of the healthy patient, in the control process, the characteristic brain area of the patient with the suicidal ideographic disorder in the first control group can be obtained by adopting a statistical analysis method through one-factor variance analysis and post-test, the characteristic brain area of the patient with the suicidal non-bipolar disorder in the second control group is obtained, and in the control process, the control is carried out based on the voxel level of the whole brain area of each test person.
It is understood that the characteristic brain region may be understood as a brain region having a difference in biological characteristic index with respect to the characteristic index of the brain region of a healthy subject, for example, a difference in DALFF of the prefrontal lobe between a patient with suicidal idenfative bipolar disorder and a healthy subject, which may be expressed as: if the DALFF of the prefrontal lobe of the patient with suicidal non-adoptive bipolar disorder is lower than the DALFF of the prefrontal lobe of the healthy subject, or if the DALFF of the prefrontal lobe of the patient with suicidal non-adoptive bipolar disorder is higher than the DALFF of the prefrontal lobe of the healthy subject, then the prefrontal lobe may serve as the characteristic brain region of the patient with suicidal non-adoptive bipolar disorder.
And S105, constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and using the regression model as an initial suicide risk prediction model.
In an embodiment of the present invention, the biometric indicator corresponding to the characteristic brain region may be a biometric indicator corresponding to each voxel of the characteristic brain region, the preset risk assessment value is a beck suicide scale score of each tester, for the DALFF value corresponding to the characteristic brain region of each first tester, the beck suicide scale score of each first tester is used as a regression indicator to perform multiple linear regression, a regression relationship between the DALFF value corresponding to the characteristic brain region and the beck suicide scale score is established, through the regression relationship, the DALFF value of any tester can be predicted to obtain the corresponding beck suicide scale score, the DALFF value can be used as a risk assessment value of the tester, the regression model established through the regression relationship can be used as an initial suicide risk prediction model, and the prediction capability of the model is improved by training the initial suicide risk prediction model, until a trained suicide risk prediction model is obtained.
And S106, training the initial suicide risk prediction model according to the corresponding biological characteristic indexes of each tester until the trained suicide risk prediction model is obtained.
In the embodiment of the present invention, a more common unbiased policy-based "leave one" method may be adopted to establish a robust and reliable suicide risk prediction model, which specifically may be: for the multiple testers, one dALFF value of the tester is reserved as a test data set during each training, the dALFF values of the other testers are used as training data sets, and through cross validation iteration (for example, the iteration time is 5000 times), the Beck suicide scale score of the test data set is predicted according to the prediction score of the training data sets until a trained suicide risk prediction model is obtained.
The training method of the suicide risk prediction model provided by the embodiment of the invention comprises the following steps: acquiring brain image data corresponding to a plurality of testers; the multiple testers are divided into a first tester and a second tester; the first tester has a suicide risk; the second tester had no suicide risk; preprocessing each image brain data and calculating biological characteristic indexes corresponding to all brain areas of each tester; determining a characteristic brain area corresponding to the first tester according to a comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject; constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and using the regression model as an initial suicide risk prediction model; and training the initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained. According to the suicide risk prediction method, the suicide risk prediction model is obtained by establishing the relation between the biological characteristic indexes of the characteristic brain area of the patient and the risk assessment, and for each person to be tested, the risk assessment value of the person to be tested can be obtained by inputting the biological characteristic indexes into the model, so that uncertainty and misdiagnosis rate caused by depending on professional knowledge and subjective judgment of doctors are avoided, and theoretical support can be provided for diagnosis of the patient.
Optionally, in order to obtain the suicide risk prediction model, it is necessary to obtain a dALFF value corresponding to a characteristic brain region, and further, a relationship between the dALFF value and a risk assessment value may be established, and in order to obtain dynamic low-frequency oscillation amplitude values corresponding to all brain regions, it is necessary to pre-process brain image data (including functional image data and structural image data) of a tester, and obtain a dynamic low-frequency oscillation amplitude corresponding to each brain region based on the processed brain image data, for convenience of understanding, in the following, taking a tester as an example, an implementation manner of pre-processing brain image data corresponding to each tester is given, referring to fig. 2, where fig. 2 is a possible implementation manner of S102 provided by an embodiment of the present invention, and includes the following steps:
s102-1, obtaining a brain resting state blood oxygen dependent level (BOLD) time sequence signal corresponding to functional brain image data of the tester.
It is to be understood that the time-series signal may include a plurality of time points, and in a practical application scenario, the length of the time-series signal may be obtained according to a practical scanning duration.
S102-2, functional brain image data of the first N time points in the blood oxygen dependent horizontal time series signal are removed.
It can be understood that, because the tester may affect the acquired brain image data due to the problem of magnetic field instability during the brain scanning process, the scanning instrument may affect the acquired brain image data at the initial scanning stage, so as to ensure the stability of longitudinal magnetization of the scanning instrument, the data at the first N time points in the time series signal may be removed, so as to avoid the influence of the magnetic field instability on the data, preferably, N may be 10, and the user may also set the value of N according to the actual situation, which is not limited herein.
And S102-3, correcting the scanning time of the functional brain image data corresponding to each residual time point.
It can be understood that, in the process of scanning the brain of the tester, the brain is divided into N layers at each scanning time point, the scanning process includes scanning the 1 st layer, the 3 rd layer, and … the nth layer, the 2 nd layer, and the 4 th layer … the N-1 th layer in sequence, that is, scanning the odd layers and then scanning the even layers, because the blood sample levels between adjacent layers are relatively close, in order to ensure the consistency between the blood sample level signals between adjacent layers, for the N layers of brain image data scanned at each time point, the scanning time corresponding to the brain image data of the remaining layers is shifted with the scanning time of the intermediate layer (the nth layer) as the reference, so that each layer is aligned to the same time origin, so as to avoid that the obtained blood oxygen level signal has a large volatility and affects the accuracy of the data.
And S102-4, taking the functional brain image data corresponding to the first time point as a reference, and performing head movement correction on the functional brain image data corresponding to the other time points.
It can be understood that, during the brain scanning process, the deviation between the scanned brain images may occur due to the body movement, the brain shaking, etc. of the tester, which affects the subsequent data processing, therefore, the head motion record of each tester in the resting state functional magnetic resonance scanning process can be obtained through alignment calibration, the head movement record can be used as a screening condition, partial brain images with larger deviation can be removed according to the screening condition so as to ensure the accuracy of data processing, in the embodiment of the present invention, the brain image screening condition may be that the maximum displacement of the tester on any plane is not more than 2 mm, the rotation angle is not more than 2 degrees, and the frame phase displacement is not more than 0.2 mm, and the screening condition that is not satisfied is eliminated, and the user may set the screening condition according to an actual application scenario, which is not limited herein.
S102-5, registering the structural brain image data of each tester into functional brain image data, and registering the junction-registered functional images into a standard space template.
It can be understood that, since the gray brain matter region is closely related to the behavior of the tester, the DARTEL segmentation method can be used to segment the gray brain matter, white matter and cerebrospinal fluid in the structural image, and then the segmented structural image is registered into the functional image through the linear transformation with 6 degrees of freedom, so that the gray brain matter in the structural image matches the gray brain matter region in the functional image, and data errors are avoided. To obtain a standard image processing template, after registration between the structural image and the functional image is completed, the brain image data may also be processed such that the voxel (similar to a pixel in the image) size is 3 × 3 × 3mm, after which the functional image is registered into a specific template. By registering the registered functional images to a specific template, for example a specific template in the MNI (montreal neurological institute, abbreviated MNI) space, the problem of differences between different tested brain structures can be overcome.
And S102-6, performing linear drift removal and low-frequency filtering on the registered functional brain image data.
It can be understood that, in order to adapt to changes such as temperature rise and the like occurring in the working process of a scanning instrument, all voxels in functional image data can be applied to perform linear drifting, and data errors caused in the processes of time alignment correction and space standardization are removed, meanwhile, a functional image signal after low-frequency filtering may reflect spontaneous neural activity, the frequency band of the low-frequency filtering is generally 0.01-0.1 Hz, and the influence of physiological noise such as heartbeat, respiration and the like can be avoided by performing signal filtering in the frequency band.
For the functional image data after the off-line drift and the low-frequency filtering, the influence of the whole brain time sequence signal, the brain gray matter signal, the brain white matter signal, the cerebrospinal fluid signal and the head movement parameter model on the time sequence signal corresponding to the functional image data can be reduced through the multivariate linear model.
Optionally, the time-layer correction, the head motion correction, the spatial normalization, the linear drift removal and the low-frequency filtering of the functional image are completed through the above steps, on the basis, in order to obtain an accurate DALFF value, an implementation manner of calculating the biometric indicators corresponding to all brain areas of each tester is given below, referring to fig. 3, where fig. 3 is a possible implementation manner of S103 provided by the embodiment of the present invention, and includes the following steps:
s103-1, obtaining a plurality of functional brain image sample data from the blood oxygen level dependent time signal sequence according to a preset time window function.
It is understood that the preset time window function may be hamming. Since the band-pass filtering frequency band is 0.01-0.1 Hz, according to the formula T ═ 1/f, it can be known that the time window size can be 10s to 180s, that is, for the blood oxygen level dependent time signal sequence, the blood oxygen level dependent time signal sequence can be intercepted every time T to obtain the brain image sample data, and the sliding time window can capture the dynamic characteristic change of the whole brain activity.
It will also be appreciated that in order to balance the rapid acquisition time series dynamics (shorter time window) and the reliable assessment of the correlation between resting blood oxygen dependent levels between brain regions (longer time window), the sliding time window may be 100s and the time window shift step may be 2 s.
S103-2, calculating biological characteristic indexes corresponding to all brain areas according to the sample data of the plurality of functional brain images.
It can be understood that each functional brain image sample data includes a plurality of brain regions, each brain region image may include a plurality of voxels, and the manner of the dynamic low-frequency oscillation amplitude values corresponding to all the brain regions may be: for each voxel corresponding to each brain region, converting the corresponding time signal sequence into a frequency domain spectrum by using a fast Fourier transform, calculating the square root of the frequency domain spectrum, then averaging the mean value in the low frequency band interval (0.01-0.1 Hz) to be used as ALFF of each voxel, then calculating the mean value and standard deviation of the ALFF value of each voxel, quantitatively describing the dynamic characteristic (dALFF) of the ALFF along with the scanning time by using the standard deviation, and in order to reduce the influence of the whole brain activity variation of the subject, calculating the z-fraction of each dALFF, and performing full width at half maximum Gaussian smoothing kernel smoothing processing of 4mm on the z-fraction brain map of each dALFF of the subject to reduce the error caused by time correction and head motion correction.
Optionally, for how to determine the performance of the suicide risk prediction model, a possible implementation is given below, referring to fig. 4, where fig. 4 provides a possible implementation of S105 for the embodiment of the present invention:
and S105-1, obtaining risk assessment predicted values corresponding to a plurality of testers.
And S105-2, calculating a correlation coefficient between the risk assessment predicted values corresponding to the multiple testers and a preset risk assessment observation value.
In the embodiment of the present invention, in each training, the correlation coefficient of the observed value and the predicted value of the beck suicide scale score of the patient with suicidal unthreaded bipolar disorder and the patient with suicidal idetic bipolar disorder in the first test subject is calculated, and in a possible implementation, the correlation coefficient can be obtained by the following calculation method:
wherein, ynIs the observed value of the nth tester, f (x)n) Predicted value of the nth tester, μyAnd mufThe observation value average value and the prediction value average value of the n testers are respectively represented, and the regression performance of the model can be evaluated through the correlation between the prediction value and the observation value.
And S105-2, when the correlation coefficient meets the set condition, obtaining a trained suicide risk prediction model.
In an embodiment of the present invention, the setting conditions include: -1 < CORR < 1, characterization represents the strength of the linear correlation between observed and predicted values, with near 0 representing a weak or even no correlation, near 1 representing a positive correlation, and near-1 representing a negative correlation. When the CORR value is less than 0, the model performance is poor, which means that larger observations tend to yield less predicted values than smaller observations.
It should be noted that, in an embodiment, the suicide risk prediction model may be a pre-trained model, and when suicide risk prediction needs to be performed on a person to be tested, the suicide risk prediction model can be directly put into use; in another embodiment, the suicide risk model may be trained before suicide risk prediction is performed on a to-be-tested person, and after training is completed, the suicide risk model may be used to perform suicide risk prediction on the to-be-tested person, and a user may select the suicide risk model according to an actual scene, which is not limited herein.
Based on the suicide risk prediction model provided in the above embodiment, an embodiment of the present invention provides a suicide risk prediction method, referring to fig. 5, and fig. 5 is a schematic flow chart of a magnetic resonance image-based bipolar disorder suicide risk prediction method provided in an embodiment of the present invention, including the following steps:
s501, acquiring image data of a characteristic brain area of a to-be-tested object; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state.
And S502, inputting the image data into the trained suicide risk prediction model, and determining a risk assessment value corresponding to the biological characteristic index.
And S503, when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested.
In the embodiment of the present invention, the preset threshold is 5, when the object to be tested is a healthy person, the corresponding risk assessment value is 5, when the object to be tested is a bipolar disorder patient, the obtained risk assessment value is greater than 5, and the greater the risk assessment value, the higher the risk of suicide.
It can be understood that the suicide risk observation value is obtained from a questionnaire for quantifying and evaluating suicide, the answers of the scale are 3, the score is higher, the suicide idea is stronger, the objects to be tested complete the first 5 questions first, if the answers of the 4 th and 5 th items are 'none', the suicide idea is considered to be absent, the scale is completed, and the preset threshold value of 5 is returned; if any 1 of the 4 th or 5 th items selects the answer "weak" or "medium to strong", then the remaining items continue to be completed, the corresponding suicide risk assessment value is greater than 5, and the greater the risk assessment value, the higher the risk of suicide.
The suicide risk prediction method provided by the embodiment of the invention comprises the following steps: acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index, and a characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state; inputting the image data into a trained suicide risk prediction model, and determining a risk assessment value corresponding to the biological characteristic index; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested. The invention can avoid the uncertainty and misdiagnosis rate caused by depending on the professional knowledge and subjective judgment of doctors, and can provide theoretical support for the diagnosis of patients.
Optionally, in order to obtain the biometric indicator of the characteristic brain region of the subject to be tested, the following provides one possible implementation manner, wherein one possible implementation manner of step S501 may be:
s501-1, collecting brain image data of a to-be-tested object; the brain image data includes functional brain image data and structural brain image data.
S501-2, preprocessing the functional brain image data according to the structural brain image data.
S501-3, calculating biological characteristic indexes corresponding to all brain areas of the object to be tested according to the preprocessed functional brain image data.
S501-4, acquiring image data of a brain area corresponding to the biological characteristic index different from the preset biological characteristic index as image data of the characteristic brain area.
It is understood that the implementation manners of the above step S501-2 to the step S501-4 may be similar to the implementation manners of the sub-steps of the step 102 and the sub-steps of the step 103 in the above embodiments, and are not described herein again.
Optionally, in an implementation, before obtaining the biological characteristic index of the characteristic brain region of the object to be tested, model training may be performed by collecting brain image data of a plurality of testers to obtain a trained suicide risk prediction model, and a training process may refer to fig. 1, fig. 2, fig. 3, and fig. 4, which is not described herein again.
In order to achieve the above-mentioned embodiments and achieve the corresponding technical effects, a suicide risk prediction apparatus is provided below, referring to fig. 6, where fig. 6 is a functional block diagram of a magnetic resonance image-based dual-phase disorder suicide risk prediction apparatus according to an embodiment of the present invention, and the apparatus 60 includes: an acquisition module 601 and a determination module 602.
An obtaining module 601, configured to obtain image data of a characteristic brain region of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state.
A determining module 602, configured to input the image data into a trained suicide risk prediction model, and determine a risk assessment value corresponding to the biological characteristic index; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested.
It is to be appreciated that the obtaining module 601 and the determining module 602 may be utilized to perform steps 501-503 to achieve corresponding technical effects.
Optionally, the obtaining module 601 may also be configured to perform steps S501-2 to S501-4 to achieve corresponding technical effects.
An embodiment of the present invention further provides a training apparatus for a bipolar disorder suicide risk prediction model based on a magnetic resonance image, referring to fig. 7, fig. 7 is a functional block diagram of the training apparatus for a bipolar disorder suicide risk prediction model based on a magnetic resonance image according to an embodiment of the present invention, where the apparatus 70 includes: an acquisition module 701, a processing module 702, a calculation module 703, a determination module 704, a construction module 705 and a training module 706.
An acquisition module 701, configured to acquire brain image data corresponding to multiple testers; the brain image data comprises functional brain image data and structural brain image data; the multiple testers are divided into a first tester and a second tester; the first tester has a suicide risk; the second tester had no suicide risk.
A processing module 702, configured to pre-process the functional brain image data according to the structural brain image data of each tester.
The calculating module 703 calculates the biological characteristic indexes corresponding to all brain regions of each tester according to the preprocessed functional brain image data of each tester.
A determining module 704, configured to determine a characteristic brain region corresponding to the first tester according to a comparison result between the biological characteristic indicators corresponding to all brain regions of the first tester and the biological characteristic indicators corresponding to all brain regions of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject.
The building module 705 is configured to build a regression model between the biological characteristic index and the risk assessment value as an initial suicide risk prediction model based on the biological characteristic index corresponding to the characteristic brain region and a preset risk assessment value.
And the training module 706 is configured to train the initial suicide risk prediction model according to the biological characteristic index corresponding to each tester until a trained suicide risk prediction model is obtained.
It is to be appreciated that the acquisition module 701, the processing module 702, the calculation module 703, the determination module 704, the construction module 705 and the training module 706 may be used to perform the steps 101 to 106 to achieve the corresponding technical effect.
Optionally, the processing module 702 may be further configured to execute the steps shown in S102-1 to S102-6, the computing module 703 may be further configured to execute the steps shown in S103-1 to S103-2, and the training module 706 may be further configured to execute the steps shown in S105-1 to S105-3 to achieve the corresponding technical effects.
An embodiment of the present invention further provides an electronic device, as shown in fig. 8, and fig. 8 is a block diagram illustrating a structure of the electronic device according to the embodiment of the present invention. The electronic device 80 includes a communication interface 801, a processor 802, and a memory 803. The processor 802, memory 803, and communication interface 801 are electrically connected to one another, directly or indirectly, to enable the transfer or interaction of data. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 803 may be used for storing software programs and modules, such as program instructions/modules corresponding to the suicide risk prediction method and the training method of the suicide risk prediction model provided by the embodiment of the present invention, and the processor 802 executes various functional applications and data processing by executing the software programs and modules stored in the memory 803. The communication interface 801 may be used for communicating signaling or data with other node devices. The electronic device 80 may have a plurality of communication interfaces 801 in the present invention.
The memory 803 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a programmable read-only memory (PROM), an erasable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), and the like.
The processor 802 may be an integrated circuit chip having signal processing capabilities. The processor may be a general-purpose processor including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc.
Alternatively, the respective modules of the magnetic resonance image-based bipolar disorder suicide risk prediction apparatus 60 and the training apparatus 70 for the magnetic resonance image-based bipolar disorder suicide risk prediction model may be stored in the form of software or Firmware (Firmware) in the memory 803 of the electronic device 80, and may be executed by the processor 802. Meanwhile, data, codes of programs, and the like required to execute the above modules may be stored in the memory 803.
An embodiment of the present invention provides a storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the method according to any one of the foregoing embodiments. The computer readable storage medium may be, but is not limited to, various media that can store program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a PROM, an EPROM, an EEPROM, a magnetic or optical disk, etc.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (8)
1. A bipolar disorder suicide risk prediction method based on magnetic resonance imaging is characterized by comprising the following steps:
acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state;
the step of obtaining image data of a characteristic brain region of a subject to be tested comprises:
acquiring brain image data of the object to be tested; the brain image data comprises functional brain image data and structural brain image data;
preprocessing the functional brain image data according to the structural brain image data;
the pretreatment comprises the following steps: segmenting the gray matter, white matter and cerebrospinal fluid in the structural brain image according to a DARTEL segmentation method, registering the segmented structural brain image into the functional brain image through linear transformation to enable the gray matter in the structural brain image to be matched with the gray matter region in the functional brain image, and registering the registered functional brain image to a specific template;
calculating biological characteristic indexes corresponding to all brain areas of the object to be tested according to the preprocessed functional brain image data, wherein the functional brain image data is obtained from a time signal sequence according to a preset time window function, and the time window function captures fluctuation of functional activities of all brain areas; the method comprises the following steps: for each voxel corresponding to each brain region, converting the time signal sequence corresponding to the voxel into a frequency domain spectrum by using fast Fourier transform, determining the standard deviation of ALFF of all the voxels according to the square root of the frequency domain spectrum and a set frequency band interval, and taking the z-fraction value of the standard deviation of ALFF as a corresponding biological characteristic index of the brain region; acquiring brain image data of a brain area corresponding to a biological characteristic index different from a preset biological characteristic index as image data of the characteristic brain area;
inputting the image data into a trained suicide risk prediction model, and determining a risk assessment value corresponding to the biological characteristic index; the trained suicide risk prediction model is obtained by the following method: acquiring brain image data corresponding to a plurality of testers; the brain image data comprises functional brain image data and structural brain image data; the plurality of testers are divided into a first tester and a second tester; the first tester is at risk of suicide; the second tester is not at risk of suicide; preprocessing the functional brain image data according to the structural brain image data of each tester; calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester; determining the characteristic brain area corresponding to the first tester according to the comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject; constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and using the regression model as an initial suicide risk prediction model; training an initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained;
and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested.
2. The method of claim 1, wherein the step of training an initial suicide risk prediction model according to the corresponding biometric indicator of each tester until obtaining the trained suicide risk prediction model comprises:
obtaining risk assessment predicted values corresponding to the multiple testers;
calculating correlation coefficients between the risk assessment predicted values corresponding to the multiple testers and preset risk assessment observed values;
when the correlation coefficient meets a set condition, obtaining the trained suicide risk prediction model; wherein the setting condition is that the correlation coefficient is within a preset range.
3. A training method of a bipolar disorder suicide risk prediction model based on magnetic resonance images is characterized by comprising the following steps:
acquiring brain image data corresponding to a plurality of testers; the image data comprises functional brain image data and structural brain image data; the plurality of testers are divided into a first tester and a second tester; the first tester is at risk of suicide; the second tester is not at risk of suicide;
preprocessing the functional brain image data according to the structural brain image data of each tester;
the pretreatment comprises the following steps: segmenting the gray matter, white matter and cerebrospinal fluid in the structural brain image according to a DARTEL segmentation method, registering the segmented structural brain image into the functional brain image through linear transformation to enable the gray matter in the structural brain image to be matched with the gray matter region in the functional brain image, and registering the registered functional brain image to a specific template;
calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester; the method comprises the following steps: for each voxel corresponding to each brain region, converting the time signal sequence corresponding to the voxel into a frequency domain spectrum by using fast Fourier transform, determining the standard deviation of ALFF of all the voxels according to the square root of the frequency domain spectrum and a set frequency band interval, and taking the standard deviation of ALFF as a corresponding biological characteristic index of the brain region;
determining the characteristic brain area corresponding to the first tester according to the comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject;
constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and using the regression model as an initial suicide risk prediction model;
and training the initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained.
4. The method of claim 3, wherein the step of training an initial suicide risk prediction model according to the corresponding biometric indicator of each tester until obtaining the trained suicide risk prediction model comprises:
obtaining risk assessment predicted values corresponding to the multiple testers;
calculating correlation coefficients between the risk assessment predicted values corresponding to the multiple testers and preset risk assessment observed values;
when the correlation coefficient meets a set condition, obtaining the trained suicide risk prediction model; wherein the setting condition is that the correlation coefficient is within a preset range.
5. A bipolar disorder suicide risk prediction device based on magnetic resonance imaging is characterized by comprising: the device comprises an acquisition module and a determination module;
the acquisition module is used for acquiring image data of a characteristic brain area of a subject to be tested; the image data comprises a biological characteristic index; the characteristic brain area represents a corresponding brain area when the biological characteristic index is different from a preset biological characteristic index; the biological characteristic index represents the fluctuation intensity of the brain region functional activity in the resting state; the acquisition module is specifically configured to: acquiring brain image data of the object to be tested; the brain image data comprises functional brain image data and structural brain image data;
preprocessing the functional brain image data according to the structural brain image data;
the pretreatment comprises the following steps: segmenting the gray matter, white matter and cerebrospinal fluid in the structural brain image according to a DARTEL segmentation method, registering the segmented structural brain image into the functional brain image through linear transformation to enable the gray matter in the structural brain image to be matched with the gray matter region in the functional brain image, and registering the registered functional brain image to a specific template;
calculating biological characteristic indexes corresponding to all brain areas of the object to be tested according to the preprocessed functional brain image data; the method comprises the following steps: for each voxel corresponding to each brain region, converting the time signal sequence corresponding to the voxel into a frequency domain spectrum by using fast Fourier transform, determining the standard deviation of ALFF of all the voxels according to the square root of the frequency domain spectrum and a set frequency band interval, and taking the standard deviation of ALFF as a corresponding biological characteristic index of the brain region;
acquiring brain image data of a brain area corresponding to a biological characteristic index different from a preset biological characteristic index as image data of the characteristic brain area;
the determining module is used for inputting the image data into a trained suicide risk prediction model and determining a risk assessment value corresponding to the biological characteristic index; the trained suicide risk prediction model is obtained by the following method: acquiring brain image data corresponding to a plurality of testers; the brain image data comprises functional brain image data and structural brain image data; the plurality of testers are divided into a first tester and a second tester; the first tester is at risk of suicide; the second tester is not at risk of suicide; preprocessing the functional brain image data according to the structural brain image data of each tester; calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester; determining the characteristic brain area corresponding to the first tester according to the comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject; constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and using the regression model as an initial suicide risk prediction model; training an initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained; and when the risk assessment value is larger than a preset threshold value, determining that the suicide risk exists in the object to be tested.
6. A training device of a bipolar disorder suicide risk prediction model based on magnetic resonance images is characterized by comprising: the device comprises an acquisition module, a processing module, a calculation module, a determination module, a construction module and a training module;
the acquisition module is used for acquiring brain image data corresponding to a plurality of testers; the image data comprises functional brain image data and structural brain image data; the plurality of testers are divided into a first tester and a second tester; the first tester is at risk of suicide; the second tester is not at risk of suicide;
the processing module is used for preprocessing the functional brain image data according to the structural brain image data of each tester; the pretreatment comprises the following steps: segmenting the gray matter, white matter and cerebrospinal fluid in the structural brain image according to a DARTEL segmentation method, registering the segmented structural brain image into the functional brain image through linear transformation to enable the gray matter in the structural brain image to be matched with the gray matter region in the functional brain image, and registering the registered functional brain image to a specific template;
the calculation module is used for calculating biological characteristic indexes corresponding to all brain areas of each tester according to the preprocessed functional brain image data of each tester; the calculation module is specifically configured to: for each voxel corresponding to each brain region, converting the time signal sequence corresponding to the voxel into a frequency domain spectrum by using fast Fourier transform, determining the standard deviation of ALFF of all the voxels according to the square root of the frequency domain spectrum and a set frequency band interval, and taking the standard deviation of ALFF as a corresponding biological characteristic index of the brain region;
the determining module is used for determining the characteristic brain area corresponding to the first tester according to the comparison result between the biological characteristic indexes corresponding to all brain areas of the first tester and the biological characteristic indexes corresponding to all brain areas of the second tester; the characteristic brain region is one or more of the total brain regions of the first test subject;
the construction module is used for constructing a regression model between the biological characteristic indexes and the risk assessment value based on the biological characteristic indexes corresponding to the characteristic brain areas and the preset risk assessment value, and the regression model is used as an initial suicide risk prediction model;
and the training module is used for training the initial suicide risk prediction model according to the biological characteristic indexes corresponding to each tester until the trained suicide risk prediction model is obtained.
7. An electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor to implement the method of any one of claims 1-2 or to implement the training method of any one of claims 3-4.
8. A storage medium having stored thereon a computer program, characterized in that the computer program, when being executed by a processor, implements the method according to any of the claims 1-2 or implements the training method according to any of the claims 3-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010793478.XA CN111739647B (en) | 2020-08-07 | 2020-08-07 | Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010793478.XA CN111739647B (en) | 2020-08-07 | 2020-08-07 | Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111739647A CN111739647A (en) | 2020-10-02 |
| CN111739647B true CN111739647B (en) | 2021-08-03 |
Family
ID=72658312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010793478.XA Active CN111739647B (en) | 2020-08-07 | 2020-08-07 | Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111739647B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112690777B (en) * | 2021-01-13 | 2023-03-21 | 常州大学 | Neurological disorder diagnosis system based on state transition dynamic brain network algorithm |
| CN113538464A (en) * | 2021-07-22 | 2021-10-22 | 脑玺(苏州)智能科技有限公司 | Brain image segmentation model training method, segmentation method and device |
| CN113947580B (en) * | 2021-10-18 | 2022-07-19 | 上海市精神卫生中心（上海市心理咨询培训中心） | System for realizing auxiliary diagnosis processing of neuroimaging aiming at mental diseases |
| CN115376694B (en) * | 2022-10-24 | 2023-02-28 | 佛山科学技术学院 | Multi-mode self-adaptive assessment system and method for limb dysfunction |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512454B (en) * | 2015-07-28 | 2018-12-18 | 东南大学 | Suicide of Depression Patients risk objective assessment system based on function nuclear magnetic resonance |
| CN111462773B (en) * | 2020-03-26 | 2023-05-05 | 心图熵动科技(苏州)有限责任公司 | Suicide risk prediction model generation method and prediction system |
-
2020
- 2020-08-07 CN CN202010793478.XA patent/CN111739647B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| 晚发性抑郁患者随访一年的认知功能变化及静息态fMRI研究;许洪敏;《放射学实践》;20190228;130-135页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111739647A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111739647B (en) | Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device | |
| Martin et al. | Translating state-of-the-art spinal cord MRI techniques to clinical use: A systematic review of clinical studies utilizing DTI, MT, MWF, MRS, and fMRI | |
| US11915825B2 (en) | Systems and methods of analyte measurement analysis | |
| CN109276241B (en) | Pressure identification method and equipment | |
| CN110097968B (en) | Baby brain age prediction method and system based on resting state functional magnetic resonance image | |
| CN113712569B (en) | High-frequency QRS wave group data analysis method and device | |
| CN109997200A (en) | Stroke diagnosis and method for prediction of prognosis and system | |
| Chen et al. | Computerized wrist pulse signal diagnosis using modified auto-regressive models | |
| CN111631715B (en) | Method for predicting early cognitive function decline of Alzheimer's disease | |
| JP2009541863A (en) | System and method for analyzing and evaluating dementia and dementia-type disorders | |
| CN110969614B (en) | Brain age prediction method and system based on three-dimensional convolutional neural network | |
| Kinany et al. | Towards reliable spinal cord fMRI: Assessment of common imaging protocols | |
| JP2016163698A (en) | Mental condition determination method and mental condition determination program | |
| CN113160994A (en) | Construction method, prediction system, device and storage medium for noninvasive screening of non-alcoholic steatohepatitis model | |
| KR101388546B1 (en) | System and method for assessing brain dyfunction using functional magnetic resonance imaging | |
| CN112990266B (en) | Method, device, equipment and storage medium for processing multi-mode brain image data | |
| CN113456033A (en) | Physiological index characteristic value data processing method and system and computer equipment | |
| CN110801228B (en) | Brain effect connection measurement method based on neural network prediction | |
| CN115337000B (en) | Machine learning method for evaluating brain aging caused by diseases based on brain structure images | |
| Lombardi et al. | Detecting sepsis from photoplethysmography: strategies for dataset preparation | |
| EP3655921B1 (en) | Method and computer program for estimating shape parameters of the fovea by optical coherence tomography | |
| Hodzic et al. | Analysis of inter-system variability of systolic and diastolic intraventricular pressure gradients derived from color Doppler M-mode echocardiography | |
| Molchanova et al. | Novel structural-scale uncertainty measures and error retention curves: application to multiple sclerosis | |
| CN108720870A (en) | A kind of fatty liver detecting system based on ultrasonic attenuation coefficient | |
| US20220378380A1 (en) | Method and system for determining cardiovascular parameters |
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 |