A method for quantitatively evaluating the treatment effect of Alzheimer's disease based on nuclear magnetic resonance

Abstract

Resting-state magnetic resonance imaging (fMRI) technology records brain activity by measuring the global blood oxygen saturation of the subject's brain, providing high spatial resolution data. Although the temporal resolution is lower, in elderly individuals at rest, the introduction of time series data still allows for good observation of the hemodynamic direction of individual pixels, making it one of the important evaluation indicators for intervention and treatment of brain diseases. The neural circuit features constructed based on a single brain region towards the whole brain are also an important evaluation method for fMRI data, namely, the quantification of the physiological interaction of the brain region towards the whole brain in a certain state. This invention constructs a method for quantifying the treatment effect of Alzheimer's disease (AD) based on resting-state magnetic resonance imaging (fMRI) technology. This method mainly addresses: (1) the difficulty in quantifying the rehabilitation effect during the treatment of AD patients; and (2) the quantification of the rehabilitation effect of different treatment modes during the treatment of AD patients.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of signal processing, and particularly relates to an AD disease treatment effect quantitative evaluation method, which is suitable for quantitative evaluation of AD clinical patient rehabilitation and treatment. BACKGROUND

[0002] AD is a common senile disease in which patients show memory decline as the highlight, cognitive function reduction, and emotional personality change. With the deepening of the disease, the morbidity and mortality of the patient group will increase. At present, the diagnosis of the disease has mature technology, and precise diagnosis can be obtained in the early stage of the disease through various means, and certain means can be used to intervene and treat the disease.

[0003] But the brain itself is a very complex nervous organ, which contains complex neural tissues that work together in parallel. When the disease occurs, the function of the lesion area decreases, and other areas will compensate, so that the functional regression of the lesion is masked, and after intervention and treatment, the recovery of the lesion may cause the function of other compensatory brain areas to decrease, which will interfere with data analysis. Therefore, single-dimensional data analysis cannot meet the rehabilitation quantification of the brain neural disease group, especially the elderly group whose brain is in regression during treatment.

[0004] The fMRI imaging technology can provide high spatial resolution data by measuring the global blood oxygen saturation of the subject's brain to describe the brain activity. Although the time resolution is low, in the resting state of the elderly group, the time sequence can still better observe the blood flow dynamics of a single pixel, which is one of the important evaluation indicators for brain disease intervention and treatment. The neural circuit features based on single brain region facing the whole brain are also an important evaluation method of fMRI data, that is, the physiological interaction of the brain region facing the whole brain in a certain state is quantified.

[0005] In recent years, the cerebellum is considered to be an important compensatory regulatory center in the development of AD disease. When the brain appears regressive changes due to AD disease, the cerebellum will appear compensatory regulation, and the interaction of most brain regions will be enhanced, which can be significantly observed in healthy control group and AD disease group. If the disease symptoms are relieved and the brain function of the lesion is restored, the compensatory regulation of the cerebellum will be weakened. SUMMARY

[0006] In order to overcome the deficiencies in the prior art, the present application constructs a method for quantitatively evaluating the treatment effect based on resting state nuclear magnetic resonance (fMRI) technology during the treatment of Alzheimer's disease (AD), which mainly solves the problems of (1) the difficulty in quantifying the rehabilitation effect of AD patients during treatment; (2) the quantification of the rehabilitation effect of different treatment modes for AD patients during treatment. Specifically, it includes:

[0007] Step 1: Constructing fMRI data

[0008] (1.1) Divide the data into AD group and healthy control group, extract resting state fMRI data, and extract pre-treatment and post-treatment data for the AD group;

[0009] (1.2) Preprocess the data based on the matlab toolkit spm;

[0010] Step 2: Cerebellar neural circuit evaluation

[0011] (2.1) Use the matlab toolkit dpabi to construct a neural circuit feature with the cerebellum as the seed node;

[0012] (2.2) Select the 10 brain regions with the largest difference in neural circuit between the AD group before treatment and the healthy control group, and reorganize the numerical values of the 10 brain regions with the largest lesion specificity brain region vector values;

[0013] (2.3) Calculate the weight of each vector value for each group, and calculate the weight proportion of each brain region data in all data;

[0014] (2.4) Calculate the absolute difference between the 10 brain region data obtained in (2.2) and the healthy control group before and after treatment for the AD group;

[0015] (2.5) Calculate the difference between the absolute difference of the AD group after treatment relative to the healthy control group and the absolute difference of the AD group before treatment relative to the control group, and distribute the total score of 100 according to the weight obtained in (2.3);

[0016] (2.6) Sum the values distributed according to the weight to obtain evaluation value 1;

[0017] Step 3: Cerebral blood flow dynamic evaluation

[0018] (3.1) Introduce time series to construct a single-pixel-based blood flow dynamics function;

[0019] (3.2) Calculate the time series coverage area, select the top ten area values in the healthy control group, and locate the area with the largest specific blood flow direction;

[0020] (3.3) According to the area located in (3.2), arrange the data of the AD group before and after treatment;

[0021] (3.4) Calculate the first derivative of the ten blood flow functions of the subjects in the healthy control group, normalize it, and obtain the weight distribution;

[0022] (3.5) According to the weight obtained in step (3.4), calculate the data score of the hemodynamics function in the first derivative of the AD group before and after treatment according to the method of steps (2.4) and (2.5), and obtain evaluation value 2;

[0023] Step 4: Data extreme evaluation

[0024] (4.1) Introduce time series, and construct a wide window with a width of 5 sampling time points;

[0025] (4.2) Use the wide window to perform windowed data superposition on the data, construct short-time fusion data, and calculate the window number average of the data;

[0026] (4.3) Select the maximum and minimum of 4 values in the healthy control group respectively and locate the value position, and construct the maximum value group and the minimum value group;

[0027] (4.4) According to the value positioning in (4.3), construct the maximum value group and the minimum value group of the AD group before and after treatment;

[0028] (4.5) Calculate the normalized weight distribution of the maximum value group and the minimum value group in the healthy control group;

[0029] (4.6) According to the weight obtained in step (4.5), calculate the data score of the maximum value and the minimum value of the brain activity of the AD group before and after treatment according to the method of steps (2.4) and (2.5), and obtain evaluation value 3;

[0030] Step 5: Perform treatment quantitative evaluation

[0031] Fusion the scores of evaluation values 1, 2 and 3 with a weight of 4:3:3 to obtain the treatment effect score of the subject.

[0032] The beneficial effects of the present application are:

[0033] The present application provides a quantitative standard for treatment effect in the clinical treatment of AD disease, and brings certain guidance and auxiliary effect for accurate treatment of AD disease. DETAILED DESCRIPTION

[0034] The following description of specific embodiments is provided to better understand the present application. It should be noted that the following description is merely illustrative of the present application and is not intended to limit the present application.

[0035] The present application is a treatment effect quantitative evaluation method for AD group, which constructs a multi-dimensional data fusion evaluation method based on resting state fMRI data. The specific steps are as follows.

[0036] Step 1: Constructing fMRI data of two types of people

[0037] (1.1) Screening of clinical patients, AD group and HC group, selecting people aged 60-75 years old, and resting state fMRI data of people with biochemical indicators meeting AD patients, wherein the data before and after treatment of AD group is extracted;

[0038] (1.2) Data preprocessing based on matlab toolkit spm;

[0039] Step 2: Cerebellar neural circuit evaluation

[0040] (2.1) Constructing neural circuit features with cerebellum as seed node using matlab toolkit dpabi;

[0041] (2.2) Screening the 10 brain regions with the largest difference in neural circuit between AD group before treatment and healthy control group, and screening and reorganizing the numerical values of the brain regions with the largest lesion specificity vector values;

[0042] (2.3) Numerical weight calculation for each group vector, calculating the weight proportion of each brain region data in all data;

[0043] (2.4) Calculating the absolute difference between the 10 brain region data obtained in (2.2) and the healthy control group before and after treatment of AD group;

[0044] (2.5) Calculate the difference between the absolute difference of AD group after treatment relative to healthy control group and the absolute difference of AD group before treatment relative to control group, and distribute the total score of 100 according to the weight obtained in (2.3);

[0045] (2.6) Summing up the numerical values of the test objects according to the weight distribution to obtain evaluation value 1;

[0046] Step 3: Cerebral blood flow dynamic evaluation

[0047] (3.1) Introducing time series, constructing single-pixel-based blood flow dynamics function;

[0048] (3.2) Calculate the function to get the time series coverage area, and select the top ten area values in the healthy control group and reorganize them as the areas with the largest specificity blood flow direction;

[0049] (3.3) According to the pixel region positioned in (3.2), the data before and after AD group treatment is sorted out;

[0050] (3.4) The first order derivative of the ten blood flow dynamic functions of each subject in the healthy control group is calculated, and the normalized weight distribution is assigned;

[0051] (3.5) According to the weight obtained in step (3.4), the data score of the subject's hemodynamics under the first order dynamics before and after treatment is calculated according to the method of steps (2.4) and (2.5), and the evaluation value 2 is obtained;

[0052] Step 4: Data extreme evaluation

[0053] (4.1) Introduce time series, and construct a wide window with a width of 5 sampling time points;

[0054] (4.2) The windowed data superposition is performed on the data using the wide window, the short-time fusion data is constructed, and the window number average is calculated;

[0055] (4.3) The maximum and minimum of 4 values in the healthy control group are selected respectively and the value position is located, and the maximum value group and the minimum value group are constructed;

[0056] (4.4) According to the value positioning in (4.3), the maximum value group and the minimum value group of the AD group before and after treatment are constructed;

[0057] (4.5) The normalized weight distribution of the maximum value group and the minimum value group in the healthy control group is calculated;

[0058] (4.6) According to the weight obtained in step (4.5), the data score of the maximum value and the minimum value of the subject's brain activity before and after treatment is calculated according to the method of steps (2.4) and (2.5), and the evaluation value 3 is obtained;

[0059] Step 5: Perform treatment quantitative evaluation

[0060] The scores of evaluation values 1, 2 and 3 are fused with a weight of 4:3:3 to obtain the subject treatment effect score.

[0061] The above has introduced the present application in detail, but the description of the specific embodiment is only used to explain the method of the present application and its core idea, so as to facilitate the technical personnel in the technical field to understand the present application, but it should be clear that the present application is not limited to the scope of the specific embodiment, for the ordinary technical personnel in the technical field, as long as various changes are within the spirit and scope of the present application limited and determined by the appended claims, these changes are obvious, all the inventions and creations using the concept of the present application are within the scope of protection.

Claims

1. A method for quantitatively evaluating the treatment efficacy of Alzheimer's disease based on nuclear magnetic resonance imaging, characterized in that: Comprising the following steps: Step 1: Constructing fMRI data (1.1) Divide the data into two groups of AD group and healthy control group, extract resting state fMRI data, and extract pre-treatment and post-treatment data for the AD group; (1.2) Preprocess the data based on the matlab toolkit spm; Step 2: Cerebellar neural circuit evaluation (2.1) Use the matlab toolkit dpabi to construct a neural circuit feature with the cerebellum as the seed node; (2.2) Screen the 10 brain regions with the largest difference in neural circuit between the AD group before treatment and the healthy control group, and reorganize the numerical values into a brain region vector with the largest lesion specificity; (2.3) Calculate the weight of each group vector value, and calculate the weight proportion of each brain region data in all data; (2.4) Calculate the absolute difference between the 10 brain region data obtained in (2.2) and the healthy control group before and after treatment for the AD group respectively; (2.5) Calculate the difference between the absolute difference of the AD group after treatment relative to the healthy control group and the absolute difference before treatment relative to the control group, and distribute according to the weight obtained in (2.3) with a total score of 100; (2.6) Sum up the values allocated according to the weight to obtain evaluation value 1; Step 3: Cerebral blood flow dynamic evaluation (3.1) Introduce time series to construct a single-pixel-based blood flow dynamics function; (3.2) Calculate the function to obtain the time series coverage area, and select the top ten area values in the healthy control group and locate them as the areas with the largest specific blood flow dynamic flow direction; (3.3) Organize the data of the AD group before and after treatment according to the areas located in (3.2); (3.4) Calculate the first-order derivative of the ten blood flow dynamics functions of the subjects in the healthy control group, normalize them, and obtain the weight distribution; (3.5) According to the weight obtained in step (3.4), calculate the data score of the AD group before and after treatment in the first-order derivative of the blood flow dynamics function according to the methods of steps (2.4) and (2.5), and obtain evaluation value 2; Step 4: Data extreme value evaluation (4.1) Introduce time series to construct a wide window with a width of 5 sampling time points; (4.2) Use the wide window to perform windowed data superposition on the data, construct short-time fusion data, and calculate the window number average of the data; (4.3) Select the maximum and minimum 4 values in the healthy control group respectively and locate the numerical values to construct the maximum value group and the minimum value group; (4.4) According to the numerical value positioning in (4.3), construct the maximum value group and the minimum value group of the AD group before and after treatment; (4.5) Calculate the normalized weight distribution of the maximum value group and the minimum value group in the healthy control group; (4.6) According to the weight obtained in step (4.5), calculate the data score of the maximum value and the minimum value of brain activity of the AD group before and after treatment according to the methods of steps (2.4) and (2.5), and obtain evaluation value 3; Step 5: Perform treatment quantitative evaluation Fusion the scores of evaluation values 1, 2, and 3 with a weight of 4:3:3 to obtain the subject treatment effect score.