CN115101204A - A model, equipment and storage medium for quantitative assessment of depression risk based on blood biochemical indicators - Google Patents

Abstract

The invention discloses a model, equipment and a storage medium for quantitatively evaluating depression risk based on blood biochemical indexes, which are used for respectively collecting blood biochemical test indexes and demographic variables of clinical depression patients and healthy people; the biochemical index (such as the biochemical index larger than the Cohen's d set value) and the demographic variable which can distinguish the depression patients and healthy people are used as independent variables, depression groups are used as dependent variables, a logistic regression model of the depression groups is established, depression risk values are calculated through the established logistic regression model of the depression groups, and risk quantitative evaluation is carried out on the depression patients according to the judgment threshold value. The method is beneficial to constructing standardized, objectively measurable depression risk assessment indexes and quantitative models in addition to scale testing and subjective judgment of doctors, assists the doctors in providing more accurate and objective depression assessment for patients, and overcomes the limitation of only using subjective assessment and expert assessment in the existing method.

Description

A model, device and method for quantitative assessment of depression risk based on blood biochemical indicators storage medium

technical field

The invention relates to the technical field of depression, in particular to a model, equipment and storage medium for quantitatively evaluating depression risk based on blood biochemical indexes.

Background technique

Depression is a serious mental illness with the highest incidence in the world, with a lifetime prevalence of 10-20%. In terms of the socioeconomic burden caused, depression currently ranks fourth among all diseases. It will improve to second place by 2020 (Holden, 2000). So far, clinical pathology is still the basis for the diagnosis of depression, but due to the complex and diverse clinical symptoms of depression and the lack of corresponding objective indicators, the diagnosis and differential diagnosis of the disease largely depend on psychiatrists Subjective experience and speculation, lack of objective and scientific basis for clinical diagnosis, depends on the doctor's subjective judgment.

The diagnosis of depression relies mainly on clinical phenomenology and lacks the utilization of biochemical parameters. Thus, the current dilemma is that the ten-year diagnostic agreement for major depressive disorder (MDD) is only 45.5%, while the long-term agreement in the seven-year cohort study for bipolar disorder (BD) The sex rate is also only 71.9%. Clinical application, translation is still limited due to the strong subjectivity in the clinical assessment process or the complexity of the specific tracking experimental design involved.

SUMMARY OF THE INVENTION

In order to solve the limitation of only using subjective evaluation and expert evaluation in the existing method, the present invention provides a model, equipment and storage medium for quantitatively evaluating the risk of depression based on blood biochemical indicators, which provide doctors or experts with More objective and accurate depression risk assessment.

The present invention adopts following technical scheme:

On the one hand, the present invention provides a model for quantitative assessment of depression risk based on blood biochemical indicators, and collects blood biochemical test indicators and demographic variables of clinical depression patients and healthy people respectively; Biochemical indicators and demographic variables were used as independent variables, and depression grouping was used as the dependent variable to establish a logistic regression model and judgment threshold for depression grouping; the depression risk value was calculated through the established logistic regression model for depression grouping; according to the judgment threshold Quantitative risk assessment of patients with depression.

Preferably, when selecting a biochemical index capable of distinguishing patients with depression and healthy people, by setting the Cohen's d value of the effect size, select the biochemical indexes of patients with depression and healthy people that are greater than the set value of the effect size Cohen's d.

Further preferably, the independent variable parameters in the established depression group regression model include: adrenocorticotropic hormone, brain-derived neurotrophic factor, interleukin 18 (IL-18), progesterone, serotonin biochemical test indicators and gender, age;

The logistic regression model of depression grouping was established as follows:

Y=1/(1+exp(-(-2.263* adrenocorticotropic hormone-1.113*brain-derived neurotrophic factor-0.13*interleukin 18+0.747*progesterone-0.451*serotonin+0.894*sex+0.002 *Age+21.221*Whether taking antidepressant or antimanic medication+20.498*Whether fasting before blood draw -2.441))).

Among the above parameters, gender: female=1, male=0; whether to take medicine: taking=1, not taking=0; fasting before blood drawing: eating before blood drawing=1, fasting before blood drawing=0.

Or further preferably, the independent variable parameters in the logistic regression model for grouping depression may also include: C-reactive protein (CRP), cortisol, testosterone and tumor necrosis factor alpha (TNF-α);

The logistic regression model of depression grouping can also be expressed by the following mathematical formula: Y=1/(1+exp(-(-3.632*corticotropin-1.1*brain-derived neurotrophic factor-0.117*interleukin 18 +0.835*progesterone-0.935*serotonin+0.671*C-reactive protein+0.38*cortisol+0.283*testosterone-0.006*tumor necrosis factor alpha+0.834*sex-0.01*age+20.137*are you taking antidepressants or Anti-manic drugs +19.641*Whether fasting before blood draw -2.124))).

Among the above parameters, gender: female=1, male=0; whether to take medicine: taking=1, not taking=0; fasting before blood drawing: eating before blood drawing=1, fasting before blood drawing=0.

Further, the collected biochemical test indicators of patients with clinical depression and healthy people may also include: corticosterone (adrenal ketone), thyroxine, gamma aminobutyric acid, dopamine, oxytocin, cortisone, neuropeptide Y, endogenous Cannabinoids, mature brain-derived neurotrophic factor, norepinephrine NE, interleukin 10 (IL-10), interleukin 12, interleukin 4, nerve growth factor and serotonin, the selection can distinguish Several biochemical indicators of depression patients and healthy people were used as independent variable parameters to establish a logistic regression model and judgment threshold for depression grouping.

The demographic variables also include marriage, children, education, occupation, income, and religious beliefs of the clinically depressed patients and healthy people.

In another aspect, the present invention provides an electronic device, the electronic device includes a processor and a memory, and the memory stores at least one piece of program, the at least one piece of program is loaded and executed by the processor to realize the A model for the quantitative assessment of depression risk based on blood biochemical indicators.

In another aspect, the present invention also provides a computer-readable storage medium, in which at least one program is stored, and the at least one program is loaded and executed by a processor to realize the quantitative evaluation based on blood biochemical indexes A model of depression risk.

In another aspect, the present invention also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the model for quantitatively assessing depression risk based on blood biochemical indicators.

The technical scheme of the present invention has the following advantages:

A. The present invention is based on the quantitative assessment model of depression risk based on blood biochemical test indicators and demographic variables, which helps to construct standardized, objectively measurable depression risk assessment indicators and quantitative models in addition to scale tests and subjective judgments of doctors, Assisting doctors to provide a more accurate and objective assessment of depression to patients overcomes the limitations of existing methods that only use subjective assessments and expert assessments.

B. The present invention establishes a depression grouping model after performing logistic regression on a plurality of selected blood biochemical indicators, and can establish a plurality of different models according to the number and type of the selected biochemical indicators, and input the tested blood biochemical test indicators respectively. and gender, age and other parameters, quickly get the calculated value, and the obtained value can quickly determine the risk probability of depression, which is helpful for the rapid screening of depression in the blood test population.

C. Since depression is a symptom cluster with multiple complex causes, including multiple subtypes, and even its production mechanism may be completely different, the symptomatic drugs required are also completely different. The current medical practice is to try different drugs by to observe the effect. Establishing a discriminant model based on blood biochemical test indexes by the method of the present invention helps to accurately distinguish the subtypes of depression, and can help doctors to select symptomatic drugs more quickly and accurately.

Description of drawings

In order to illustrate the specific embodiments of the present invention more clearly, the following will briefly introduce the accompanying drawings used in the specific embodiments. As far as technical personnel are concerned, other drawings can also be obtained based on these drawings without any creative effort.

FIG. 1 is a method for establishing a depression risk model provided by the present invention.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

The invention provides a model for quantitative assessment of depression risk based on blood biochemical indexes, respectively collecting blood biochemical test indexes and demographic variables of clinical depression patients and healthy people; selecting depression patients and healthy people with significantly different biochemical indexes and demographic variables are independent variables, the present invention is preferred, the effect size Cohen's d>0.3 is selected, of course, other statistical methods can be used to find biochemical indicators with significant differences, and different selection criteria can be formulated according to sample differences; As a dependent variable, a logistic regression model of depression grouping was established, and a judgment threshold was set. The depression risk value was calculated through the established depression grouping regression model, and a quantitative risk assessment was made for patients with depression combined with the judgment threshold.

The independent variables of biochemical indicators established in the present invention include serotonin, brain-derived neurotrophic factor, nerve growth factor (NGF), adrenocortical hormone, C-reactive protein (CRP), interleukin 18 (IL-18), tumor Key biochemical indicators and demographic variables such as necrosis factor alpha (TNF-α), cortisol, progesterone, testosterone, etc. are used to quantitatively assess the risk of depression. The specifically established depression grouping model is as follows:

The logistic regression model of depression grouping was established as follows:

Y=1/(1+exp(-(-2.263* adrenocorticotropic hormone-1.113*brain-derived neurotrophic factor-0.13*interleukin 18+0.747*progesterone-0.451*serotonin+0.894*sex+0.002 *Age+21.221*Whether taking antidepressant or antimanic medication+20.498*Whether fasting before blood draw -2.441))).

In the above formula, gender: female=1, male=0; taking medicine or not: taking=1, not taking=0; fasting before blood drawing: eating before blood drawing=1, fasting before blood drawing=0. Under the premise that the judgment threshold is 0.5, the model hit rate is 0.714, the correct rejection rate is 0.912, and the comprehensive accuracy rate is 0.858.

Of course, when establishing a logistic regression model for depression grouping, the model also includes the following independent variable parameters: C-reactive protein (CRP), cortisol, testosterone and tumor necrosis factor alpha (TNF-α), so that the The evaluation of depression by the model is more accurate. Therefore, the grouped logistic regression model for depression established for the above independent variable parameters can also be expressed by the following mathematical formula:

Y=1/(1+exp(-(-3.632*corticotropin-1.1*brain-derived neurotrophic factor-0.117*interleukin 18+0.835*progesterone-0.935*serotonin+0.671*C-reaction Protein+0.38*cortisol+0.283*testosterone-0.006*tumor necrosis factor alpha+0.834*sex-0.01*age+20.137*whether you take antidepressant or antimanic drugs+19.641*whether fasting before blood draw -2.124)) ).

In the above formula, gender: female=1, male=0; taking medicine or not: taking=1, not taking=0; fasting before blood drawing: eating before blood drawing=1, fasting before blood drawing=0.

Under the premise that the judgment threshold is 0.5, the model hit rate of the above model is 0.781, the correct rejection rate is 0.929, and the comprehensive accuracy rate is 0.875.

The set threshold range is between 0 and 1, which should be selected according to the proportion of depression in the tested sample. When the proportion of patients with depression in the overall sample is less than 50%, the judgment threshold should be set to be greater than or equal to 0.5. When the proportion of patients with depression in the overall sample is >50%, the judgment threshold should be set to be less than or equal to 0.5, that is, when the proportion of patients with depression in the overall sample increases, the judgment threshold should decrease accordingly.

By collecting the values of biochemical test indicators, and substituting the obtained values into the logistic regression model of one of the above depression groups, the patient's depression risk value is obtained. When the obtained value is greater than the set judgment threshold, the patient is considered to be depressed. The risk is higher, and vice versa, the risk of depression is lower. The objective index calculated by the logistic regression model of depression grouping makes the present invention help to construct standardized and objectively measurable depression risk assessment index and quantitative model in addition to the scale test, which overcomes the problem of only using the existing method. Subjective and expert assessments have limitations.

During modeling, in addition to collecting the above biochemical test indicators of clinical depression patients and healthy people, the collection of epinephrine, corticosterone (adrenal ketone), thyroxine, oxytocin, neuropeptide Y, endocannabinoids, Forebrain-derived neurotrophic factor, mature brain-derived neurotrophic factor, norepinephrine NE, interleukin 10 (IL-10), interleukin 12, interleukin 4, gamma aminobutyric acid, dopamine and other biochemical The test indicators can be selected according to the set selection requirements. The selected biological indicators are of course indicators with obvious differences between healthy people and patients with depression. According to the selected biochemical indicators combined with demographic variables as independent variables, the establishment of Corresponding depression risk model. The demographic variables can also include independent variables such as marriage, children, education, occupation, income, and religious beliefs of clinically depressed patients and healthy people.

Of course, the more independent variables and parameters involved in the established logistic regression model, the more accurate the evaluation of patients with depression will be. For example, all the biochemical indicators listed above are used to establish the model. The present invention can select different biochemical indexes to form a discriminant function according to biochemical indexes, availability of population variables and test cost requirements, and is also the fastest way to model and evaluate depression risk.

The present invention also provides a computer-readable storage medium, where the storage medium stores at least one instruction, at least one piece of program, code set or instruction set, the at least one instruction, the at least one piece of program, the code set or the instruction set Loaded and executed by the processor to run the above model and automatically calculate the depression assessment value Y.

The present invention also provides a computer program product, which, when running on the electronic device, enables the electronic device to perform the above-mentioned model establishment and model calculation.

The present invention can store the above-established model in a computer-readable storage medium, and the above-mentioned storage medium can be a read-only memory, a magnetic disk or an optical disk.

Obviously, the above-mentioned embodiments are only examples for clear description, and are not intended to limit the implementation manner. For those of ordinary skill in the art, changes or modifications in other different forms can also be made on the basis of the above description. There is no need and cannot be exhaustive of all implementations here. And the obvious changes or changes derived from this are still within the protection scope of the present invention.

Claims (9)

1. A model for quantitatively evaluating the depression risk based on blood biochemical indexes is characterized in that blood biochemical test indexes and demographic variables of clinical depression patients and healthy people are respectively collected; taking biochemical indexes and demographic variables capable of distinguishing depression patients and healthy people as independent variables, taking depression groups as dependent variables, and establishing a logistic regression model and a judgment threshold value of the depression groups; calculating a depression risk value by the established logistic regression model of the depression group; and (5) performing quantitative risk assessment on the depression patients according to the judgment threshold value.

2. The model for quantitatively evaluating the risk of depression based on biochemical indicators of blood as claimed in claim 1, wherein when the biochemical indicators capable of distinguishing patients with depression from healthy people are selected, the biochemical indicators of patients with depression and healthy people are selected by setting the value of Cohen's d, and the biochemical indicators of patients with depression and healthy people are selected to be larger than the set value of Cohen's d.

3. The model for quantitatively evaluating the risk of depression based on blood biochemical indicators as set forth in claim 2, wherein the independent variable parameters in the logistic regression model establishing the depression group include: adrenocorticotropic hormone, brain-derived neurotrophic factor, interleukin 18(IL-18), progesterone, serotonin biochemical test indexes, sex and age;

a logistic regression model for the depression group was established as follows:

y ═ 1/(1+ exp ((-2.263 × adrenocorticotropic hormone-1.113 × -brain-derived neurotrophic factor-0.13 × interleukin 18+0.747 × progesterone-0.451 × serotonin +0.894 × + gender +0.002 × -age +21.221 × -fasting before taking antidepressant or antimanic +20.498 × -2.441))).

4. The model for quantitatively evaluating the risk of depression based on blood biochemical indicators as recited in claim 3, wherein the establishing of the independent variable parameters in the logistic regression model for the grouping of depression further comprises: c-reactive protein (CRP), cortisol, testosterone, and tumor necrosis factor alpha (TNF-alpha);

the logistic regression model for establishing a group of depression can also be expressed using the following mathematical formula: y ═ 1/(1+ exp ((-3.632 × adrenocorticotropic hormone-1.1 × brain-derived neurotrophic factor-0.117 × interleukin 18+0.835 × progesterone-0.935 × serotonin +0.671 × C-reactive protein +0.38 × cortisol +0.283 × testosterone-0.006 tumor necrosis factor α +0.834 × sex-0.01 × age +20.137 × whether to take antidepressants or antimanic drugs +19.641 × whether to empty stomach-2.124 before drawing blood))).

5. The model for quantitatively evaluating the risk of depression according to the biochemical blood index as claimed in claim 4, wherein the collected biochemical test indexes of clinical depression patients and healthy people can further comprise: corticosterone (adrenalone), thyroxine, gamma aminobutyric acid, dopamine, oxytocin, cortisone, neuropeptide Y, endocannabinoid, mature brain-derived neurotrophic factor, norepinephrine NE, interleukin 10(IL-10), interleukin 12, interleukin 4, nerve growth factor and 5-hydroxytryptamine, a plurality of biochemical indexes capable of distinguishing depression patients and healthy people are selected as independent variable parameters, and a logistic regression model and a judgment threshold value of depression groups are established.

6. The model for quantitative assessment of depression risk based on blood biochemical indicators as claimed in claim 5, wherein the demographic variables further include marital, child, education, occupation, income and presence or absence of religious beliefs of the collected clinical depression patients and healthy population.

7. An electronic device comprising a processor and a memory, wherein the memory stores at least one program that is loaded and executed by the processor to implement the model for quantitative assessment of depression risk based on blood biochemical indicators as claimed in any one of claims 1 to 6.

8. A computer readable storage medium having stored thereon at least one program which is loaded into and executed by a processor to implement the model for quantitative assessment of depression risk based on blood biochemical indicators according to any one of claims 1 to 6.

9. A computer program product comprising computer instructions which, when executed by a processor, implement the model for quantitatively evaluating the risk of depression based on blood biochemical indicators as claimed in any one of claims 1 to 6.

Images