Method and device for providing evaluation score of subject in each single mental disease dimension based on FNIRS

Abstract

The invention provides a method and a device for providing an evaluation score of a subject in each single mental disease dimension based on FNIRS. The method comprises the following steps: at least based on acquired near-infrared data of a target brain region of a subject, an evaluation score of the subject in the dimension of each single mental disease is determined by using each single mental disease analysis model through a processor, each obtained evaluation score represents the tendency of the subject to suffer from the corresponding single mental disease, and each evaluation score represents the tendency of the subject to suffer from the corresponding single mental disease. Wherein each single mental disease analysis model is configured to distinguish the health condition from the corresponding single mental disease; and comparing and presenting the evaluation score of the subject suffering from each single mental disease. Therefore, the evaluation score of each single mental disease can be provided, and the mental health condition of the subject can be truly and comprehensively represented from multiple dimensions.

Description

Technical Field

[0001] This application relates to the field of mental illness assessment technology, specifically to a method and apparatus for providing assessment scores of subjects on various individual dimensions of mental illness based on FNIRS. Background Technology

[0002] Near-infrared spectroscopy brain functional imaging (FNIRS) has been widely used in clinical research on various mental illnesses, assisting doctors in the diagnosis and classification of mental illnesses. Breakthroughs in machine learning, deep learning, and other technologies have provided a more advanced and widely adaptable approach for automatically extracting key features from complex FNIRS data and constructing disease-specific predictive indicators.

[0003] However, the clinical manifestations of various mental illnesses are complex and varied. Different types of mental illnesses often have similar and difficult-to-distinguish symptoms or even physiological manifestations. For example, both bipolar disorder and depression have symptoms such as depressed mood and slowed thinking. Clinically, the greatest hope is to effectively distinguish the type of disease the examinee has, such as whether the examinee has simple depression or bipolar disorder, in order to avoid misdiagnosis as much as possible.

[0004] To achieve this goal, existing FNIRS-based machine learning models for the diagnosis and classification of mental illnesses typically train the same model by inputting sample data from various mental illnesses and healthy individuals together. In application, the subject's near-infrared data or related features are used as input to the trained machine learning model. The model outputs the probability values ​​of the subject having different mental illnesses, and the type of mental illness is determined based on the probability values. For example, the output might show a 50% probability of depression, a 32% probability of anxiety disorder, and a 10% probability of bipolar disorder, indicating a higher probability of depression. Alternatively, relevant features such as brain activation integral values ​​and centroid values ​​can be extracted from the subject's near-infrared data, and the type of mental illness can be identified based on the threshold range of different related features, as illustrated in CN116098591A.

[0005] These methods all aim to use a technological means to absolutely distinguish a single mental illness from other types of mental illnesses. However, in reality, the accuracy of this method in absolutely distinguishing various mental illnesses is very low, and its reliability is insufficient. It cannot effectively assist doctors in diagnosis. For example, the output result of the machine learning model indicates that the probability of the examinee having depression is 50%, but in reality, according to this result, the probability of the examinee having depression is still low. This can easily lead to significant biases in the process of doctors diagnosing the type of mental illness the examinee has based on the assessment results of these methods. Summary of the Invention

[0006] This application addresses the aforementioned technical problems existing in the prior art. The purpose of this application is to provide a method and apparatus for providing assessment scores for subjects on various individual dimensions of mental illness based on FNIRS, which can more comprehensively and objectively reflect the subject's brain function health status based on the provided assessment scores across multiple dimensions of mental illness.

[0007] According to the first aspect of this application, a method is provided for providing assessment scores of subjects on various individual mental illness dimensions based on FNIRS. The method includes: via a processor, at least based on acquired near-infrared data of target brain regions of the subject, using individual mental illness analysis models to determine the assessment scores of the subject on each individual mental illness dimension, wherein each assessment score represents the subject's tendency to have the corresponding individual mental illness, wherein each individual mental illness analysis model is configured to distinguish between health status and the corresponding individual mental illness; and presenting the assessment scores of the subject for each individual mental illness in a comparative manner.

[0008] According to a second aspect of this application, an apparatus is provided for determining the assessment scores of a subject on various individual mental illness dimensions based on FNIRS. The apparatus includes a processor configured to: display the assessment scores of the subject for each individual mental illness on a display interface, wherein each assessment score represents the subject's tendency to have the corresponding individual mental illness; wherein, based at least on the near-infrared data of the target brain region of the subject, the assessment scores of the subject on each individual mental illness dimension are determined using individual mental illness analysis models, wherein each individual mental illness analysis model is configured to distinguish between the health status and the corresponding individual mental illness.

[0009] Compared with the prior art, the beneficial effects of the embodiments of this application are as follows:

[0010] The method provided in this application utilizes multiple single-mental-disorder analysis models to distinguish between health and corresponding single mental illnesses. These models analyze near-infrared data of target brain regions of the examinee, determining the examinee's assessment score for each single mental illness dimension. The assessment scores from each analysis model for each single mental illness are presented together for comparison, allowing for the determination of the examinee's predisposition to various mental illnesses based on these scores, without the need for absolute distinctions in which illness the examinee belongs. This assists doctors in conducting a more comprehensive and objective assessment of the examinee. Specifically, in the embodiments of this application, each single-mental-disorder analysis model is a binary classification model, capable of distinguishing between health status and corresponding single mental illnesses. Thus, each single mental illness can be analyzed independently, and each single-mental-disorder analysis model outputs only the assessment result for that single mental illness. The simultaneous comparison of assessment scores from multiple single mental illnesses comprehensively and accurately reflects the examinee's brain function health level across multiple mental illness dimensions.

[0011] By presenting the assessment scores of the examinee for each specific mental illness, doctors can comprehensively analyze and assess the examinee's brain function health level based on the assessment scores of each specific mental illness. This helps doctors to make an objective diagnosis of the examinee's mental illness and avoid serious biases that could lead to misdiagnosis.

[0012] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above description and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0013] In drawings that are not necessarily drawn to scale, the same reference numerals may describe similar parts in different views. Similar reference numerals with different letter suffixes may indicate different examples of similar components. The drawings generally illustrate various embodiments by way of example rather than limitation, and are used together with the specification and claims to illustrate the disclosed embodiments. Such embodiments are illustrative and exemplary, and are not intended to be exhaustive or exclusive embodiments of the method, apparatus, system, or non-transitory computer-readable medium having instructions for implementing the method.

[0014] Figure 1 A flowchart is shown of a method for providing assessment scores for subjects on various individual mental illness dimensions based on FNIRS, according to an embodiment of this application.

[0015] Figure 2A flowchart illustrating the determination of an examinee's assessment scores on each individual mental illness dimension according to embodiments of this application is shown.

[0016] Figure 3 This diagram illustrates a nonlinear mapping of the output results of each individual mental illness classification model according to an embodiment of this application.

[0017] Figure 4 A schematic diagram of an apparatus for determining an examinee's assessment score on each individual mental illness dimension based on FNIRS, according to an embodiment of this application, is shown. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments. The embodiments of this application will be further described in detail below with reference to the accompanying drawings and specific examples, but these are not intended to limit the scope of this application.

[0019] The terms "first," "second," and similar words used in this application do not indicate any order, quantity, or importance, but are merely used for distinction. The terms "including" or "comprising," etc., used in this application mean that the element preceding the word encompasses the elements listed after the word, and do not exclude the possibility of encompassing other elements. In this application, the arrows shown in the figures for each step are merely examples of the execution order, not limitations. The technical solution of this application is not limited to the execution order described in the embodiments. The steps in the execution order can be combined, broken down, or rearranged, as long as the logical relationship of the executed content is not affected.

[0020] All terms used in this application (including technical or scientific terms) have the same meaning as understood by one of ordinary skill in the art to which this application pertains, unless otherwise specifically defined. It should also be understood that terms defined in general dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant art, and not as idealized or highly formalized, unless expressly defined herein. Technologies and equipment known to one of ordinary skill in the art may not be discussed in detail, but where appropriate, such technologies and equipment should be considered part of the specification.

[0021] Figure 1A flowchart illustrating a method for providing assessment scores for subjects on various individual mental illness dimensions based on FNIRS, according to an embodiment of this application, is shown. The method executes steps S101 to S102 via a processor. In step S101, based at least on acquired near-infrared data of the target brain region of the subject, assessment scores for each individual mental illness dimension are determined using individual mental illness analysis models. Each assessment score represents the subject's predisposition to the corresponding individual mental illness. Each individual mental illness analysis model is configured to distinguish between a health condition and the corresponding individual mental illness. Exemplarily, the target brain region is the prefrontal cortex, or in other embodiments, the target brain region may be the prefrontal cortex and temporal lobe. The prefrontal cortex is a region of the brain closely related to cognitive function, emotion regulation, and decision-making, responsible for many higher-order cognitive functions such as executive function, attention, planning, and organization. Mental illnesses such as depression, schizophrenia, and attention deficit hyperactivity disorder (ADHD) are accompanied by impairments in these cognitive functions. Furthermore, the prefrontal cortex plays a crucial role in regulating emotional responses. For example, patients with depression and anxiety disorders often exhibit prefrontal cortex dysfunction, affecting their ability to control and regulate emotions. In addition, the temporal lobe is also critical in mental illness research; conditions such as depression and anxiety disorders often involve temporal lobe dysfunction.

[0022] Specifically, near-infrared data acquisition devices can be used to collect near-infrared data from subjects performing various tasks or in a resting state. For example, the near-infrared data acquisition device includes at least a headgear worn on the subject's head. The headgear may have multiple probes for transmitting and / or receiving near-infrared light, each of which can be configured as either a transmitting or receiving probe, and each pair of probes can form a detection channel. In some embodiments, one transmitting probe may correspond to multiple receiving probes, or vice versa, wherein the pairing of transmitting and receiving probes depends on the specific requirements of the probe placement, the brain functional region to be detected, etc.

[0023] In some embodiments, each individual mental illness includes at least two of depression, schizophrenia, anxiety disorder, mania, and bipolar disorder. The analysis model for each individual mental illness is a binary classification model; for example, it may include a health-bipolar disorder analysis model, a health-depression analysis model, a health-schizophrenia analysis model, a health-mania analysis model, and a health-anxiety disorder analysis model. Based on the individual mental illness analysis model, healthy individuals can be distinguished from their corresponding individual mental illness, without considering the possibility of other illnesses. For example, the health-bipolar disorder analysis model can distinguish healthy individuals from bipolar disorder patients, and the health-schizophrenia analysis model can distinguish healthy individuals from schizophrenia patients. It is understood that in other embodiments, each individual mental illness may also include other mental illnesses, particularly those with comorbid symptoms that are difficult to distinguish from a single illness, such as depression combined with anxiety disorder.

[0024] Each individual mental illness analysis model is a pre-trained binary classification model, used only to distinguish whether the examinee is in a healthy state or suffers from a specific mental illness. In some embodiments, each individual mental illness analysis model can be trained separately and independently; that is, the model training and the use of the trained model are independent of each other, with no correlation. For example, for the health-bipolar disorder analysis model, only sample data from healthy individuals and bipolar disorder patients are used to train the model. In other embodiments, the individual mental illness analysis models may also have some correlation; for example, during the model training phase, the output of one model may be used to correct the input of another model. This application does not specifically limit this, as long as each individual mental illness analysis model is a binary classification model that only distinguishes whether the examinee is in a healthy state or suffers from a specific mental illness.

[0025] Specifically, the structures of the individual mental illness analysis models can be the same or different. For example, the health-depression analysis model uses a residual neural network as the main structure, and the health-schizophrenia analysis model can use a similar residual network structure. However, the model parameters are based on the task scenario of schizophrenia identification, and other network architectures adapted to this task, such as MobileNet, can also be used. This application does not make any specific limitations on this.

[0026] This is merely an example and does not constitute a specific limitation on the analysis model for any individual mental illness.

[0027] After obtaining near-infrared data of the target brain region of the subject, the near-infrared data can be simultaneously input into each individual mental illness analysis model to process and analyze the near-infrared data at the same time.

[0028] By processing and analyzing the near-infrared data using individual mental illness analysis models, assessment scores for each individual mental illness dimension can be determined. These scores are then used to characterize the subject's predisposition to the corresponding mental illness. In some embodiments, each individual mental illness analysis model analyzes the near-infrared data simultaneously and independently, and the assessment scores determined based on the outputs of each model have the same initial weight. That is, multiple individual mental illness analysis models can be used simultaneously, i.e., the input near-infrared data is identified and analyzed in parallel. These models are equally important; that is, the outputs of each model have equal initial weight, and the assessment scores determined based on each output also have equal initial weight.

[0029] For example, by inputting the collected near-infrared data of the subjects into the health-bipolar disorder analysis model, the health-depression analysis model, and the health-anxiety disorder analysis model, the health-bipolar disorder analysis model can determine the subject's assessment score A on the bipolar disorder dimension, for example, assessment score A is 50 points; the health-depression analysis model can determine the subject's assessment score B on the depression dimension, for example, assessment score B is 90 points; and the health-anxiety disorder analysis model can determine the subject's assessment score C on the anxiety dimension, for example, assessment score C is 95 points.

[0030] Among them, assessment scores A, B, and C are unrelated and equally important, each independently reflecting the examinee's brain function health level in the corresponding mental illness dimension.

[0031] Based on this embodiment, for example, doctors can comprehensively understand the examinee's health status across various dimensions of mental illness using assessment scores A, B, and C, with a focus on the examinee's performance in the dimensions of depression and anxiety. Specifically, it can also guide doctors to combine other relevant medical data (such as focusing on indicators related to depression and anxiety, such as scale scores and subjective complaints) to determine whether the examinee has a mental illness involving both anxiety and depression. In other words, if there are diagnostic features that are not significantly different between at least two mental illnesses, such as depression and bipolar disorder, the individual mental illness analysis models can be used to objectively and realistically assess the examinee's mental state across each individual mental illness dimension. The assessment scores for each individual mental illness dimension are presented to the doctor in comparison, aiming to inform the doctor about the examinee's performance in these different mental illnesses, so that the doctor can make their own judgment or make a comprehensive judgment based on other clinical manifestations, rather than making an absolute judgment that the examinee has a particular illness. Some mental illnesses share similar or identical clinical manifestations, symptoms, and even ranges of similar or identical indicator parameters. Therefore, it is difficult to differentiate between various mental illnesses using a single model. Existing mental illness classification models generally have an accuracy rate of less than 60%, indicating low reliability. This can easily lead to significant biases in doctors' diagnoses of the mental illness type of examinees based on these methods' assessment results, hindering their widespread clinical application. In the proposed solution, each individual mental illness analysis model is a binary classification model, which can distinguish between health status and the corresponding individual mental illness. Each individual mental illness is analyzed independently, comprehensively and accurately reflecting the examinee's brain function health level across multiple dimensions of mental illness.

[0032] In other embodiments, the specific process of determining the assessment scores of the examinee on each individual mental illness dimension using individual mental illness analysis models is as follows: Figure 2 As shown. In step S201, the near-infrared data of the target brain region of the subject are input into each trained single mental illness classification model to obtain the output results of each single mental illness classification model; in step S202, each output result is nonlinearly mapped to an evaluation score within a threshold range.

[0033] Specifically, the output results of each individual mental illness analysis model are not directly used to characterize the examinee's tendency to have a mental illness, nor can they be directly used as a reference for doctors to diagnose the examinee's mental illness type. Instead, the output results need to be non-linearly mapped to a threshold range to obtain an assessment score, and the examinee's mental health status is comprehensively assessed based on the assessment score.

[0034] The method provided in this embodiment differs from existing methods that use the same analytical model to output the probability of different types of mental illnesses. It can comprehensively reflect the brain function health level of the examinee in multiple dimensions of mental illness, and avoid the misleading effect of absolute probability indicators on doctors' diagnosis.

[0035] The threshold range can be 0-100 points, 0-200 points, etc., and is not limited thereto. Specifically, the threshold range can be a score range used to assess the mental health status of the examinee, and the output of the single mental illness analysis model is non-linearly mapped to the score value within the threshold range. Preferably, the score value presented after non-linear mapping of the parameter value representing a higher probability of the patient having a single mental illness output by the single mental illness analysis model is lower than the probability score value linearly corresponding to that parameter value. For example, the output of the single mental illness analysis model shows a probability of 98% for having that single mental illness, indicating that the patient has a high probability of having the disease, but the score presented to the doctor is only 85 points, which is lower than the probability score of 98 linearly corresponding to 98%.

[0036] In some embodiments, the threshold score range includes multiple score intervals, with different score intervals corresponding to different probability intervals of the output results of a single mental illness analysis model. In some embodiments, the threshold range may include a score range for the normal state, a score range for the intermediate state, and a score range for the predisposition. The nonlinear mapping may map the output result in a first probability segment to the score range for the normal state, the output result in a second probability segment to the score range for the intermediate state, and the output result in a third probability segment to the score range for the predisposition. For example, as... Figure 3 As shown, in the first probability range of 0%-10%, the output results within this range are mapped to a score of 0-50. In the second probability range of 10%-90%, the output results within this range are mapped to a score of 50-70. In the third probability range of 90%-100%, the output results within this range are mapped to a score of 70-90. In other words, even if the probability value of the output result from a single mental illness analysis model is 100%, it will not directly show the doctor an absolute assessment result of a 100% probability of having a single mental illness. Instead, through non-linear mapping, it displays an assessment score with a maximum of 90 points. This avoids misleading the doctor's diagnosis with absolute probability indicators and also allows the doctor to combine other indicators for a more accurate comprehensive assessment of the examinee's mental health status.

[0037] In step S102, the assessment scores for each individual mental illness of the examinee are presented in comparison. For example, each type of mental illness is displayed in relation to its corresponding assessment score. Specifically, all types of mental illnesses and their corresponding assessment scores can be presented in the form of a bar chart, or the assessment scores for each dimension of each mental illness can be presented in the form of a tree diagram, for example, showing the position of the assessment scores for each dimension of each mental illness in the tree diagram. In some embodiments, the assessment scores for each dimension of each mental illness can also be presented in the form of a table. This application does not specifically limit the presentation method of the assessment scores, as long as they can be presented to the doctor together so that the doctor can simultaneously observe the examinee's performance in these different mental illnesses and make a comprehensive assessment.

[0038] Alternatively, the individual outputs of each mental illness analysis model and the threshold range of the assessment scores can be displayed individually or simultaneously, along with the position of each output within the threshold range through non-linear mapping. For example, as... Figure 3 As shown, taking the output of the health-bipolar disorder analysis model as M1, the output of the health-depression analysis model as M2, and the output of the health-anxiety disorder analysis model as M3 as examples, after nonlinear mapping, the assessment score obtained based on M1 is N1, the assessment score obtained based on M2 is N2, and the assessment score obtained based on M3 is N3. M1, M2, and M3 are then nonlinearly mapped to assessment scores N1, N2, and N3 within a threshold range, and the threshold range of assessment scores is 0-100 points, respectively, to facilitate users' comprehensive consideration and assessment of the examinee's condition.

[0039] This is provided as an example only and does not constitute a limitation on any specific solution.

[0040] Based on the method provided in this application, assessment scores for each individual mental illness dimension can be obtained. Each assessment score is analyzed by a separate individual mental illness analysis model. Therefore, each assessment score can independently characterize the examinee's tendency to have each individual mental illness. The method provided in this application does not directly output the probability value of the examinee having a single mental illness, nor does it directly output the probability of the examinee having comorbidities (i.e., it does not distinguish between comorbid symptoms, such as anxiety combined with depression). Instead, each individual mental illness analysis model provides assessment scores for the examinee in each individual mental illness dimension, allowing users such as doctors to comprehensively consider the examinee's condition from multiple mental illness dimensions based on the provided assessment scores.

[0041] By presenting the assessment scores for each individual mental illness output by each analysis model together, it is possible to determine the examinee's tendency for various mental illnesses based on the assessment scores, without having to make absolute distinctions about which type of illness the examinee belongs to. This can assist doctors and other users in making a more comprehensive and objective assessment of the examinee's mental illness, and improve the accuracy and objectivity of the diagnosis of the examinee's mental illness type.

[0042] In some embodiments of this application, the method further includes obtaining representative values ​​of near-infrared data for each subdivided brain region based on a preset number of subdivided brain regions targeting the target brain region; and determining the subject's assessment score on each individual mental illness dimension using individual mental illness analysis models based on the representative values ​​of the near-infrared data for each subdivided brain region. For example, 10 brain regions can be divided (or representative brain regions can be divided, for example, first selecting 10 brain regions with a high correlation to mental illness from 15 subdivided brain regions). After dividing these 10 subdivided brain regions, representative values ​​of near-infrared data for each subdivided brain region are obtained. For example, the representative value can be the average value of the near-infrared data for each subdivided brain region, or a segment of near-infrared data, or the maximum value of the near-infrared data, or data obtained based on near-infrared data to characterize the activation level and / or brain functional connectivity of each subdivided brain region. Specifically, the activation level of each sub-brain region can be characterized by relevant parameters such as mean, variance, integral value, and centroid value obtained from near-infrared data. The status of brain functional connectivity can also be characterized by relevant parameters such as brain functional connectivity parameters and dynamic functional connectivity coefficient of variation.

[0043] Determining the assessment scores for each individual mental illness dimension based on representative values ​​of near-infrared data from various sub-brain regions can reduce the computational load on individual mental illness analysis models and improve computational efficiency. Furthermore, using representative values ​​of near-infrared data, which are highly correlated with mental illnesses, to determine the assessment scores for each individual mental illness dimension can assist physicians in improving assessment accuracy.

[0044] In one specific embodiment, the target brain region may be the prefrontal lobe, or the prefrontal lobe and temporal lobe. In the case where the target brain region is the prefrontal lobe and temporal lobe, preferably, the subdivided brain regions include the left and right ventral prefrontal lobes, the left and right dorsal prefrontal lobes, the left and right ventrolateral prefrontal lobes, the left and right dorsolateral prefrontal lobes, and the left and right lateral temporal lobes.

[0045] In other embodiments of this application, the method further includes: based on the near-infrared data of the acquisition channels of the target brain region or each sub-brain region, using Gram angle field transformation to obtain Gram angle field image data of the corresponding target brain region or each sub-brain region; based on the Gram angle field image data, determining the individual output results of the subject for each single mental illness, and mapping each output result nonlinearly to an evaluation score within a threshold range. For example, the time series data of the original acquisition channels of the target brain region or each sub-brain region can be preprocessed as necessary, such as motion artifact removal, filtering, calculation of the relative content of oxyhemoglobin (HBO) and deoxyhemoglobin (HBR), standardization, etc., before performing Gram angle field transformation.

[0046] Specifically, for example, data at specific time points can be mapped onto a unit circle to obtain a series of angle values. A Gram matrix can then be constructed by calculating a function of the angle difference between any two time points, such as using a complex exponential function or a weighted form combining sine and cosine functions. By performing specific operations on the Gram matrix, a Gram angle field matrix is ​​obtained. Finally, the Gram angle field matrix can be visualized as a heatmap, thus showing the intensity relationships between samples in the near-infrared data. Specifically, the time-series data from each acquisition channel, after Gram angle field transformation, becomes two-dimensional image data. By stitching together the two-dimensional image data from each acquisition channel with the acquisition channel as the third dimension, a three-dimensional data similar to an RGB image is obtained. The Gram angle field image data obtained from near-infrared data from multiple acquisition channels using Gram angle field transformation is three-dimensional data with the acquisition channel as the third dimension. This is merely an illustrative example and does not constitute a limitation on any specific solution.

[0047] The inventors discovered through verification that by processing and analyzing Gram field image data using individual mental illness analysis models, the output results of each individual mental illness analysis model can be obtained, thereby improving the accuracy of analyzing and diagnosing the type of mental illness of the examinee.

[0048] Furthermore, by using Gram angle field transformation to obtain the corresponding Gram angle field data based on near-infrared data from multiple acquisition channels of each subdivided brain region, compared to using near-infrared data from each acquisition channel of the target brain region, the computational load of the system can be reduced and the data processing efficiency can be improved.

[0049] Figure 4An apparatus for determining an examinee's assessment scores on various individual mental illness dimensions based on FNIRS is shown. The apparatus includes a processor 401 and a display 402. The processor 401 is configured to display the examinee's assessment scores for each individual mental illness on the display interface of the display 402, each assessment score representing the examinee's predisposition to the corresponding individual mental illness. The assessment scores are determined based at least on near-infrared data of the examinee's target brain regions using individual mental illness analysis models, each of which is configured to distinguish between a health condition and the corresponding individual mental illness.

[0050] Each individual mental illness analysis model is analyzed independently, and each outputs the subject's assessment score on each individual mental illness dimension. The output results of each individual mental illness analysis model are not correlated. This device can comprehensively and realistically reflect the subject's brain function health level on multiple mental illness dimensions.

[0051] In some embodiments of this application, the processor 401 is further configured to display each individual mental illness type and its corresponding assessment score on a display interface; and / or to display on the display interface the various output results of each individual mental illness analysis model and the threshold range of the assessment index, as well as the position of each output result nonlinearly mapped to the threshold range. By presenting the assessment scores of the examinee for each individual mental illness on the display interface, doctors can comprehensively analyze and consider the examinee's brain function health level based on the assessment scores of each individual mental illness. This facilitates objective diagnosis of the examinee's mental illness by doctors, avoiding serious bias and misdiagnosis.

[0052] The processor 401 may be a processing device including one or more general-purpose processing devices, such as a microprocessor, a central processing unit (CPU), a graphics processing unit (GPU), etc. More specifically, the processor 401 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor running other instruction sets, or a processor running a combination of instruction sets. The processor 401 may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a system-on-a-chip (SoC), etc. The processor 401 may be included within a near-infrared brain functional imaging device, or it may be located outside the near-infrared brain functional imaging device, cooperating with the near-infrared brain functional imaging device to perform relevant data analysis and processing. The processor 401 in the device is configured to execute the method described in the various embodiments of this application for providing assessment scores of subjects on various individual mental illness dimensions based on FNIRS.

[0053] This application describes various operations or functions that can be implemented as software code or instructions, or defined as software code or instructions. Such content can be directly executable source code or differential code (“incremental” or “patch” code) (“object” or “executable” form). The software code or instructions can be stored in a computer-readable storage medium and, when executed, can cause a machine to perform the described functions or operations, and include any mechanism for storing information in a machine-accessible form, such as recordable or non-recordable media (e.g., read-only memory (ROM), random access memory (RAM), disk storage media, optical storage media, flash memory devices, etc.).

[0054] The exemplary methods described in this application can be implemented, at least in part, by a machine or computer. In some embodiments, a computer-readable storage medium stores computer program instructions that, when executed by a processor, cause the processor to perform the methods for providing assessment scores for subjects on various individual dimensions of mental illness based on FNIRS, as described in the various embodiments of this application.

[0055] Each step of the method for providing assessment scores of subjects on each individual mental illness dimension based on FNIRS in the various embodiments described above can be incorporated herein and implemented by a processor running computer program instructions, which will not be elaborated here.

[0056] Implementations of such methods may include software code, such as microcode, assembly language code, high-level language code, etc. Various software programming techniques can be used to create various programs or program modules. For example, program parts or program modules can be designed using or with the aid of Java, Python, C, C++, assembly language, or any known programming language. One or more of such software parts or modules can be integrated into a computer system and / or a computer-readable medium. Such software code may include computer-readable instructions for performing various methods. This software code can form part of a computer program product or a computer program module. Furthermore, in the example, the software code may be tangibly stored on one or more volatile, non-transitory, or non-volatile tangible computer-readable media, for example, during execution or at other times. Examples of such tangible computer-readable media may include, but are not limited to, hard disks, removable disks, removable optical discs (e.g., optical discs and digital video discs), magnetic tape cassettes, memory cards or memory sticks, random access memory (RAM), read-only memory (ROM), etc.

[0057] Furthermore, although exemplary embodiments have been described herein, their scope includes any and all embodiments based on this application that have equivalent elements, modifications, omissions, combinations (e.g., schemes involving intersections of various embodiments), adaptations, or alterations. Elements in the claims will be interpreted broadly based on the language used in the claims and are not limited to the examples described in this specification or during the implementation of this application, which will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered illustrative only, and the true scope and spirit are indicated by the full scope of the following claims and their equivalents.

[0058] The above description is intended to be illustrative and not restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. Other embodiments may be used by those skilled in the art upon reading the above description. Furthermore, in the above detailed description, various features may be grouped together to simplify the application. This should not be construed as an intention that a disclosed feature not claimed is necessary for any claim. Rather, the subject matter of the application may be less than all the features of a particular disclosed embodiment. Thus, the claims are incorporated herein by reference as examples or embodiments, wherein each claim is an independent, separate embodiment, and these embodiments are contemplated as being combined with each other in various combinations or arrangements. The scope of this application should be determined by reference to the appended claims and the full scope of their equivalents.

[0059] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for providing assessment scores for subjects on various individual dimensions of mental illness based on the FNIRS, characterized in that, The method includes: via a processor, Based at least on the near-infrared data of the target brain region of the subject, the assessment scores of the subject in each single mental illness dimension are determined by using the single mental illness analysis model. The obtained assessment scores represent the subject's tendency to have the corresponding single mental illness. The single mental illness analysis model is configured to distinguish between the health status and the corresponding single mental illness. The assessment scores of the subjects for each corresponding individual mental illness are presented in comparison.

2. The method according to claim 1, characterized in that, Each individual mental illness includes at least two of the following: depression, schizophrenia, anxiety disorder, mania, and bipolar disorder.

3. The method according to claim 1 or 2, characterized in that, There are at least two mental illnesses that have indistinguishable diagnostic features.

4. The method according to claim 1 or 2, characterized in that, The process of using individual mental illness analysis models to determine the assessment scores of the examinees on each individual mental illness dimension specifically includes: The near-infrared data of the target brain region of the subject were input into each of the trained single mental illness classification models to obtain the output results of each single mental illness classification model. Each output result is non-linearly mapped to an evaluation score within a threshold range.

5. The method according to claim 1 or 2, characterized in that, The comparison presents the assessment scores of the subjects for each corresponding individual mental illness, specifically including: Display the corresponding assessment scores for each individual mental illness type; and / or, The output results of each individual mental illness analysis model and the threshold range of the assessment scores are displayed, as well as the position of each output result in the threshold range through nonlinear mapping.

6. The method according to claim 1 or 2, characterized in that, Each individual mental illness analysis model analyzes near-infrared data simultaneously and independently, and the assessment scores determined based on the output results of each individual mental illness analysis model have the same initial weights.

7. The method according to claim 1 or 2, characterized in that, The method further includes: Based on a preset number of subdivided brain regions targeting the target brain region, representative values ​​of near-infrared data for each subdivided brain region are obtained. Based on the representative values ​​of near-infrared data from each subdivided brain region, the assessment scores of the examinees on each individual mental illness dimension were determined using individual mental illness analysis models.

8. The method according to claim 7, characterized in that, The representative value is data obtained from near-infrared data to characterize the activation level and / or brain functional connectivity of each subdivided brain region.

9. The method according to claim 1 or 2, characterized in that, The method further includes: Based on the near-infrared data of the acquisition channels of the target brain region or each sub-brain region, Gram angle field transformation is used to obtain Gram angle field image data of the corresponding target brain region or each sub-brain region; based on the Gram angle field image data, the various output results of the examinee for each individual mental illness are determined.

10. A device for determining an examinee's assessment scores on each individual dimension of mental illness based on the FNIRS, characterized in that, The device includes a processor, the processor being configured to: The assessment scores of the examinee for each corresponding individual mental illness are displayed on the screen, and each assessment score represents the examinee's tendency to have each corresponding individual mental illness. Specifically, based at least on the near-infrared data of the target brain region of the examinee, the individual mental illness analysis models are used to determine the examinee's assessment score on each individual mental illness dimension. The individual mental illness analysis models are configured to distinguish between the health status and the corresponding individual mental illness.

11. The apparatus according to claim 10, characterized in that, The processor is further configured as follows: The display interface shows each individual mental illness type and its corresponding assessment score; and / or The display interface shows the output results of each individual mental illness analysis model and the threshold range of the assessment index, as well as the position of each output result in the threshold range through nonlinear mapping.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the method for providing assessment scores for subjects on each individual dimension of mental illness based on FNIRS as described in any one of claims 1-9.

Images