Early cognitive impairment screening method and system based on multi-mode electromagnetic field biological detection

Abstract

The invention belongs to the technical field of artificial intelligence, and particularly relates to an early cognitive impairment screening method and system based on multi-mode electromagnetic field biological detection.The early cognitive impairment screening method comprises the steps that a first electrode plate and a second electrode plate are arranged at the forehead position and the occipital position of the brain of a user to be detected in advance; the third electrode plate and the fourth electrode plate are arranged on the left side of the brain side by side, and the fifth electrode plate and the sixth electrode plate are arranged on the right side of the brain side by side. And obtaining a disturbance coefficient set of the to-be-detected user under the reference detection mode combination, inputting the disturbance coefficient set into a pre-trained early cognitive impairment screening model, and predicting to obtain a prediction probability value of the early cognitive impairment of the to-be-detected user, and by adopting the scheme to screen the early cognitive impairment, the accuracy is high, the universality is strong, and the accuracy is high. The method can be applied to screening of people in communities, nursing homes and the like on a large scale.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and specifically relates to an early cognitive impairment screening method and system based on multi-mode electromagnetic field biodetection. Background Art

[0002] Mild cognitive impairment (MCI) refers to a syndrome of cognitive impairment in a transitional stage between normal aging and dementia, in which individuals have significant impairment of memory and / or other cognitive functions compared with a control group of individuals matched in age and education, but their daily living abilities are normal and do not meet the usual diagnostic criteria for dementia. In-depth research on MCI will help to identify and screen high-risk groups for dementia and provide the best time window for the treatment of dementia.

[0003] MCI can be divided into degenerative, vascular and somatic types according to the cause; MCI can be divided into white matter lesions (MCI with white matter hyperintensities) and hippocampal atrophy (MCI with hippocampal atrophy) according to neuroimaging manifestations; MCI can also be divided into progressive MCI and stable MCI according to the changes in the disease.

[0004] Currently, the world is looking for simple and effective methods for MCI screening. The main methods are as follows: 1. Scale method; 2. Biological testing; 3. Imaging testing.

[0005] The above methods can achieve screening and diagnosis to a certain extent, but they all have their shortcomings, such as long time, consumption of expert resources, high cost, limited accuracy, etc., and they cannot be used as routine means for large-scale screening. Each of the above methods has many fatal shortcomings, such as rapid, non-invasive, digital, portable, low-cost, accurate sensitivity, etc. The above characteristics are the main content of early screening pursued by the medical community in the future. Electromagnetic field biodetection technology is currently a cutting-edge technology in the world. In particular, the application of this technology in the detection of craniocerebral diseases is an important direction in the field of neurology in the future. At present, the application of electromagnetic field biodetection technology has made progress in realizing a relatively rough detection method for craniocerebral diseases, such as using detection parameters, such as disturbance coefficient, bioimpedance coefficient and other parameters, to realize abnormal detection of the whole brain caused by various diseases, and then diagnose through CT / MRI and other methods, and determine the treatment plan based on the diagnosis results.

[0006] For example, the applicant has proposed a device for monitoring hydrocephalus and cerebral edema based on electromagnetic field biodetection technology (see CN102525458B). For another example, the applicant has also proposed a method and device for processing data on cranial brain abnormalities caused by cerebral infarction (see CN117688430A), which respectively sets electrodes at the forehead, occipital position, left front side of the cranium, left rear side of the cranium, right front side of the cranium and right rear side of the cranium, generates disturbance coefficients between the electrodes, and determines the abnormality of the subject's cerebral infarction according to the disturbance coefficients. Specifically, it performs machine learning by pre-collecting disturbance coefficient data sets of healthy subjects and abnormal subjects with abnormalities in the corresponding areas due to early cerebral infarction under a preset detection mode combination, thereby obtaining a multi-modal abnormality prediction model, and then inputs the collected disturbance coefficient data set under the preset detection mode combination into the multi-modal abnormality prediction model to obtain the corresponding prediction results, thereby realizing early prediction of abnormalities caused by early cerebral infarction, etc.

[0007] Although the above schemes provide abnormal prediction schemes for some brain diseases, they are not suitable for abnormal prediction of early cognitive impairment. Summary of the invention

[0008] The purpose of the present invention is to provide a method and system for screening early cognitive impairment based on multi-modal electromagnetic field biodetection, so as to partially solve or alleviate the deficiencies in the prior art and meet the needs of large-scale early cognitive screening in communities, welfare institutions, elderly care service centers and other places. In order to solve the above-mentioned technical problems, the present invention specifically adopts the following technical solutions: The first aspect of the present invention is to provide a method for screening early cognitive impairment based on multi-mode electromagnetic field biodetection, which comprises: S101 pre-arranges a first electrode sheet and a second electrode sheet at the forehead and occipital position of the user to be detected, respectively, and arranges a third electrode sheet and a fourth electrode sheet in parallel on the left side of the brain, and arranges a fifth electrode sheet and a sixth electrode sheet in parallel on the right side of the brain; S102 obtains a disturbance coefficient set of the user to be detected under the reference detection mode combination, and inputs the disturbance coefficient set into a pre-trained early cognitive impairment screening model to predict the predicted probability value of the user to be detected having early cognitive impairment; The perturbation coefficient set under the reference detection mode combination includes: the first perturbation coefficient set and the second perturbation coefficient set under the preset first four-electrode detection mode, the second four-electrode detection mode, and the third to sixth perturbation coefficient sets under the first to fourth three-electrode detection modes; In the first four-electrode detection mode, the first electrode sheet located at the forehead is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the third electrode sheet and the fifth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the first disturbance coefficient set includes: the seventh disturbance coefficient R7, the fifth disturbance coefficient R5 and the sixth disturbance coefficient R6 on the propagation path between the first electrode sheet and the second electrode sheet, the third electrode sheet and the fifth electrode sheet respectively; In the second four-electrode detection mode, the second electrode sheet located at the occipital position is used as an excitation transmitting end, the first electrode sheet located at the forehead position and the fourth electrode sheet and the sixth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the second disturbance coefficient set includes: the eighth disturbance coefficient R7', the third disturbance coefficient R3 and the fourth disturbance coefficient R4 on the propagation path between the second electrode sheet and the first electrode sheet, the fourth electrode sheet and the sixth electrode sheet respectively; In the first three-electrode detection mode, the third electrode sheet located on the left side of the skull is used as an excitation transmitting end, the first electrode sheet located at the forehead and the fifth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the three disturbance coefficient sets include: the twelfth disturbance coefficient R5' and the first disturbance coefficient R1 on the propagation path between the third electrode sheet and the first electrode sheet and the fifth electrode sheet respectively; In the second three-electrode detection mode, the fourth electrode sheet located on the left side of the skull is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the sixth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the four disturbance coefficient sets include: the tenth disturbance coefficient R3' and the second disturbance coefficient R2 on the propagation path between the fourth electrode sheet and the second electrode sheet and the sixth electrode sheet respectively; In the third three-electrode detection mode, the fifth electrode sheet located on the right side of the skull is used as an excitation transmitting end, and the first electrode sheet located at the forehead and the third electrode sheet located on the left side of the skull are used as excitation receiving ends. Accordingly, the five disturbance coefficient sets include: the thirteenth disturbance coefficient R6' and the eighth disturbance coefficient R1' on the propagation path between the fifth electrode sheet and the first electrode sheet and the third electrode sheet respectively; In the fourth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends, and accordingly, the six disturbance coefficient sets include: the eleventh disturbance coefficient R4' and the ninth disturbance coefficient R2' on the propagation path between the sixth electrode sheet and the second electrode sheet and the fourth electrode sheet respectively; The step of training the early cognitive impairment screening model specifically includes: S201 obtains a health disturbance coefficient data set of multiple healthy subjects under the combination of the benchmark detection modes; S202 obtaining a data set of abnormal disturbance coefficients of a plurality of subjects with early cognitive impairment under the combination of the benchmark detection modes; S203 obtaining scale test results of the plurality of early cognitive impairment subjects obtained through testing using a designated scale method; S204: Input the healthy disturbance coefficient data set, the abnormal disturbance coefficient data set and the scale detection result into a pre-built neural network model for learning to obtain the early cognitive impairment screening model.

[0009] Preferably, the step of training to obtain the early cognitive impairment screening model further comprises: S205: obtaining and identifying relevant past medical histories of the plurality of subjects with early cognitive impairment; the relevant past medical histories comprising at least one of hypertension, stroke, cerebral infarction and genetic factors; S206a If there are two or more previous medical histories, obtain the disturbance coefficient set of the subject under the reference detection mode combination, and input it, the health disturbance coefficient data set and the scale detection result into a pre-constructed neural network model for training to obtain a first prediction model; If S206b is a single past medical history, obtain the disturbance coefficient sets of the subject under the fifth to eighth three-electrode detection modes, and the disturbance coefficient set obtained under the baseline detection mode combination, and input them together with the health disturbance coefficient data set and the scale detection results into a pre-constructed neural network model for training to obtain a second prediction model.

[0010] Preferably, step S102 specifically includes: Obtaining and identifying relevant past medical history of the user to be detected; If there are two or more previous medical histories, obtaining a disturbance coefficient set of the user to be detected under the reference detection mode combination, and then inputting the set into the first prediction model for prediction; If it is a single past medical history, the disturbance coefficient set of the user to be detected under the reference detection mode combination and the disturbance coefficient set under the fifth to eighth three-electrode detection modes are obtained respectively, and input into the second prediction model for prediction.

[0011] Preferably, in the fifth three-electrode detection mode, the third electrode sheet located on the left side of the skull serves as an excitation transmitting end, the first electrode sheet located on the forehead and the sixth electrode sheet located on the right side of the skull serve as excitation receiving ends. Accordingly, the seventh disturbance coefficient set in the fifth three-electrode detection mode includes: the fifteenth disturbance coefficient R5' and the sixteenth disturbance coefficient R16 on the propagation path between the third electrode sheet and the first electrode sheet and the sixth electrode sheet respectively.

[0012] Preferably, in the sixth three-electrode detection mode, the fourth electrode sheet located on the left side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fifth electrode sheet located on the right side of the skull serve as excitation receiving ends. Accordingly, the eighth disturbance coefficient set in the sixth three-electrode detection mode includes: the tenth disturbance coefficient R3' and the seventeenth disturbance coefficient R17 on the propagation path between the fourth electrode sheet and the second electrode sheet and the fifth electrode sheet respectively.

[0013] Preferably, in the seventh three-electrode detection mode, the fifth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the first electrode sheet located on the forehead and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends, and accordingly, the ninth disturbance coefficient set in the seventh three-electrode detection mode includes: the thirteenth disturbance coefficient R6' and the eighteenth disturbance coefficient R18 on the propagation path between the fifth electrode sheet and the first electrode sheet and the fourth electrode sheet respectively.

[0014] Preferably, in the eighth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode located at the occipital position and the third electrode sheet located on the left side of the skull serve as excitation receiving ends, and accordingly, the tenth disturbance coefficient set in the eighth three-electrode detection mode includes: the eleventh disturbance coefficient R4' and the nineteenth disturbance coefficient R19 on the propagation path between the sixth electrode sheet and the second electrode sheet and the third electrode sheet respectively.

[0015] Preferably, if the subject to be tested has a previous medical history of cerebral stroke, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the four disturbance coefficient sets in the first and second three-electrode detection modes is greater than the ratio of the sum of the five disturbance coefficient sets and the six disturbance coefficient sets in the third and fourth three-electrode detection modes; If the subject to be tested has a past medical history of cerebral infarction, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the five disturbance coefficient sets in the first and third three-electrode detection modes is greater than the ratio of the sum of the four disturbance coefficient sets and the six disturbance coefficient sets in the second and fourth three-electrode detection modes.

[0016] The second aspect of the present invention is an early cognitive impairment screening system based on multi-mode electromagnetic field biodetection, which comprises: The data acquisition device includes a data communication module and an electrode group for users to obtain data, wherein the electrode group includes a first electrode sheet and a second electrode sheet corresponding to the forehead position and the occipital position of the skull, respectively, a third electrode sheet and a fourth electrode sheet arranged in parallel corresponding to the left side of the skull, and a fifth electrode sheet and a sixth electrode sheet arranged in parallel corresponding to the right side of the skull; A screening device, which communicates with the data acquisition device to obtain a disturbance coefficient set of the data acquisition device under a preset reference detection mode combination, and inputs the disturbance coefficient set into a pre-trained early cognitive impairment screening model to predict the predicted probability value of the early cognitive impairment of the user to be detected; Specifically, the screening device comprises: A first data acquisition module is used to communicate with the data acquisition device to respectively acquire a plurality of healthy subjects' health disturbance coefficient data sets under the reference detection mode combination; and respectively acquire a plurality of early cognitive impairment subjects' abnormal disturbance coefficient data sets under the reference detection mode combination; and acquire a disturbance coefficient set of the user to be detected under the reference detection mode combination; A second data acquisition module is used to obtain the scale test results of the plurality of early cognitive impairment subjects obtained by the specified scale method test; A screening model construction module, used to train a pre-constructed neural network model based on the healthy disturbance coefficient data set, the abnormal disturbance coefficient data set and the scale detection result to obtain the early cognitive impairment screening model; A prediction module, configured to make a prediction based on the early cognitive impairment screening model constructed by the screening model construction module and the disturbance coefficient set of the user to be detected under the reference detection mode combination acquired by the first data acquisition module, to obtain a predicted probability value of the user to be detected having early cognitive impairment; The perturbation coefficient set under the reference detection mode combination includes: the first perturbation coefficient set and the second perturbation coefficient set under the preset first four-electrode detection mode, the second four-electrode detection mode, and the third to sixth perturbation coefficient sets under the first to fourth three-electrode detection modes; In the first four-electrode detection mode, the first electrode located at the forehead is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the third electrode sheet and the fifth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the first disturbance coefficient set includes: the seventh disturbance coefficient R7, the fifth disturbance coefficient R5 and the sixth disturbance coefficient R6 on the propagation path between the first electrode sheet and the second electrode sheet, the third electrode sheet and the fifth electrode sheet respectively; In the second four-electrode detection mode, the second electrode sheet located at the occipital position is used as an excitation transmitting end, the first electrode sheet located at the forehead position and the fourth electrode sheet and the sixth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the second disturbance coefficient set includes: the eighth disturbance coefficient R7', the third disturbance coefficient R3 and the fourth disturbance coefficient R4 on the propagation path between the second electrode sheet and the first electrode sheet, the fourth electrode sheet and the sixth electrode sheet respectively; In the first three-electrode detection mode, the third electrode sheet located on the left side of the skull is used as an excitation transmitting end, and the first electrode located at the forehead and the fifth electrode sheet located on the right side of the skull are used as excitation receiving ends. Accordingly, the three disturbance coefficient sets include: the twelfth disturbance coefficient R5' and the first disturbance coefficient R1 on the propagation path between the third electrode sheet and the first electrode sheet and the fifth electrode sheet respectively; In the second three-electrode detection mode, the fourth electrode sheet located on the left side of the skull is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the sixth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the four disturbance coefficient sets include: the tenth disturbance coefficient R3' and the second disturbance coefficient R2 on the propagation path between the fourth electrode sheet and the second electrode sheet and the sixth electrode sheet respectively; In the third three-electrode detection mode, the fifth electrode sheet located on the right side of the skull is used as an excitation transmitting end, and the first electrode sheet located at the forehead and the third electrode sheet located on the left side of the skull are used as excitation receiving ends. Accordingly, the five disturbance coefficient sets include: the thirteenth disturbance coefficient R6' and the eighth disturbance coefficient R1' on the propagation path between the fifth electrode sheet and the first electrode sheet and the third electrode sheet respectively; In the fourth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the six disturbance coefficient sets include: the eleventh disturbance coefficient R4' and the ninth disturbance coefficient R2' on the propagation path between the sixth electrode sheet and the second electrode sheet and the fourth electrode sheet respectively.

[0017] Preferably, the screening device further comprises: A third data acquisition module is used to acquire and identify the associated past medical history of the multiple early cognitive impairment subjects; the associated past medical history includes at least one of hypertension, stroke, cerebral infarction and genetic factors; a fourth data acquisition module, for acquiring the perturbation coefficient set of the subject under the reference detection mode combination when the subject has two or more related past medical histories, and for acquiring the perturbation coefficient set of the subject under the fifth to eighth three-electrode detection modes respectively when the subject has a single past medical history, Correspondingly, the screening model construction module is also used to, when the subject has two or more related past medical histories, input the perturbation coefficient set of the subject under the baseline detection mode combination, the healthy perturbation coefficient data set and the scale detection result into the model for training to obtain a first prediction model; when the subject has a single past medical history, input the perturbation coefficient set of the subject under the fifth to eighth three-electrode detection modes, the perturbation coefficient set obtained under the baseline detection mode combination, the healthy perturbation coefficient data set and the scale detection result into the model for training to obtain a second prediction model.

[0018] The principle and beneficial technical effects of the present invention are: In the prior art (see CN117688430A), electrodes are respectively arranged at the forehead, occipital region, left front side of the skull, left rear side of the skull, right front side of the skull and right rear side of the skull, disturbance coefficients are generated between the electrodes, and predictions are made in combination with the default detection mode combination to obtain the probability of cerebral infarction in the user to be detected. However, due to the complexity of human brain tissue and clear division of labor, there are differences in the lesion locations (i.e., detection locations) that induce different brain diseases. The above-mentioned device is not suitable for screening of early cognitive impairment. In addition, the scheme performs indiscriminate detection of users through a fixed mode (default detection mode combination) and is not suitable for large-scale detection.

[0019] Based on this, the present application proposes an early cognitive impairment screening method and system with multi-modal electromagnetic field biodetection with parallel collection points for early cognitive impairment, which has the characteristics of high accuracy and strong universality.

[0020] On the one hand, this scheme sets two groups of electrodes in parallel in the parts corresponding to the left brain and the right brain, and sets multiple detection modes based on multiple electrodes (corresponding to electrodes in different positions and different electrode propagation paths), and detects the users to be detected separately to obtain a multivariate disturbance data set, thereby predicting the probability of the user suffering from early cognitive impairment. The samples are rich and the prediction results are highly reliable.

[0021] On the other hand, the possible influencing factors of the onset of early cognitive impairment (associated with past medical history; elderly people who usually have or have suffered from hypertension, stroke, genetic disease history, etc., have a higher risk of early cognitive impairment) are introduced into the model training process, and the model is divided into a first model and a second model. For users with a history of more past diseases and a higher risk of illness, the first model with a smaller amount of data (measured by fewer detection modes) is used for prediction, which can effectively reduce the amount of calculation, and for users with a single medical history with a lower prevalence, the second model with a larger amount of data (measured by more detection modes) is used for prediction, which improves the accuracy of the prediction. In other words, the present application effectively improves the work efficiency of the prediction process by distinguishing user categories and quickly matching the corresponding prediction models, and compared with equipment such as CT, ultrasound and MRI, it has almost no negative impact on the user's body and can be continuously monitored (for example, for several hours or even for several consecutive days). Therefore, it can be widely used in public medical systems such as communities. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale. Obviously, the drawings described below are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without paying creative labor.

[0023] Figure 1 This is a flow chart of the method for screening early cognitive impairment in Embodiment 1 of the present invention; Figure 2a is a schematic diagram of a first four-electrode detection mode in Embodiment 1 of the present invention; Figure 2b is a schematic diagram of the second four-electrode detection mode in the first embodiment of the present invention; Figure 3a is a schematic diagram of a first three-electrode detection mode in Embodiment 1 of the present invention; Figure 3b is a schematic diagram of the second three-electrode detection mode in the first embodiment of the present invention; Figure 3c is a schematic diagram of the third three-electrode detection mode in the first embodiment of the present invention; Figure 3d is a schematic diagram of a fourth three-electrode detection mode in the first embodiment of the present invention; Figure 3e is a schematic diagram of the fifth three-electrode detection mode in the first embodiment of the present invention; Figure 3fis a schematic diagram of the sixth three-electrode detection mode in the first embodiment of the present invention; Figure 3g is a schematic diagram of the seventh three-electrode detection mode in the first embodiment of the present invention; Figure 3h is a schematic diagram of an eighth three-electrode detection mode in the first embodiment of the present invention; Figure 4a is a schematic diagram of a first four-electrode grounding detection mode in Embodiment 2 of the present invention; Figure 4b is a schematic diagram of the second four-electrode grounding detection mode in the second embodiment of the present invention; Figure 5a is a schematic diagram of a first three-electrode grounding detection mode in Embodiment 2 of the present invention; Figure 5b Schematic diagram of the second three-electrode grounding detection mode in the second embodiment of the present invention; Figure 5c is a schematic diagram of the third three-electrode grounding detection mode in the second embodiment of the present invention; Figure 5d Schematic diagram of the fourth third-electrode grounding detection mode in the second embodiment of the present invention; Figure 6 This is a schematic diagram of the modular structure of the early cognitive impairment screening system in Embodiment 3 of the present invention; Figures 7 to 12 The present invention provides the prediction results of clinical trials using the early cognitive impairment screening system based on multi-mode electromagnetic field biodetection, as well as the corresponding verified clinical cases obtained through CT diagnosis / MRI diagnosis.

[0024] Markings in the figure: 1, first electrode sheet; 2, second electrode sheet; 3, third electrode sheet; 4, fourth electrode sheet; 5, fifth electrode sheet; 6, sixth electrode sheet. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] Herein, the suffixes such as "module", "component" or "unit" used to represent elements are used only to facilitate the description of the present invention and have no specific meaning in themselves. Therefore, "module", "component" or "unit" can be used in a mixed manner. Herein, the orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", "front", "back", "one end", "the other end", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0027] In this document, unless otherwise clearly specified and limited, the terms "installed", "provided with", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In this document, "plurality" means two or more, that is, it includes two, three, four, five, etc.

[0028] In this article, "abnormality" means that when an excitation signal of a specific frequency is used to transmit an excitation signal from the transmitting end, and it reaches the receiving end through a specific transmission path (i.e., the propagation path between the transmitting end and the receiving end), due to the presence of diseased tissue corresponding to a certain disease in the transmission path, the detection signal received by the receiving end is different from the detection signal (i.e., electromagnetic field biological signal, including receiving impedance, transmission differential impedance and disturbance coefficient, etc.) received when there is no diseased tissue in the transmission path or when there is diseased tissue corresponding to other diseases. Therefore, it is called abnormality.

[0029] In this document, "excitation signal" includes a sinusoidal curve having a frequency selected from a specified frequency range; or, a pulse or pulse sequence having a pulse width and a pulse repetition frequency selected from a specified pulse width range and a specified pulse repetition frequency range; or a coded signal, such as a linear frequency modulation signal, which has a frequency that varies with time to obtain a tissue response within a specified frequency range and minimize interference, multipath effects, and radio frequency (RF) noise and other noise.

[0030] The "detection mode" in this article refers to the electrode patch at a specified position acting as an excitation transmitter to transmit an excitation signal of a specific frequency range to the subject's brain to obtain the disturbance coefficient of a specific area.

[0031] Embodiment 1: This embodiment provides a method for screening early cognitive impairment using multi-mode electromagnetic field biodetection, see Figure 1-Figure 3h ,include: S101 pre-sets the first electrode sheet 1 and the second electrode sheet 2 on the forehead and the occipital position of the user to be tested, respectively, and sets the third electrode sheet 3 and the fourth electrode sheet 4 in parallel on the left side of the skull, and sets the fifth electrode sheet 5 and the sixth electrode sheet 6 in parallel on the right side of the skull.

[0032] S102 obtains a disturbance coefficient set of the user to be detected under a combination of reference detection modes, and inputs the disturbance coefficient set into a pre-trained early cognitive impairment screening model to predict a predicted probability value of the user to be detected having early cognitive impairment.

[0033] In this embodiment, the perturbation coefficient set under the reference detection mode combination includes: the first perturbation coefficient set and the second perturbation coefficient set under the preset first four-electrode detection mode, the second four-electrode detection mode, and the third to sixth perturbation coefficient sets under the first to fourth three-electrode detection modes.

[0034] Among them, in the first four-electrode detection mode, the first electrode located at the forehead serves as the excitation transmitting end, the second electrode located at the occipital position and the third electrode and the fifth electrode located on the left side and the right side of the skull respectively serve as the excitation receiving ends. Accordingly, the first disturbance coefficient set includes: the seventh disturbance coefficient R7, the fifth disturbance coefficient R5 and the sixth disturbance coefficient R6 on the propagation path between the first electrode and the second electrode, the third electrode and the fifth electrode respectively.

[0035] Among them, in the second four-electrode detection mode, the second electrode sheet located at the occipital position serves as the excitation transmitting end, the first electrode sheet located at the forehead position and the fourth electrode sheet and the sixth electrode sheet located on the left side and the right side of the skull respectively serve as the excitation receiving ends, and accordingly, the second disturbance coefficient set includes: the eighth disturbance coefficient R7', the third disturbance coefficient R3 and the fourth disturbance coefficient R4 on the propagation path between the second electrode sheet and the first electrode sheet, the fourth electrode sheet and the sixth electrode sheet respectively.

[0036] Among them, in the first three-electrode detection mode, the third electrode located on the left side of the brain serves as the excitation transmitting end, the first electrode located at the forehead and the fifth electrode located on the right side of the brain serve as the excitation receiving ends, and accordingly, the three disturbance coefficient sets include: the twelfth disturbance coefficient R5' and the first disturbance coefficient R1 on the propagation path between the third electrode and the first electrode and the fifth electrode respectively.

[0037] Among them, in the second three-electrode detection mode, the fourth electrode located on the left side of the skull serves as the excitation transmitting end, the second electrode located at the occipital position and the sixth electrode located on the right side of the skull serve as the excitation receiving ends, and accordingly, the four disturbance coefficient sets include: the tenth disturbance coefficient R3' and the second disturbance coefficient R2 on the propagation path between the fourth electrode and the second electrode and the sixth electrode respectively.

[0038] Among them, in the third three-electrode detection mode, the fifth electrode located on the right side of the brain serves as the excitation transmitting end, the first electrode located at the forehead and the third electrode located on the left side of the brain serve as the excitation receiving ends, and accordingly, the five disturbance coefficient sets include: the thirteenth disturbance coefficient R6' and the eighth disturbance coefficient R1' on the propagation path between the fifth electrode and the first electrode and the third electrode respectively.

[0039] Among them, in the fourth three-electrode detection mode, the sixth electrode located on the right side of the skull serves as the excitation transmitting end, the second electrode located at the occipital position and the fourth electrode located on the left side of the skull serve as the excitation receiving ends, and accordingly, the six disturbance coefficient sets include: the eleventh disturbance coefficient R4' and the ninth disturbance coefficient R2' on the propagation path between the sixth electrode and the second electrode and the fourth electrode respectively.

[0040] The step of training the early cognitive impairment screening model specifically includes: S201 obtains a health disturbance coefficient data set of multiple healthy subjects under the combination of the benchmark detection modes; S202 obtaining a data set of abnormal disturbance coefficients of a plurality of subjects with early cognitive impairment under the combination of the benchmark detection modes; S203 obtaining scale test results of the plurality of early cognitive impairment subjects obtained through testing using a designated scale method; S204: Input the healthy disturbance coefficient data set, the abnormal disturbance coefficient data set and the scale detection result into a pre-built neural network model for learning to obtain the early cognitive impairment screening model.

[0041] It is worth noting that for different brain diseases, the locations of lesions or diseased tissues vary greatly, and in the process of electromagnetic field biological detection, setting up transmission and reception at different locations will produce different detection data. Therefore, slight differences in detection locations may have a great impact on the detection results.

[0042] In view of the characteristics of the location of the lesions or pathological tissues of early cognitive impairment, the present application selects two electrode sheets at the forehead and occipital positions of the skull, and sets two electrode sheets in parallel on the left and right sides of the skull, respectively. A model is constructed in advance according to the interference coefficient of the early cognitive impairment population, so that in subsequent detections, it is only necessary to use a preset benchmark detection mode to detect the user to be detected to obtain the corresponding interference coefficient set, and the predicted probability of the user having early cognitive impairment can be obtained, so that users who may have early cognitive impairment can be screened out according to the predicted probability, and reminded to go to a medical institution for timely treatment for diagnosis. For example, if the predicted probability is greater than the preset threshold range, it will be marked and reminded to go to a medical institution for timely treatment for diagnosis. Specifically, for example, if the predicted probability of illness belongs to the first preset threshold range (for example, the predicted probability ≥ 90%), it means that it has a risk of early cognitive impairment; if it is less than the minimum value of the first preset threshold range, it means that it is risk-free.

[0043] Furthermore, if the predicted probability is less than the minimum value of the first preset threshold range, and the difference between the predicted probability and the minimum value is less than the preset difference value, it means that the user may actually have early cognitive impairment, but it is difficult to detect. Therefore, the test is performed for multiple consecutive days to obtain multiple test results, and the average value of the multiple test results is compared with the first preset threshold range. If it belongs to the first preset threshold range, it means that the user has the risk of early cognitive impairment; if it is less than the minimum value of the first preset threshold range, it means that the user has no risk. Preferably, if the volatility of multiple test results is greater than the preset volatility threshold, the average value is compared with the first preset threshold range, otherwise, it is determined that the user has no risk.

[0044] In some embodiments, the step of training to obtain the early cognitive impairment screening model further comprises: S205: obtaining and identifying relevant past medical histories of the plurality of subjects with early cognitive impairment; the relevant past medical histories comprising at least one of hypertension, stroke, cerebral infarction and genetic factors; S206a If there are two or more previous medical histories, obtain the disturbance coefficient set of the subject under the reference detection mode combination, and input it, the health disturbance coefficient data set and the scale detection result into a pre-constructed neural network model for training to obtain a first prediction model; If S206b is a single past medical history, obtain the disturbance coefficient sets of the subject under the fifth to eighth three-electrode detection modes, and the disturbance coefficient set obtained under the baseline detection mode combination, and input them together with the health disturbance coefficient data set and the scale detection results into a pre-constructed neural network model for training to obtain a second prediction model.

[0045] Accordingly, S102 specifically includes: Obtain and identify relevant past medical history of the user to be tested; If there are two or more previous medical histories, a disturbance coefficient set of the user to be detected under the reference detection mode combination is obtained, and then input into the first prediction model for prediction.

[0046] In some embodiments, when the predicted probability value detected by a user with two or more past medical histories meets the first preset threshold range, the detection is terminated. Of course, if the predicted probability value is not within the preset threshold range, the perturbation coefficient set of the user to be detected in the fifth to eighth three-electrode detection modes can also be obtained, and it is input into the second prediction model together with the perturbation coefficient set under the reference detection mode combination for secondary prediction, and the predicted probability value predicted by the second prediction model is obtained, and the predicted value is determined as the prediction result. For example, if the predicted probability value is greater than 90% (the minimum value in the first preset threshold range), it is determined that the patient has a risk of illness, and further diagnosis is required for confirmation.

[0047] The preset range threshold can be obtained by calculating the probability of early cognitive impairment in the elderly group with a medical history based on a pre-statisticed medical database. Different medical histories correspond to different probability ranges. When there are multiple medical histories, the probability of early cognitive impairment in multiple medical histories is obtained by combining the disease probabilities of the multiple medical histories. Different combinations of medical histories correspond to different probability ranges. Generally speaking, the more types of medical histories, the higher the probability of early cognitive impairment.

[0048] If it is a single past medical history, the disturbance coefficient set of the user to be detected under the reference detection mode combination and the disturbance coefficient set under the fifth to eighth three-electrode detection modes are obtained respectively, and input into the second prediction model for prediction.

[0049] The relevant past medical history of the user to be detected can be directly extracted from a database pre-collected by a hospital or other institution, or can be directly obtained from the user to be detected through a questionnaire survey or on-site questioning.

[0050] This solution divides users into a single medical history group and a multiple medical history group in advance, and builds corresponding prediction models in advance according to user types, which can effectively reduce the amount of calculation while ensuring prediction accuracy.

[0051] In some embodiments, in the fifth three-electrode detection mode, the third electrode sheet located on the left side of the skull serves as an excitation transmitting end, the first electrode located at the forehead and the sixth electrode sheet located on the right side of the skull serve as excitation receiving ends, and accordingly, the seventh disturbance coefficient set in the fifth three-electrode detection mode includes: the fifteenth disturbance coefficient R5' and the sixteenth disturbance coefficient R16 on the propagation path between the third electrode sheet and the first electrode sheet and the sixth electrode sheet respectively.

[0052] In some embodiments, in the sixth three-electrode detection mode, the fourth electrode sheet located on the left side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fifth electrode sheet located on the right side of the skull serve as excitation receiving ends. Accordingly, the eighth disturbance coefficient set in the sixth three-electrode detection mode includes: the tenth disturbance coefficient R3' and the seventeenth disturbance coefficient R17 on the propagation path between the fourth electrode sheet and the second electrode sheet and the fifth electrode sheet respectively.

[0053] In some embodiments, in the seventh three-electrode detection mode, the fifth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the first electrode located at the forehead and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the ninth disturbance coefficient set in the seventh three-electrode detection mode includes: the thirteenth disturbance coefficient R6' and the eighteenth disturbance coefficient R18 on the propagation path between the fifth electrode sheet and the first electrode sheet and the fourth electrode sheet respectively.

[0054] In some embodiments, in the eighth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode located at the occipital position and the third electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the tenth disturbance coefficient set in the eighth three-electrode detection mode includes: the eleventh disturbance coefficient R4' and the nineteenth disturbance coefficient R19 on the propagation path between the sixth electrode sheet and the second electrode sheet and the third electrode sheet respectively.

[0055] In some embodiments, if the subject to be tested has a past medical history of stroke, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the four disturbance coefficient sets in the first and second three-electrode detection modes is greater than the ratio of the sum of the five disturbance coefficient sets and the six disturbance coefficient sets in the third and fourth three-electrode detection modes.

[0056] If the subject to be tested has a past medical history of cerebral infarction, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the five disturbance coefficient sets in the first and third three-electrode detection modes is greater than the ratio of the sum of the four disturbance coefficient sets and the six disturbance coefficient sets in the second and fourth three-electrode detection modes.

[0057] Adjusting the proportion of data sets under different modes according to high-risk lesions with different medical histories can effectively improve the detection rate. For example, stroke lesions on the left side of the brain have a high risk of cognitive impairment. Therefore, the detection proportion of the first and second three-electrode detection modes is greater than the detection proportion of the third and fourth three-electrode detection modes. For another example, when cerebral infarction occurs in the left hemisphere or anterior hemisphere, the risk of cognitive impairment in the front of the brain is high. Therefore, when collecting data, increasing the detection proportion of the first and third three-electrode detection modes is greater than the detection proportion of the second and fourth three-electrode detection modes.

[0058] Embodiment 2: Different from Embodiment 1, based on the early cognitive impairment screening method provided in Embodiment 1, this embodiment proposes a grounded reference detection mode combination for people with no previous medical history or with a small number of previous medical history (for example, less than or equal to 2 types). For details, see Figure 4a-Figure 5d , the ground reference detection mode combination includes: In a first four-electrode grounding detection mode, the first electrode sheet is used as an excitation transmitting end, the second electrode sheet on the opposite side of the first electrode sheet is used as a grounding end, and the third electrode sheet and the fifth electrode sheet are used as excitation receiving ends, to obtain an eleventh disturbance coefficient set, wherein the eleventh disturbance coefficient set includes: a fifth disturbance coefficient R5 and a sixth disturbance coefficient R6 on a propagation path between the first electrode sheet and the third electrode sheet and the fifth electrode sheet respectively; In a second four-electrode grounding detection mode, the second electrode sheet is used as an excitation transmitting end, the first electrode sheet is used as a grounding end, and the fourth electrode sheet and the sixth electrode sheet are used as excitation receiving ends, to obtain a twelfth disturbance coefficient set, wherein the twelfth disturbance coefficient set includes: a third disturbance coefficient R3 and a fourth disturbance coefficient R4 on a propagation path between the second electrode sheet and the fourth electrode sheet and the sixth electrode sheet respectively; In the first three-electrode grounding detection mode, the third electrode sheet is used as an excitation transmitting end, the fifth electrode sheet on the opposite side of the third electrode sheet is used as a grounding end, and the first electrode sheet is used as an excitation receiving end, to obtain a thirteenth disturbance coefficient set, wherein the thirteenth disturbance coefficient set includes: a twelfth disturbance coefficient R5' on the propagation path between the third electrode sheet and the first electrode sheet; In the second three-electrode grounding detection mode, the fourth electrode sheet is used as the excitation transmitting end, the sixth electrode sheet is used as the grounding end, and the second electrode sheet is used as the excitation receiving end, and a fourteenth disturbance coefficient set is obtained, wherein the fourteenth disturbance coefficient set includes: a tenth disturbance coefficient R3' on the propagation path between the fourth electrode sheet and the second electrode sheet; In the third three-electrode grounding detection mode, the fifth electrode sheet is used as the excitation transmitting end, the third electrode sheet is used as the grounding end, and the first electrode sheet is used as the excitation receiving end, and a fifteenth disturbance coefficient set is obtained, wherein the fifteenth disturbance coefficient set includes: a fifteenth disturbance coefficient R6' on the propagation path between the fifth electrode sheet and the first electrode sheet; In the fourth three-electrode grounding detection mode, the sixth electrode sheet is used as the excitation transmitting end, the fourth electrode sheet is used as the grounding end, and the second electrode sheet is used as the excitation receiving end to obtain the sixteenth disturbance coefficient set, which includes: the eleventh disturbance coefficient R4' on the propagation path between the sixth electrode sheet and the second electrode sheet.

[0059] Furthermore, compared with patients with a history of disease, patients without a history of disease are more difficult to be screened. Therefore, in order to avoid missed detection to a certain extent, when the predicted probability value predicted by the above screening method meets the second preset threshold range, a secondary screening is performed. Specifically, it includes the above-mentioned various detection modes, and the difference is that it also includes alternative modes: In the fifth three-electrode grounding detection mode, the sixth electrode sheet is used as the excitation transmitting end, the second electrode sheet is used as the grounding end, and the third electrode sheet is used as the excitation receiving end, and a seventeenth disturbance coefficient set is obtained, wherein the seventeenth disturbance coefficient set includes: a nineteenth disturbance coefficient R19 on the propagation path between the sixth electrode sheet and the third electrode sheet; In the sixth three-electrode grounding detection mode, the fourth electrode sheet is used as the excitation transmitting end, the second electrode sheet is used as the grounding end, and the fifth electrode sheet is used as the excitation receiving end, and an eighteenth disturbance coefficient set is obtained, wherein the eighteenth disturbance coefficient set includes: a seventeenth disturbance coefficient R17 on the propagation path between the fourth electrode sheet and the fifth electrode sheet; In the seventh three-electrode grounding detection mode, the third electrode sheet is used as the excitation transmitting end, the first electrode sheet is used as the grounding end, and the sixth electrode sheet is used as the excitation receiving end, and a nineteenth disturbance coefficient set is obtained, wherein the nineteenth disturbance coefficient set includes: a sixteenth disturbance coefficient R16 on the propagation path between the third electrode sheet and the sixth electrode sheet; In the eighth three-electrode grounding detection mode, the fifth electrode sheet is used as the excitation transmitting end, the first electrode sheet is used as the grounding end, and the fourth electrode sheet is used as the excitation receiving end to obtain the twentieth disturbance coefficient set, which includes: the eighteenth disturbance coefficient R18 on the propagation path between the fifth electrode sheet and the fourth electrode sheet.

[0060] Specifically, the second preset threshold range is smaller than the first preset threshold range. For example, the first preset threshold range is ≥90%; the second preset threshold range is: [85%, 90%).

[0061] Accordingly, the early cognitive impairment screening model obtained through training also includes: A disturbance coefficient set of multiple subjects with no history of early cognitive impairment under a grounded benchmark detection mode combination is obtained, and it is input into a pre-constructed neural network model for training together with the health disturbance coefficient data set and the scale detection results to obtain a third prediction model.

[0062] The grounded reference detection mode combination with the above-mentioned mode can meet the requirements of whole-brain detection while having a small amount of data and can be used and promoted on a large scale.

[0063] Example 3: See Figure 6 The present invention also provides an early cognitive impairment screening system based on multi-mode electromagnetic field biodetection, comprising: The data acquisition device includes a data communication module and an electrode group for users to obtain data, wherein the electrode group includes a first electrode sheet and a second electrode sheet corresponding to the forehead position and the occipital position of the skull, respectively, a third electrode sheet and a fourth electrode sheet arranged in parallel corresponding to the left side of the skull, and a fifth electrode sheet and a sixth electrode sheet arranged in parallel corresponding to the right side of the skull; A screening device communicates data with the data acquisition device to obtain a set of disturbance coefficients of the data acquisition device under a preset baseline detection mode combination, and inputs the set of disturbance coefficients into a pre-trained early cognitive impairment screening model to predict the predicted probability value of the user to be detected having early cognitive impairment.

[0064] Specifically, the screening device comprises: A first data acquisition module is used to communicate with the data acquisition device to respectively acquire a plurality of healthy subjects' health disturbance coefficient data sets under the reference detection mode combination; and respectively acquire a plurality of early cognitive impairment subjects' abnormal disturbance coefficient data sets under the reference detection mode combination; and acquire a disturbance coefficient set of the user to be detected under the reference detection mode combination; A second data acquisition module is used to obtain the scale test results of the plurality of early cognitive impairment subjects obtained by the specified scale method test; The screening model construction module is used to train the pre-constructed neural network model based on the healthy disturbance coefficient data set, the abnormal disturbance coefficient data set and the scale detection results to obtain the early cognitive impairment screening model.

[0065] A prediction module, configured to make a prediction based on the early cognitive impairment screening model constructed by the screening model construction module and the disturbance coefficient set of the user to be detected under the reference detection mode combination acquired by the first data acquisition module, to obtain a predicted probability value of the user to be detected having early cognitive impairment; The perturbation coefficient set under the reference detection mode combination includes: the first perturbation coefficient set and the second perturbation coefficient set under the preset first four-electrode detection mode, the second four-electrode detection mode, and the third to sixth perturbation coefficient sets under the first to fourth three-electrode detection modes; In the first four-electrode detection mode, the first electrode located at the forehead is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the third electrode sheet and the fifth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the first disturbance coefficient set includes: the seventh disturbance coefficient R7, the fifth disturbance coefficient R5 and the sixth disturbance coefficient R6 on the propagation path between the first electrode sheet and the second electrode sheet, the third electrode sheet and the fifth electrode sheet respectively; In the second four-electrode detection mode, the second electrode sheet located at the occipital position is used as an excitation transmitting end, the first electrode sheet located at the forehead position and the fourth electrode sheet and the sixth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the second disturbance coefficient set includes: the eighth disturbance coefficient R7', the third disturbance coefficient R3 and the fourth disturbance coefficient R4 on the propagation path between the second electrode sheet and the first electrode sheet, the fourth electrode sheet and the sixth electrode sheet respectively; In the first three-electrode detection mode, the third electrode sheet located on the left side of the skull is used as an excitation transmitting end, and the first electrode located at the forehead and the fifth electrode sheet located on the right side of the skull are used as excitation receiving ends. Accordingly, the three disturbance coefficient sets include: the twelfth disturbance coefficient R5' and the first disturbance coefficient R1 on the propagation path between the third electrode sheet and the first electrode sheet and the fifth electrode sheet respectively; In the second three-electrode detection mode, the fourth electrode sheet located on the left side of the skull is used as an excitation transmitting end, the second electrode located at the occipital position and the sixth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the four disturbance coefficient sets include: the tenth disturbance coefficient R3' and the second disturbance coefficient R2 on the propagation path between the fourth electrode sheet and the second electrode sheet and the sixth electrode sheet respectively; In the third three-electrode detection mode, the fifth electrode sheet located on the right side of the skull is used as an excitation transmitting end, and the first electrode located at the forehead and the third electrode sheet located on the left side of the skull are used as excitation receiving ends. Accordingly, the five disturbance coefficient sets include: the thirteenth disturbance coefficient R6' and the eighth disturbance coefficient R1' on the propagation path between the fifth electrode sheet and the first electrode sheet and the third electrode sheet respectively; In the fourth three-electrode detection mode, the sixth electrode located on the right side of the skull serves as an excitation transmitting end, the second electrode located at the occipital position and the fourth electrode located on the left side of the skull serve as excitation receiving ends. Accordingly, the six disturbance coefficient sets include: the eleventh disturbance coefficient R4' and the ninth disturbance coefficient R2' on the propagation path between the sixth electrode and the second electrode and the fourth electrode respectively.

[0066] In some embodiments, the screening device also includes a third data acquisition module, which is used to acquire and identify the associated past medical histories of the multiple early cognitive impairment subjects; the associated past medical histories include at least one of hypertension, stroke, cerebral infarction and genetic factors; a fourth data acquisition module, which is used to acquire the perturbation coefficient set of the subject under the benchmark detection mode combination when the subject has two or more associated past medical histories, and to acquire the perturbation coefficient sets of the subject under the fifth to eighth three-electrode detection modes when the subject has a single past medical history. Correspondingly, the screening model construction module is also used to input the perturbation coefficient set of the subject under the benchmark detection mode combination, the healthy perturbation coefficient data set and the scale detection result into the model for training to obtain a first prediction model when the subject has two or more associated past medical histories; when the subject has a single past medical history, the perturbation coefficient set of the subject under the fifth to eighth three-electrode detection modes, the perturbation coefficient set obtained under the benchmark detection mode combination, the healthy perturbation coefficient data set and the scale detection result are input into the model for training to obtain a second prediction model.

[0067] In some embodiments, when the past medical history of the subject to be tested is stroke, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the four disturbance coefficient sets under the first and second three-electrode detection modes is greater than the ratio of the sum of the five disturbance coefficient sets and the six disturbance coefficient sets under the third and fourth three-electrode detection modes; when the past medical history of the subject to be tested is cerebral infarction, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the five disturbance coefficient sets under the first and third three-electrode detection modes is greater than the ratio of the sum of the four disturbance coefficient sets and the six disturbance coefficient sets under the second and fourth three-electrode detection modes.

[0068] In the fifth three-electrode detection mode, the third electrode located on the left side of the skull serves as an excitation transmitting end, the first electrode located on the forehead and the sixth electrode located on the right side of the skull serve as excitation receiving ends. Accordingly, the seventh disturbance coefficient set in the fifth three-electrode detection mode includes: the fifteenth disturbance coefficient R5' and the sixteenth disturbance coefficient R16 on the propagation path between the third electrode and the first electrode and the sixth electrode respectively.

[0069] In the sixth three-electrode detection mode, the fourth electrode sheet located on the left side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fifth electrode sheet located on the right side of the skull serve as excitation receiving ends. Accordingly, the eighth disturbance coefficient set in the sixth three-electrode detection mode includes: the tenth disturbance coefficient R3' and the seventeenth disturbance coefficient R17 on the propagation path between the fourth electrode sheet and the second electrode sheet and the fifth electrode sheet respectively.

[0070] In the seventh three-electrode detection mode, the fifth electrode located on the right side of the skull serves as an excitation transmitting end, the first electrode located on the forehead and the fourth electrode located on the left side of the skull serve as excitation receiving ends. Accordingly, the ninth disturbance coefficient set in the seventh three-electrode detection mode includes: the thirteenth disturbance coefficient R6' and the eighteenth disturbance coefficient R18 on the propagation path between the fifth electrode and the first electrode and the fourth electrode respectively.

[0071] In the eighth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode located at the occipital position and the third electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the tenth disturbance coefficient set in the eighth three-electrode detection mode includes: the eleventh disturbance coefficient R4' and the nineteenth disturbance coefficient R19 on the propagation path between the sixth electrode sheet and the second electrode sheet and the third electrode sheet respectively.

[0072] In some embodiments, the data acquisition device further includes an excitation generator, which is configured to generate an excitation signal under a corresponding detection mode according to a control signal sent by a host computer, and control the working state of the electrode group (including: as an excitation transmitting end, or as an excitation receiving end, or as a grounding end, or as a suspended end; wherein the suspended end means that the electrode sheet is neither grounded nor transmits an excitation signal, nor receives an excitation signal, but is suspended).

[0073] Of course, in some other embodiments, the excitation generator may not be set in the data acquisition device, that is, the data acquisition device is externally connected to an excitation generator, and the data acquisition device communicates data with a host computer (preferably, the above-mentioned screening device is integrated in the host computer) through a communication module, and the excitation generator is externally connected to the host computer, and the host computer controls the excitation generator to generate a corresponding excitation signal and then sends it to the corresponding electrode in the data acquisition device; of course, the host computer generates a corresponding control signal according to the corresponding detection mode and sends it to the data acquisition device to control the working state of each electrode sheet.

[0074] In order to verify the effectiveness of the early cognitive impairment screening method and system in this program, the following clinical verification was conducted: Specifically, in order to verify the sensitivity of the screening method and system of the present invention, mental health centers, welfare institutions, elderly care service centers and communities use general self-defined scales and community dementia scales (GSI-D), cognitive function self-rating scales (AD8), daily living ability rating scales (ADL), and simple cognitive assessment (Mini-Cog) scales to screen the cognitive function and emotional state of the elderly population, and compare the screening results obtained by the above scales with the screening results obtained by the present device. The total number of elderly people screened in this activity was 1,502, of which 1,023 were found to have cognitive impairment according to the scales, and 745 were found to have cognitive impairment using the present device, accounting for 72.83%.

[0075] In order to verify the specificity of this device, 180 young healthy people from related companies were screened using this device, and the screening results were evaluated as normal in 168 cases, accounting for 93.33%.

[0076] In summary, the device has a sensitivity of 72.83% and a specificity of 93.33% for early screening of elderly cognitive impairment. In other words, the device has high sensitivity and specificity. The following is an explanation of the prediction of several subjects: Example 1: See Figure 7 Patient I was diagnosed with mild cognitive impairment by a health center in Chongqing, and the detection disturbance coefficient measured by the screening method of the present invention was 169, which was not within the disturbance coefficient range of 120-160 for healthy people, and the predicted probability of suffering from early cognitive impairment was 93.56% (greater than the preset probability threshold of the disease), which was consistent with the actual diagnosis result.

[0077] Example 2: See Figure 8 Patient II was diagnosed with mild cognitive impairment by a health center in Chongqing. The detection disturbance coefficient measured by this device was 183, which was not within the disturbance coefficient range of 120-160 for healthy people. The predicted probability of suffering from early cognitive impairment was 95.51% (greater than the preset probability threshold), which was consistent with the actual diagnosis result.

[0078] Example 3: See Fig. 9 Patient III was diagnosed with mild cognitive impairment by a health center in Chongqing, and the detection disturbance coefficient measured by this device was 114, which was not within the disturbance coefficient range of 120-160 for healthy people, and the predicted probability of suffering from early cognitive impairment was 92.68% (greater than the preset probability threshold of illness), which was consistent with the actual diagnosis result.

[0079] Example 4: See Fig.10Patient IV was diagnosed with mild cognitive impairment by a health center in Chongqing, and the detection disturbance coefficient measured by this device was 163, which was not within the disturbance coefficient range of 120-160 for healthy people, and the predicted probability of suffering from early cognitive impairment was 94.61% (greater than the preset probability threshold), which was consistent with the actual diagnosis result.

[0080] Example 5: See Fig.11 Patient V was diagnosed with mild cognitive impairment by a health center in Chongqing. The detection disturbance coefficient measured by this device was 166, which was not within the disturbance coefficient range of 120-160 for healthy people. The predicted probability of suffering from early cognitive impairment was 93.88% (greater than the preset probability threshold), which was consistent with the actual diagnosis result.

[0081] Example 6: See Fig.12 Patient VI was diagnosed with mild cognitive impairment by a health center in Chongqing. The detection disturbance coefficient measured by this device was 168, which was not within the disturbance coefficient range of 120-160 for healthy people. The predicted probability of suffering from early cognitive impairment was 92.73% (greater than the preset probability threshold), which was consistent with the actual diagnosis result.

[0082] In summary, this scheme constructs a rich prediction model by collecting data from the subjects in advance (including the disturbance data set collected by this device and the data set measured by the scale, etc.); then, the user type is obtained during the specific use process, and the corresponding detection mode is matched for it, and then the disturbance data set of the current user collected using the corresponding mode is input into the pre-built early cognitive impairment screening model to obtain the predicted probability value of early cognitive impairment. The result has high accuracy and small data processing volume, and can be used for large-scale early cognitive impairment screening. It has strong universality and is particularly suitable for places where elderly people gather, such as communities and welfare homes.

[0083] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0084] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for screening early cognitive impairment based on multi-modal electromagnetic field biodetection, characterized in that: include: S101 pre-arranges a first electrode sheet and a second electrode sheet at the forehead and occipital position of the user to be detected, respectively, and arranges a third electrode sheet and a fourth electrode sheet in parallel on the left side of the brain, and arranges a fifth electrode sheet and a sixth electrode sheet in parallel on the right side of the brain; S102 obtains a disturbance coefficient set of the user to be detected under the reference detection mode combination, and inputs the disturbance coefficient set into a pre-trained early cognitive impairment screening model to predict the predicted probability value of the user to be detected having early cognitive impairment; The perturbation coefficient set under the reference detection mode combination includes: the first perturbation coefficient set and the second perturbation coefficient set under the preset first four-electrode detection mode, the second four-electrode detection mode, and the third to sixth perturbation coefficient sets under the first to fourth three-electrode detection modes; In the first four-electrode detection mode, the first electrode sheet located at the forehead is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the third electrode sheet and the fifth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the first disturbance coefficient set includes: the seventh disturbance coefficient R7, the fifth disturbance coefficient R5 and the sixth disturbance coefficient R6 on the propagation path between the first electrode sheet and the second electrode sheet, the third electrode sheet and the fifth electrode sheet respectively; In the second four-electrode detection mode, the second electrode sheet located at the occipital position is used as an excitation transmitting end, the first electrode sheet located at the forehead position and the fourth electrode sheet and the sixth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the second disturbance coefficient set includes: the eighth disturbance coefficient R7', the third disturbance coefficient R3 and the fourth disturbance coefficient R4 on the propagation path between the second electrode sheet and the first electrode sheet, the fourth electrode sheet and the sixth electrode sheet respectively; In the first three-electrode detection mode, the third electrode sheet located on the left side of the skull is used as an excitation transmitting end, the first electrode sheet located at the forehead and the fifth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the three disturbance coefficient sets include: the twelfth disturbance coefficient R5' and the first disturbance coefficient R1 on the propagation path between the third electrode sheet and the first electrode sheet and the fifth electrode sheet respectively; In the second three-electrode detection mode, the fourth electrode sheet located on the left side of the skull is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the sixth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the four disturbance coefficient sets include: the tenth disturbance coefficient R3' and the second disturbance coefficient R2 on the propagation path between the fourth electrode sheet and the second electrode sheet and the sixth electrode sheet respectively; In the third three-electrode detection mode, the fifth electrode sheet located on the right side of the skull is used as an excitation transmitting end, the first electrode sheet located at the forehead and the third electrode sheet located on the left side of the skull are used as excitation receiving ends, and accordingly, the five disturbance coefficient sets include: the thirteenth disturbance coefficient R6' and the eighth disturbance coefficient R1' on the propagation path between the fifth electrode sheet and the first electrode sheet and the third electrode sheet respectively; In the fourth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends, and accordingly, the six disturbance coefficient sets include: the eleventh disturbance coefficient R4' and the ninth disturbance coefficient R2' on the propagation path between the sixth electrode sheet and the second electrode sheet and the fourth electrode sheet respectively; The step of training the early cognitive impairment screening model specifically includes: S201 obtains a health disturbance coefficient data set of multiple healthy subjects under the combination of the reference detection modes; S202 obtaining a data set of abnormal disturbance coefficients of a plurality of subjects with early cognitive impairment under the combination of the benchmark detection modes; S203 obtaining scale test results of the plurality of early cognitive impairment subjects obtained through testing using a designated scale method; S204: Input the healthy disturbance coefficient data set, the abnormal disturbance coefficient data set and the scale detection result into a pre-built neural network model for learning to obtain the early cognitive impairment screening model.

2. The method for early cognitive impairment screening based on multi-modal electromagnetic field biodetection according to claim 1, characterized in that: The step of training to obtain the early cognitive impairment screening model also includes: S205: obtaining and identifying relevant past medical histories of the plurality of subjects with early cognitive impairment; the relevant past medical histories comprising at least one of hypertension, stroke, cerebral infarction and genetic factors; S206a If there are two or more previous medical histories, obtain the disturbance coefficient set of the subject under the reference detection mode combination, and input it, the health disturbance coefficient data set and the scale detection result into a pre-constructed neural network model for training to obtain a first prediction model; If S206b is a single past medical history, obtain the disturbance coefficient sets of the subject under the fifth to eighth three-electrode detection modes, and the disturbance coefficient set obtained under the baseline detection mode combination, and input them together with the health disturbance coefficient data set and the scale detection results into a pre-constructed neural network model for training to obtain a second prediction model.

3. The method for early cognitive impairment screening based on multi-modal electromagnetic field biodetection according to claim 2, characterized in that: S102 specifically includes: Obtaining and identifying relevant past medical history of the user to be detected; If there are two or more previous medical histories, obtaining a disturbance coefficient set of the user to be detected under the reference detection mode combination, and then inputting the set into the first prediction model for prediction; If it is a single past medical history, the disturbance coefficient set of the user to be detected under the reference detection mode combination and the disturbance coefficient set under the fifth to eighth three-electrode detection modes are obtained respectively, and input into the second prediction model for prediction.

4. The method for early cognitive impairment screening based on multi-mode electromagnetic field biodetection according to claim 3, characterized in that: In the fifth three-electrode detection mode, the third electrode sheet located on the left side of the skull serves as an excitation transmitting end, the first electrode sheet located on the forehead and the sixth electrode sheet located on the right side of the skull serve as excitation receiving ends. Accordingly, the seventh disturbance coefficient set in the fifth three-electrode detection mode includes: the fifteenth disturbance coefficient R5' and the sixteenth disturbance coefficient R16 on the propagation path between the third electrode sheet and the first electrode sheet and the sixth electrode sheet respectively.

5. The method for early cognitive impairment screening based on multi-mode electromagnetic field biodetection according to claim 3, characterized in that: In the sixth three-electrode detection mode, the fourth electrode sheet located on the left side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fifth electrode sheet located on the right side of the skull serve as excitation receiving ends. Accordingly, the eighth disturbance coefficient set in the sixth three-electrode detection mode includes: the tenth disturbance coefficient R3' and the seventeenth disturbance coefficient R17 on the propagation path between the fourth electrode sheet and the second electrode sheet and the fifth electrode sheet respectively.

6. The method for early cognitive impairment screening based on multi-mode electromagnetic field biodetection according to claim 3, characterized in that: In the seventh three-electrode detection mode, the fifth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the first electrode sheet located on the forehead and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the ninth disturbance coefficient set in the seventh three-electrode detection mode includes: the thirteenth disturbance coefficient R6' and the eighteenth disturbance coefficient R18 on the propagation path between the fifth electrode sheet and the first electrode sheet and the fourth electrode sheet respectively.

7. The method for early cognitive impairment screening based on multi-mode electromagnetic field biodetection according to claim 3, characterized in that: In the eighth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode located at the occipital position and the third electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the tenth disturbance coefficient set in the eighth three-electrode detection mode includes: the eleventh disturbance coefficient R4' and the nineteenth disturbance coefficient R19 on the propagation path between the sixth electrode sheet and the second electrode sheet and the third electrode sheet respectively.

8. The method for early cognitive impairment screening based on multi-mode electromagnetic field biodetection according to claim 2, characterized in that: If the subject to be tested has a history of stroke, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the four disturbance coefficient sets in the first and second three-electrode detection modes is greater than the ratio of the sum of the five disturbance coefficient sets and the six disturbance coefficient sets in the third and fourth three-electrode detection modes; If the subject to be tested has a past medical history of cerebral infarction, in the corresponding disturbance coefficient set, the ratio of the sum of the three disturbance coefficient sets and the five disturbance coefficient sets in the first and third three-electrode detection modes is greater than the ratio of the sum of the four disturbance coefficient sets and the six disturbance coefficient sets in the second and fourth three-electrode detection modes.

9. An early cognitive impairment screening system based on multi-mode electromagnetic field biodetection, characterized in that: include: The data acquisition device includes a data communication module and an electrode group for users to obtain data, wherein the electrode group includes a first electrode sheet and a second electrode sheet corresponding to the forehead position and the occipital position of the skull, respectively, a third electrode sheet and a fourth electrode sheet arranged in parallel corresponding to the left side of the skull, and a fifth electrode sheet and a sixth electrode sheet arranged in parallel corresponding to the right side of the skull; A screening device, which communicates with the data acquisition device to obtain a disturbance coefficient set of the data acquisition device under a preset reference detection mode combination, and inputs the disturbance coefficient set into a pre-trained early cognitive impairment screening model to predict the predicted probability value of the early cognitive impairment of the user to be detected; Specifically, the screening device comprises: A first data acquisition module is used to communicate with the data acquisition device to respectively acquire a plurality of healthy subjects' health disturbance coefficient data sets under the reference detection mode combination; and respectively acquire a plurality of early cognitive impairment subjects' abnormal disturbance coefficient data sets under the reference detection mode combination; and acquire a disturbance coefficient set of the user to be detected under the reference detection mode combination; A second data acquisition module is used to obtain the scale test results of the plurality of early cognitive impairment subjects obtained by the specified scale method test; A screening model construction module, used to train a pre-constructed neural network model based on the healthy disturbance coefficient data set, the abnormal disturbance coefficient data set and the scale detection result to obtain the early cognitive impairment screening model; A prediction module, configured to make a prediction based on the early cognitive impairment screening model constructed by the screening model construction module and the disturbance coefficient set of the user to be detected under the reference detection mode combination acquired by the first data acquisition module, to obtain a predicted probability value of the user to be detected having early cognitive impairment; The perturbation coefficient set under the reference detection mode combination includes: the first perturbation coefficient set and the second perturbation coefficient set under the preset first four-electrode detection mode, the second four-electrode detection mode, and the third to sixth perturbation coefficient sets under the first to fourth three-electrode detection modes; In the first four-electrode detection mode, the first electrode located at the forehead is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the third electrode sheet and the fifth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the first disturbance coefficient set includes: the seventh disturbance coefficient R7, the fifth disturbance coefficient R5 and the sixth disturbance coefficient R6 on the propagation path between the first electrode sheet and the second electrode sheet, the third electrode sheet and the fifth electrode sheet respectively; In the second four-electrode detection mode, the second electrode sheet located at the occipital position is used as an excitation transmitting end, the first electrode sheet located at the forehead position and the fourth electrode sheet and the sixth electrode sheet located at the left side and the right side of the skull respectively are used as excitation receiving ends, and accordingly, the second disturbance coefficient set includes: the eighth disturbance coefficient R7', the third disturbance coefficient R3 and the fourth disturbance coefficient R4 on the propagation path between the second electrode sheet and the first electrode sheet, the fourth electrode sheet and the sixth electrode sheet respectively; In the first three-electrode detection mode, the third electrode sheet located on the left side of the skull is used as an excitation transmitting end, and the first electrode located at the forehead and the fifth electrode sheet located on the right side of the skull are used as excitation receiving ends. Accordingly, the three disturbance coefficient sets include: the twelfth disturbance coefficient R5' and the first disturbance coefficient R1 on the propagation path between the third electrode sheet and the first electrode sheet and the fifth electrode sheet respectively; In the second three-electrode detection mode, the fourth electrode sheet located on the left side of the skull is used as an excitation transmitting end, the second electrode sheet located at the occipital position and the sixth electrode sheet located on the right side of the skull are used as excitation receiving ends, and accordingly, the four disturbance coefficient sets include: the tenth disturbance coefficient R3' and the second disturbance coefficient R2 on the propagation path between the fourth electrode sheet and the second electrode sheet and the sixth electrode sheet respectively; In the third three-electrode detection mode, the fifth electrode sheet located on the right side of the skull is used as an excitation transmitting end, the first electrode sheet located at the forehead and the third electrode sheet located on the left side of the skull are used as excitation receiving ends, and accordingly, the five disturbance coefficient sets include: the thirteenth disturbance coefficient R6' and the eighth disturbance coefficient R1' on the propagation path between the fifth electrode sheet and the first electrode sheet and the third electrode sheet respectively; In the fourth three-electrode detection mode, the sixth electrode sheet located on the right side of the skull serves as an excitation transmitting end, the second electrode sheet located at the occipital position and the fourth electrode sheet located on the left side of the skull serve as excitation receiving ends. Accordingly, the six disturbance coefficient sets include: the eleventh disturbance coefficient R4' and the ninth disturbance coefficient R2' on the propagation path between the sixth electrode sheet and the second electrode sheet and the fourth electrode sheet respectively.

10. The early cognitive impairment screening system based on multi-mode electromagnetic field biodetection according to claim 9, characterized in that: The screening device also includes: A third data acquisition module is used to acquire and identify the associated past medical history of the multiple early cognitive impairment subjects; the associated past medical history includes at least one of hypertension, stroke, cerebral infarction and genetic factors; a fourth data acquisition module, for acquiring the perturbation coefficient set of the subject under the reference detection mode combination when the subject has two or more related past medical histories, and for acquiring the perturbation coefficient set of the subject under the fifth to eighth three-electrode detection modes respectively when the subject has a single past medical history, Correspondingly, the screening model construction module is also used to, when the subject has two or more related past medical histories, input the perturbation coefficient set of the subject under the baseline detection mode combination, the healthy perturbation coefficient data set and the scale detection result into the model for training to obtain a first prediction model; when the subject has a single past medical history, input the perturbation coefficient set of the subject under the fifth to eighth three-electrode detection modes, the perturbation coefficient set obtained under the baseline detection mode combination, the healthy perturbation coefficient data set and the scale detection result into the model for training to obtain a second prediction model.

Images