CN-121549772-B - Method and device for generating information based on brain image data

CN121549772BCN 121549772 BCN121549772 BCN 121549772BCN-121549772-B

Abstract

The invention provides a method and a device for generating information based on brain image data, wherein the method comprises the steps of carrying out longitudinal brain image analysis on target individual brain image data acquired at multiple time points, carrying out regional division on the brain of each time point target individual and morphological measurement extraction of each region based on longitudinal brain image analysis results and cortex regional maps, determining deviation degree of morphological measurement values of each region of the brain of each time point target individual relative to brain morphological normal mode based on brain morphological normal mode to obtain individual brain morphological deviation vectors corresponding to each time point target individual, carrying out correlation calculation on the individual brain morphological deviation vectors corresponding to each time point target individual and statistical data of various mental diseases in a whole brain range to obtain cross-disease brain morphological similarity characteristics, inputting the cross-disease brain morphological similarity characteristics into a pre-trained machine learning model, and outputting classification information and/or clinical index prediction information of whether the target individual is a patient with schizophrenia.

Inventors

WEI YONGBIN
WANG YINGCHAN
WANG JIJUN
HUANG YANMENG
ZHAO SHUWAN
JIN QIANHUI
WANG ZEYAO
LIU DONGXU
LI LONGXIN

Assignees

北京邮电大学
上海市精神卫生中心（上海市心理咨询培训中心）

Dates

Publication Date: 20260512
Application Date: 20260121

Claims (10)

1. An apparatus for generating information based on brain image data, comprising: The extraction unit is configured to perform longitudinal brain image analysis on brain image data of a target individual brain acquired at least two time points, and perform regional division and morphological measurement value extraction of each region on the brain of the target individual at each time point based on a longitudinal brain image analysis result and a preset cortex regional map, wherein the at least two time points comprise a reference and a follow-up time point, the reference represents the time of acquiring the brain image data for the first time, and the follow-up time point represents the time of acquiring the brain image data for the first time; A determining unit configured to determine, based on a pre-established brain morphology norms, a degree of deviation of morphology measurement values of each region of the brain of the target individual at each time point with respect to the brain morphology norms, and obtain individual brain morphology deviation vectors corresponding to the target individual at each time point; The calculating unit is configured to calculate the correlation between the individual brain morphology deviation vector corresponding to the target individual at each time point and the pre-acquired statistical data of various mental diseases in the whole brain range, so as to obtain the cross-disease brain morphology similarity characteristics of the target individual at each time point; And the generation unit is configured to input the cross-disease brain morphology similarity characteristics of the target individual into a pre-trained machine learning model and output classification information and/or prediction information of clinical indexes of whether the target individual is a schizophrenic patient.
2. The apparatus of claim 1, wherein the machine learning model comprises a radial basis function based support vector machine classification model for outputting classification information of whether an individual is a schizophrenic patient based on cross-disease brain morphology similarity features of the individual.
3. The apparatus of claim 1 or 2, wherein the machine learning model comprises a LASSO regression prediction model for outputting prediction information of clinical indicators of individuals based on cross-disease brain morphology similarity features of the individuals, and wherein the LASSO regression prediction model is trained and regularized with nested cross-validation.
4. The apparatus according to claim 1, wherein the determining, based on a pre-established brain morphology norms, a degree of deviation of morphology measurement values of each region of the brain of the target individual with respect to the brain morphology norms at each time point, to obtain individual brain morphology deviation vectors corresponding to the target individual at each time point, includes: based on a pre-established brain morphology normal mode, calculating the z fraction of morphology measured values of each region of the brain of the target individual at each time point relative to the measured values of the brain morphology normal mode, and obtaining a z fraction vector corresponding to the target individual at each time point as an individual brain morphology deviation vector.
5. The apparatus according to claim 1, wherein the statistical data of the mental diseases in the whole brain range includes an effect vector of morphological measurement values, and wherein the correlation calculation of the individual brain morphology deviation vector corresponding to the target individual at each time point and the statistical data of the mental diseases in the whole brain range acquired in advance, to obtain the cross-disease brain morphology similarity characteristics of the target individual at each time point, comprises: Calculating pearson correlation between individual brain morphology deviation vectors corresponding to the target individual at each time point and effect vector of morphology measurement values of various mental diseases in the plurality of mental diseases, and obtaining correlation coefficients between the individual brain morphology deviation vectors and effect vector of morphology measurement values of various mental diseases; based on the obtained plurality of correlation coefficients, cross-disease brain morphology similarity characteristics of the target individual are determined.
6. The apparatus of claim 1, wherein the brain image data comprises T1 weighted magnetic resonance imaging data, and wherein the performing the region division and the morphological measurement extraction of each region of the brain of the target individual at each time point based on the longitudinal brain image analysis result and the preset cortical partition map comprises: Based on a longitudinal brain image analysis result and a preset cortical partition map, dividing the brain of the target individual at each time point into a plurality of cortical thickness areas and a plurality of subcortical volume areas; and extracting a measured value of the cortical thickness index of each cortical thickness region, and extracting a measured value of the subcortical volume index of each subcortical volume region as a morphological measured value of each region.
7. The apparatus according to claim 1, characterized in that the apparatus further comprises means for performing the steps of: and calculating the correlation between the similarity characteristics of the cross-disease brain morphology of a plurality of schizophrenic patients and the measured values of clinical indexes by adopting a partial least squares regression method.
8. The device of claim 1, wherein the plurality of mental disorders comprises at least two of 22q11.2 deficiency syndrome, attention deficit hyperactivity disorder, autism spectrum disorders, bipolar disorder, epilepsy, major depression, obsessive compulsive disorder, and schizophrenia.
9. A computing device comprising a processor, a memory, and computer programs/instructions stored on the memory, wherein the processor is configured to execute the computer programs/instructions, which when executed implement a method of generating information based on brain image data, the method comprising: Performing longitudinal brain image analysis on brain image data of a target individual brain acquired at least two time points, and performing region division and morphological measurement value extraction of each region on the brain of the target individual at each time point based on a longitudinal brain image analysis result and a preset cortical partition map, wherein the at least two time points comprise a reference and a follow-up time point, the reference represents the time of acquiring the brain image data for the first time, and the follow-up time point represents the time of acquiring the brain image data for the first time; determining deviation degree of morphological measurement values of each region of the brain of the target individual at each time point relative to the brain morphology normal mode based on a pre-established brain morphology normal mode, and obtaining individual brain morphology deviation vectors corresponding to the target individual at each time point; performing correlation calculation on the individual brain morphology deviation vector corresponding to the target individual at each time point and the pre-acquired statistical data of various mental diseases in the whole brain range to obtain cross-disease brain morphology similarity characteristics of the target individual at each time point; inputting the cross-disease brain morphology similarity characteristics of the target individual into a pre-trained machine learning model, and outputting classification information and/or prediction information of clinical indexes of whether the target individual is a schizophrenic patient.
10. A computer readable storage medium having stored thereon a computer program/instruction which when executed by a processor implements a method of generating information based on brain image data, the method comprising: Performing longitudinal brain image analysis on brain image data of a target individual brain acquired at least two time points, and performing region division and morphological measurement value extraction of each region on the brain of the target individual at each time point based on a longitudinal brain image analysis result and a preset cortical partition map, wherein the at least two time points comprise a reference and a follow-up time point, the reference represents the time of acquiring the brain image data for the first time, and the follow-up time point represents the time of acquiring the brain image data for the first time; determining deviation degree of morphological measurement values of each region of the brain of the target individual at each time point relative to the brain morphology normal mode based on a pre-established brain morphology normal mode, and obtaining individual brain morphology deviation vectors corresponding to the target individual at each time point; performing correlation calculation on the individual brain morphology deviation vector corresponding to the target individual at each time point and the pre-acquired statistical data of various mental diseases in the whole brain range to obtain cross-disease brain morphology similarity characteristics of the target individual at each time point; inputting the cross-disease brain morphology similarity characteristics of the target individual into a pre-trained machine learning model, and outputting classification information and/or prediction information of clinical indexes of whether the target individual is a schizophrenic patient.

Description

Method and device for generating information based on brain image data Technical Field The invention relates to the technical field of medical image processing, in particular to a method and a device for generating information based on brain image data. Background In the neuroimaging study of schizophrenia (Schizophrenia, SCZ), quantitative analysis of brain morphology based on brain image data such as magnetic resonance imaging (Magnetic Resonance Imaging, MRI), magnetic resonance imaging (Nuclear Magnetic Resonance Imaging, NMRI) has become an important approach to reveal disease-related brain structural abnormalities. Numerous studies have shown that the subcutaneous regions of the hippocampus, thalamus and amygdala of schizophrenic patients have a significant volume reduction, as well as a significant reduction in frontal, temporal and parietal gray matter volume and cortical thickness. Extensive, multi-site neuroimaging studies confirm these morphological changes, which are closely related to cognitive dysfunction, symptom severity and disease progression in patients. Therefore, it is important to assist diagnosis and treatment of schizophrenia by mining information related to schizophrenia through brain image data. Disclosure of Invention In view of the foregoing, embodiments of the present invention provide a method and apparatus for generating information based on brain image data, which obviates or mitigates one or more of the disadvantages of the related art. According to a first aspect, a method for generating information based on brain image data is provided, which comprises the steps of performing longitudinal brain image analysis on brain image data of a target individual brain acquired at least at two time points, performing regional division on the target individual brain at each time point and morphological measurement extraction of each region based on a longitudinal brain image analysis result and a preset cortex regional map, determining deviation degree of the morphological measurement of each region of the target individual brain at each time point relative to the brain morphological normal mode based on a pre-established brain morphological normal mode to obtain individual brain morphological deviation vectors corresponding to the target individual at each time point, performing correlation calculation on the individual brain morphological deviation vectors corresponding to the target individual at each time point and pre-acquired statistical data of various mental diseases in a whole brain range to obtain cross-disease brain morphological similarity characteristics of the target individual at each time point, inputting the cross-disease brain morphological similarity characteristics of the target individual into a pre-trained machine learning model, and outputting classification information of whether the target individual is a schizophrenic patient and/or not and/or prediction information of clinical indexes. According to a second aspect, there is provided an apparatus for generating information based on brain image data, comprising an extraction unit configured to perform longitudinal brain image analysis on brain image data of a target individual brain acquired at least at two time points, and to perform region division and morphological measurement extraction of each region on the brain of the target individual at each time point based on a longitudinal brain image analysis result and a preset cortex partition map, a determination unit configured to determine a degree of deviation of the morphological measurement of each region of the brain of the target individual at each time point relative to the brain morphology norms based on a pre-established brain morphology norms, to obtain individual brain morphology deviation vectors corresponding to the target individual at each time point, a calculation unit configured to perform correlation calculation on the individual brain morphology deviation vectors corresponding to the target individual at each time point and pre-acquired statistical data of a plurality of mental diseases in a whole brain range, to obtain cross-disease brain morphology similarity features of the target individual at each time point, and a generation unit configured to input pre-trained machine learning model of the cross-disease brain morphology features of the target individual, and output whether the target individual brain morphology similarity features are clinical classification indexes or not, and prediction indexes of the target individual or the clinical classification indexes. According to a third aspect, there is provided an apparatus for generating information based on brain image data, comprising a processor, a memory and computer programs/instructions stored on the memory, the processor being adapted to execute the computer programs/instructions, the apparatus implementing the steps of the method as defined in any one of the first aspects when the