Method, device, processor and computer-readable storage medium for generating and processing mental illness image reports

Abstract

The present invention relates to a method for realizing the generation and processing of image reports of mental illness, including evaluating the overall resolution, noise and grayscale deviation level of the image according to the original image of the subject submitted for examination; calculating the comprehensive quality rating of the image according to the evaluation result; intercepting the coronal plane, sagittal plane and cross section of the original brain image of the submitted data, as well as the coronal plane, sagittal plane and cross section of the brain tissue probability map. The present invention also relates to a device, a processor and a computer-readable storage medium thereof for realizing the function of generating image reports of mental illness. The method, device, processor and computer-readable storage medium thereof for realizing the generation and processing of image reports of mental illness of the present invention can provide psychiatrists with image quantitative examination reports with clinical reference value and interpretability. The report generation method provides a method for organizing image analysis results according to clinical needs and efficiently presenting information required for clinical use.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to the field of medical imaging data information, and specifically refers to a method, device, processor and computer-readable storage medium thereof for realizing the generation and processing of mental illness imaging reports. Background Art

[0002] Patients with mental illness have obvious abnormalities in brain structure and function. However, the current diagnosis and treatment of mental illness mainly rely on subjective assessment of symptoms (such as interviews and questionnaire assessments), and lack objective measurement methods, especially assessment methods for the degree of brain abnormalities. Neuroimaging technology is based on the principle of magnetic resonance imaging and reflects the structural, functional and metabolic characteristics of an individual's brain with high resolution. Neuroimaging technology has been widely used in medical fields such as neurology and neurosurgery to provide necessary information for the discovery and diagnosis of organic brain lesions.

[0003] At present, research on quantitative image analysis and auxiliary diagnosis of brain diseases has been carried out at home and abroad, including the use of deep learning methods to assist in the diagnosis of Alzheimer's disease, and the analysis of brain morphological indicators using clinical or scientific research data. At present, there is no imaging reporting system that meets clinical needs in the field of mental health. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method, device, processor and computer-readable storage medium for generating and processing mental illness image reports that meet the requirements of high accuracy, high efficiency and a wide range of applications.

[0005] In order to achieve the above-mentioned purpose, the method, device, processor and computer-readable storage medium for generating and processing mental illness image reports of the present invention are as follows:

[0006] The method for realizing the generation and processing of mental illness image reports is mainly characterized in that the method comprises the following steps:

[0007] (1) Based on the original images of the subjects, evaluate the overall image resolution, noise, and grayscale deviation level;

[0008] (2) Calculate the comprehensive quality rating of the image based on the evaluation results;

[0009] (3) The coronal, sagittal, and transverse sections of the original brain images and the coronal, sagittal, and transverse sections of the brain tissue probability map are intercepted as the visualization part of the image quality assessment report;

[0010] The method further includes the step of diagnosing the association between abnormalities in various brain regions and mental illness, specifically including the following processing steps:

[0011] (2-1) For each brain structural partition, report the numerical value of the image analysis index, the variation range of the healthy population, and the abnormal distribution map of the imaging index;

[0012] (2-2) indicates values ​​outside the normal range;

[0013] (2-3) show the degree of association between abnormal features and each diagnostic type, expressed as odds ratio;

[0014] The method further comprises the step of generating a key brain image feature distribution map, which specifically comprises the following processing steps:

[0015] (3-1) Calculate the effect size of each neuroimaging feature for identifying the diagnostic type of mental illness, and extract the features with the highest effect size;

[0016] (3-2) extracting the offset of the test data relative to the healthy population on the feature;

[0017] (3-3) Adjust the population percentiles according to the effect size and brain region indicators;

[0018] (3-4) Providing a multivariate display method to present an overview of the key characteristic abnormalities of a disease and indicate the overall risk.

[0019] Preferably, the method further comprises a step of comorbidity diagnosis, which specifically comprises the following processing steps:

[0020] (1-1) Calculate the identification results of each disease separately;

[0021] (1-2) Analyze comorbidities based on the identification results of a single disease;

[0022] (1-3) The probability threshold is used as the identification standard for a single disease to determine whether the test data is identified as more than two diseases. If so, the comorbidity is reported; otherwise, the multi-category prediction probability between disease categories is reported.

[0023] Preferably, the step of comorbidity diagnosis also includes the following processing:

[0024] Based on the current samples submitted for examination, the predicted probability of binary diseases between any two categories of healthy people and common mental illnesses is provided.

[0025] Preferably, the step (3-3) is specifically as follows:

[0026] Determine whether the effect size is positive for the brain region indicator. If so, keep the population percentile unchanged; otherwise, convert the population percentile according to the following formula:

[0027] Population percentile = 1 - original population percentile.

[0028] The main feature of the device for realizing the processing function of mental illness image report is that the device comprises:

[0029] a processor configured to execute computer executable instructions;

[0030] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for generating and processing a mental illness image report are implemented.

[0031] The main feature of the processor for realizing the generation and processing of mental illness image reports is that the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the various steps of the above-mentioned method for realizing the generation and processing of mental illness image reports are realized.

[0032] The main feature of the computer-readable storage medium is that a computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the above-mentioned method for generating and processing a mental illness image report.

[0033] The method, device, processor and computer-readable storage medium for generating and processing mental illness image reports of the present invention can provide psychiatrists with clinical reference value and interpretable quantitative image examination reports. The report generation method provides a method for organizing image analysis results according to clinical needs and efficiently presenting information required by the clinic. The report generated by the method can be used as part of the medical record. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The present invention is a flowchart of a method for implementing mental illness image report generation and processing.

[0035] Figure 2 A schematic diagram of an embodiment of an image quality assessment report of a method for generating and processing a mental illness image report according to the present invention.

[0036] Figure 3 A schematic diagram of brain region information for the method of the present invention for realizing the generation and processing of mental illness image reports.

[0037] Figure 4 A schematic diagram of image feature distribution for the method of the present invention for realizing the generation and processing of mental illness image reports. DETAILED DESCRIPTION

[0038] In order to more clearly describe the technical content of the present invention, further description is given below in conjunction with specific embodiments.

[0039] See also Figures 1 to 4 As shown, the method for generating and processing a mental illness image report of the present invention includes the following steps:

[0040] (1) Based on the original images of the subjects, evaluate the overall image resolution, noise, and grayscale deviation level;

[0041] (2) Calculate the comprehensive quality rating of the image based on the evaluation results;

[0042] (3) The coronal, sagittal, and transverse sections of the original brain images of the submitted data, as well as the coronal, sagittal, and transverse sections of the brain tissue probability map, are captured as the visualization part of the image quality assessment report.

[0043] As a preferred embodiment of the present invention, the method further comprises a step of comorbidity diagnosis, which specifically comprises the following processing steps:

[0044] (1-1) Calculate the identification results of each disease separately;

[0045] (1-2) Analyze comorbidities based on the identification results of a single disease;

[0046] (1-3) The probability threshold is used as the identification standard for a single disease to determine whether the test data is identified as more than two diseases. If so, the comorbidity is reported; otherwise, the multi-category prediction probability between disease categories is reported.

[0047] As a preferred embodiment of the present invention, the step of comorbidity diagnosis further includes the following processing:

[0048] Based on the current samples submitted for examination, the predicted probability of binary diseases between any two categories of healthy people and common mental illnesses is provided.

[0049] As a preferred embodiment of the present invention, the method further comprises the step of diagnosing the association between abnormalities in various brain regions and mental illness, specifically comprising the following processing steps:

[0050] (2-1) For each brain structural partition, report the numerical value of the image analysis index, the variation range of the healthy population, and the abnormal distribution map of the imaging index;

[0051] (2-2) indicates values ​​outside the normal range;

[0052] (2-3) shows the degree of association between abnormal features and each diagnostic type, expressed as odds ratio.

[0053] As a preferred embodiment of the present invention, the method further comprises the step of generating a key brain image feature distribution map, which specifically comprises the following processing steps:

[0054] (3-1) Calculate the effect size of each neuroimaging feature for identifying the diagnostic type of mental illness, and extract the features with the highest effect size;

[0055] (3-2) extracting the offset of the test data relative to the healthy population on the feature;

[0056] (3-3) Adjust the population percentiles according to the effect size and brain region indicators;

[0057] (3-4) Providing a multivariate display method to present an overview of the key characteristic abnormalities of a disease and indicate the overall risk.

[0058] As a preferred embodiment of the present invention, the step (3-3) is specifically as follows:

[0059] Determine whether the effect size is positive for the brain region indicator. If so, keep the population percentile unchanged; otherwise, convert the population percentile according to the following formula:

[0060] Population percentile = 1 - original population percentile.

[0061] The device for realizing the processing function of mental illness image report of the present invention, wherein the device comprises:

[0062] a processor configured to execute computer executable instructions;

[0063] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for generating and processing a mental illness image report are implemented.

[0064] The processor of the present invention is used to implement the generation and processing of mental illness image reports, wherein the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the various steps of the above-mentioned method for implementing the generation and processing of mental illness image reports are implemented.

[0065] The computer-readable storage medium of the present invention stores a computer program thereon, and the computer program can be executed by a processor to implement the various steps of the above-mentioned method for generating and processing a mental illness image report.

[0066] In a specific embodiment of the present invention, a method for generating a psychiatric brain imaging report based on neuroimaging analysis results is provided, thereby providing psychiatrists with an examination report that has clinical reference value and is interpretable.

[0067] The present invention implements the function of generating a mental illness image report with the following processing steps:

[0068] 1. Image quality assessment and method for generating an image quality assessment report: Image quality affects the accuracy of feature analysis based on images, so it is necessary to evaluate and report image quality. Based on the original images of the subjects submitted for inspection, the overall resolution, noise, and grayscale deviation level of the image are evaluated, and based on the evaluation results, the comprehensive quality rating of the image is calculated; the higher the comprehensive quality level, the better the image quality, and accordingly, the higher the content validity of the subsequent image analysis report. At the same time, the coronal plane, sagittal plane, and cross-section of the original brain image of the submitted data, as well as the above three sections of the brain tissue probability map, are intercepted for the present invention as the visualization part of the image quality assessment report, so that users can intuitively check the quality of brain images.

[0069] 2. Reporting method for diagnostic recommendations:

[0070] Provide a single disease recognition result, that is, for disease category a, calculate the probability that the submitted data belongs to disease a and the probability that it does not belong to disease a based on brain images.

[0071] For common categories of mental illness, the present invention calculates the recognition results of individual diseases respectively and displays the recognition results in the intelligent imaging diagnosis report; based on the recognition results of individual diseases, the comorbidity situation is analyzed.

[0072] Use a probability threshold (such as 50%) as the identification standard for a single disease. 1) When the test data is identified as more than two diseases by the test subject at the same time, the possible comorbidity is prompted in the report. 2) If it is judged that there is no comorbidity, report the multi-class prediction probability between disease categories, that is, the probability of identifying a certain diagnostic type among different types. 3) At the same time, give the binary disease prediction probability between any two categories of healthy people and common mental illnesses for the current sample submitted for inspection.

[0073] 3. Reporting methods for abnormalities in various brain regions and their association with the diagnosis of mental illness:

[0074] For each structural division of the brain (such as hippocampus, amygdala, superior temporal gyrus, etc.), report: 1) the value of image analysis indicators such as volume / surface area / cortical thickness; 2) the range of variation in the healthy population; 3) the abnormal distribution map of imaging indicators (offset relative to the healthy population), indicating values ​​that are beyond the normal range (too high or too low); 4) the degree of association between the abnormal feature and each diagnostic type, expressed as an odds ratio.

[0075] 4. Generate key brain image feature distribution map:

[0076] For a certain type of mental illness diagnosis, based on a large-scale database, the effect size of each neuroimaging feature for identifying the mental illness is calculated, and several features with the highest effect size are extracted.

[0077] The deviation of the test data on these characteristics relative to the healthy population is extracted (such as the percentile in the healthy population).

[0078] Effect size sign correction: For brain region indicators with positive effect size values, the population percentile remains unchanged; for brain region indicators with negative effect size values, the population percentile is converted to: 1-original population percentile.

[0079] Use multivariate display methods (such as radar charts) to present an overview of abnormal key features for a certain type of disease, indicating comprehensive risks. For example, use a radar chart to display abnormalities in key features related to depression diagnosis in the submitted data. Its area reflects the risk of the submitted subject for the mental illness. The larger the area covered by the radar chart, the greater the risk of the disease.

[0080] In the present invention, the content of the image quality assessment report includes but is not limited to resolution assessment, noise assessment and deviation field assessment, and may also include image contrast assessment, brain region segmentation assessment, brain region index abnormality assessment, organic lesion assessment, etc.

[0081] The features targeted by the present invention include, but are not limited to, magnetic resonance features, and may also include CT image features, PET image features, near-infrared image features, EEG features, and the like.

[0082] The diseases targeted by the present invention include but are not limited to mental illnesses, and are also applicable to other diseases that require brain imaging examinations, such as Alzheimer's disease, Parkinson's disease, epilepsy, etc.

[0083] The imaging features targeted by the present invention include, but are not limited to, volume, cortical thickness, surface area, and may also include curvature, sphericity, folding rate, and the like.

[0084] In the present invention, the order of the image quality assessment report, the image intelligent diagnosis report, and the quantitative image report can be swapped, or they can be split separately to form independent reports.

[0085] In the present invention, the content of the intelligent imaging diagnosis report includes but is not limited to single disease identification, comorbidity analysis, five-category disease category assessment and classification between disease categories, and may also include health / disease status analysis, disease subtype identification, etc.

[0086] In the present invention, the brain image is first evaluated for quality by an image quality inspection module, and the comprehensive quality rating of the image is calculated based on the overall resolution, noise, and grayscale deviation level of the image. Since the image quality has a fundamental impact on the validity of subsequent image-based intelligent diagnosis and quantitative evaluation, providing an image quality inspection link can provide the users of the system with the confidence of the report, which can be used as a reference to better evaluate the reference significance of this image report.

[0087] For the diagnosis and reporting of mental illness, the technology disclosed in this invention realizes three core functions:

[0088] 1. Quantitatively (rather than just judging the level) report the morphological indicators of each brain region, their range of variation in healthy people, their relative positions in the individuals tested, abnormalities, and their relationship with mental illness.

[0089] The present invention reports, for each structural division of the brain (such as the hippocampus, amygdala, superior temporal gyrus, etc.): 1) the values ​​of image analysis indicators such as volume / surface area / cortical thickness 2) the range of variation in a healthy population 3) abnormal distribution maps of imaging indicators (offset relative to a healthy population), indicating values ​​that are beyond the normal range (too high or too low); 4) the degree of association between the abnormal feature and each diagnostic type, expressed as an odds ratio.

[0090] 2. Infer the disease diagnosis type based on brain image features. In the present invention, in addition to calculating the recognition results of a single disease, the intelligent diagnosis report part also analyzes the comorbidity based on the recognition results of a single disease. When the submitted data is identified as two or more diseases at the same time, the possible comorbidity is prompted in the report; if it is judged that there is no comorbidity, the multi-class prediction probability between disease categories is reported, that is, the probability of identifying a certain diagnosis type among different types.

[0091] 3. In addition, the intelligent diagnosis report also provides the binary disease prediction probability between any two categories of healthy people and common mental illnesses for the current sample. The components and operation methods of the above intelligent diagnosis report are not involved in the existing technology, and there is currently no similar solution in the field of mental illness.

[0092] In this technical solution, since image quality has a fundamental impact on the validity of subsequent image-based intelligent diagnosis and quantitative evaluation, the image quality must be checked first before the subsequent image-based disease evaluation to screen out neurological images of qualified quality for subsequent analysis; at the same time, providing an image quality inspection link can provide the confidence of the report to the users of the system, which can be used as a reference to better evaluate the reference significance of this image report. At present, the research on image quality inspection in the field of neuroimaging is mainly focused on the research of image quality assessment methods, and there is no precedent for integrating image resolution / noise level / artifact conditions into auxiliary diagnosis systems to evaluate the validity of auxiliary diagnosis results. At the same time, the public data sets currently used by technicians in this field (such as ADNI, OASIS, GSP, etc.) do not provide image quality assessment.

[0093] The specific implementation scheme of this embodiment can refer to the relevant description in the above embodiment, which will not be repeated here.

[0094] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0095] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0096] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0097] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0098] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the corresponding program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.

[0099] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0100] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0101] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0102] The method, device, processor and computer-readable storage medium for generating and processing mental illness image reports of the present invention can provide psychiatrists with clinical reference value and interpretable quantitative image examination reports. The report generation method provides a method for organizing image analysis results according to clinical needs and efficiently presenting information required by the clinic. The report generated by the method can be used as part of the medical record.

[0103] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it is apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Therefore, the specification and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A method for generating and processing a mental illness image report, characterized in that: The method comprises the following steps: (1) Based on the original images of the subjects, evaluate the overall image resolution, noise, and grayscale deviation level; (2) Calculate the comprehensive quality rating of the image based on the evaluation results; (3) The coronal, sagittal, and transverse sections of the original brain images and the coronal, sagittal, and transverse sections of the brain tissue probability map are captured as the visualization part of the image quality assessment report; The method further includes the step of diagnosing the association between abnormalities in various brain regions and mental illness, specifically including the following processing steps: (2-1) For each brain structural region, report the numerical values ​​of image analysis indicators, the variation range of healthy people, and the abnormal distribution map of imaging indicators; (2-2) indicates values ​​outside the normal range; (2-3) show the degree of association between abnormal features and each diagnostic type, expressed as odds ratio; The method further comprises the step of generating a key brain image feature distribution map, which specifically comprises the following processing steps: (3-1) Calculate the effect size of each neuroimaging feature for identifying the diagnostic type of mental illness, and extract the features with the highest effect size; (3-2) extracting the offset of the test data relative to the healthy population on the above-mentioned features; (3-3) Adjust the population percentiles according to the effect size and brain region indicators; (3-4) Providing a multivariate display to present an overview of the key characteristic abnormalities of a disease, indicating the overall risk; The method further includes a step of comorbidity diagnosis, which specifically includes the following processing: (1-1) Calculate the identification results of each disease separately; (1-2) Analyze comorbidities based on the identification results of a single disease; (1-3) Use the probability threshold as the identification standard for a single disease to determine whether the test data is identified as more than two diseases. If so, report the comorbidity; otherwise, report the multi-class prediction probability between disease categories; The step (3-3) is specifically as follows: Determine whether the effect size is positive for the brain region indicator. If so, keep the population percentile unchanged; otherwise, convert the population percentile according to the following formula: Population percentile = 1 - original population percentile.

2. The method for generating and processing a mental illness image report according to claim 1, characterized in that: The comorbidity diagnosis step also includes the following processing: Based on the current samples submitted for examination, the predicted probability of binary diseases between any two categories of healthy people and common mental illnesses is provided.

3. A device for realizing the processing function of mental illness image report, characterized in that: The device comprises: a processor configured to execute computer executable instructions; A memory storing one or more computer executable instructions, wherein when the computer executable instructions are executed by the processor, the steps of the method for generating and processing a mental illness image report described in any one of claims 1 to 2 are implemented.

4. A processor for implementing the generation and processing of a mental illness image report, characterized in that: The processor is configured to execute computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for generating and processing a mental illness image report described in any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the method for generating and processing a mental illness image report as described in any one of claims 1 to 2.

Images