Neurosurgery nurse data processing method and system based on intelligent matching

Abstract

The present invention relates to the field of medical health data processing technology, and specifically to a neurosurgery nurse data processing method and system based on intelligent matching, the method comprising: obtaining the patient's medical imaging data and historical work and rest data; taking each two adjacent frames of image images in the medical imaging data as an image frame group, and determining the frame difference result of each image frame group; determining the patient's active interval and inactive interval based on the frame difference result and the historical work and rest data; respectively calculating the image and audio abnormal values ​​of each inactive interval, and extracting the target inactive interval; respectively inputting the medical imaging data corresponding to the target inactive interval and the active interval into the medical care grade prediction model, and obtaining the medical care grade determination result; based on the medical care grade determination result, determining the patient's medical matching recommended data and neurosurgery nurse matching information. The scheme provided by the present invention can be automatically implemented without human intervention, and the allocation of neurosurgery nurse resources is more accurate and more targeted.

Description

Technical Field

[0001] The present invention relates to the technical field of medical health data processing, and in particular to a neurosurgery nurse data processing method and system based on intelligent matching. Background Art

[0002] The main duties of a neurosurgical nurse typically include a variety of tasks related to highly specialized medical care, focusing on treating patients with neurological diseases such as brain tumors, cerebral hemorrhages, spinal cord injuries and other related diseases, including disease monitoring, postoperative care, etc., thereby providing targeted nursing services to patients.

[0003] In the related technology, relevant managers usually assign corresponding neurosurgery nurses to different patients based on the patients' recovery status, pain level, etc. at different stages and subjective experience. This artificial allocation of neurosurgery nurse resources mainly relies on the subjective experience of managers, and the allocation process has strong subjective factors, resulting in poor accuracy and pertinence of the nursing process, and it is difficult to meet the real-time nursing needs of patients.

[0004] It can be seen that the traditional neurosurgery nurse allocation program has technical problems such as poor accuracy and pertinence, and difficulty in meeting patients' real-time nursing needs. Summary of the invention

[0005] In order to solve the technical problems that the traditional neurosurgery nurse allocation scheme has poor accuracy and pertinence and is difficult to meet the real-time nursing needs of patients, the purpose of the present invention is to provide a neurosurgery nurse data processing method and system based on intelligent matching. The technical scheme adopted is as follows:

[0006] On the one hand, the present invention provides a method for processing neurosurgery nurse data based on intelligent matching, the method comprising:

[0007] Obtain the patient's medical imaging data and historical daily routine data;

[0008] Taking each two adjacent frames of image images in the medical image data as an image frame group, and determining a frame difference result of each of the image frame groups;

[0009] Determine the active interval and the inactive interval of the patient based on the frame difference results of each of the image frame groups and the historical work and rest data;

[0010] Calculating the image and audio abnormality values ​​corresponding to each of the inactive intervals respectively, and extracting the target inactive intervals whose image and audio abnormality values ​​meet the preset audio threshold range;

[0011] Inputting the medical imaging data corresponding to the target inactive interval and the active interval into a pre-constructed medical care level prediction model to obtain a medical care level determination result of the patient;

[0012] Based on the medical care level determination result, the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined.

[0013] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, the frame difference result of each image frame group is determined, including:

[0014] Extracting a first grayscale value of each key pixel point in each of the image frame groups in a first image frame and a second grayscale value in a second image frame;

[0015] Subtract the first grayscale value from the second grayscale value of each key pixel to obtain a grayscale difference value of each key pixel;

[0016] The grayscale differences of each key pixel point are summed to obtain a frame difference result of each image frame group.

[0017] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, based on the frame difference results of each of the image frame groups and the historical work and rest data, the inactive interval of the patient is determined, including:

[0018] Extracting continuous time periods in which the frame difference result meets a preset first frame difference threshold range to obtain a plurality of first intervals to be verified;

[0019] Determine the start time, end time and frame difference result average of each first interval to be verified, and calculate the time difference between the time period corresponding to each first interval to be verified and the rest period in the historical work and rest data;

[0020] Based on the start time, end time, frame difference result mean and time difference of each first interval to be verified, the inactive probability of each first interval to be verified is obtained; wherein the frame difference result mean is negatively correlated with the inactive probability;

[0021] The first interval to be verified in which the inactive probability meets the first probability threshold is used as the inactive interval of the patient.

[0022] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, based on the frame difference results of each of the image frame groups and the historical work and rest data, the active interval of the patient is determined, including:

[0023] Extracting continuous time periods in which the frame difference result meets a preset second frame difference threshold range to obtain a plurality of second intervals to be verified;

[0024] Determine the start time, end time and frame difference result average of each second interval to be verified, and calculate the time difference between the time period corresponding to each second interval to be verified and the rest period in the historical work and rest data;

[0025] Based on the start time, end time, frame difference result mean and time difference of each second interval to be verified, the activity probability of each second interval to be verified is obtained; wherein the frame difference result mean is positively correlated with the activity probability;

[0026] The second interval to be verified, in which the activity probability meets the second probability threshold, is used as the active interval of the patient.

[0027] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, the image and audio abnormality values ​​corresponding to each of the inactive intervals are calculated respectively, including:

[0028] For each of the inactive intervals, extracting audio data from the medical imaging data in the inactive interval;

[0029] Based on the audio data, calculating the average audio decibel value at each moment in the inactive interval, extracting the audio decibel value at the key moment, and determining the audio signal amplitude difference at the key moment;

[0030] Based on the audio decibel mean, the audio decibel value at the key moment, and the audio signal amplitude difference, an image and audio anomaly value corresponding to each of the inactive intervals is calculated.

[0031] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, based on the audio decibel mean, the audio decibel value at the critical moment and the audio signal amplitude difference, the image audio abnormal value corresponding to each inactive interval is calculated, including:

[0032] Subtract the audio decibel value at the critical moment from the audio decibel mean and calculate the absolute value to obtain the audio difference absolute value;

[0033] The absolute value of the audio difference is multiplied by the audio signal amplitude difference to obtain an image and audio anomaly value corresponding to each of the inactive intervals.

[0034] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, the medical imaging data corresponding to the target inactive interval and the active interval are respectively input into a pre-built medical care level prediction model to obtain the medical care level determination result of the patient, including:

[0035] Inputting the medical imaging data corresponding to the target inactive interval and the active interval into a pre-built medical care level prediction model respectively, and obtaining the interval prediction result value output by the medical care level prediction model for each interval;

[0036] The predicted result values ​​of each interval are averaged to obtain the medical care level determination result of the patient.

[0037] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, based on the medical care level determination result, the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined, including:

[0038] According to the medical care level determination result, determining the target level to which the patient belongs from the preset patient levels;

[0039] The pre-stored medical care configuration data corresponding to the target level is retrieved, and the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined according to the pre-stored medical care configuration data.

[0040] According to a neurosurgery nurse data processing method based on intelligent matching provided by the present invention, preset patient levels include low risk level, medium risk level, high risk level and special needs level.

[0041] On the other hand, the present invention also provides a neurosurgery nurse data processing system based on intelligent matching, comprising:

[0042] The acquisition module is used to obtain the patient's medical imaging data and historical work and rest data;

[0043] A frame difference determination module, used for treating each two adjacent frames of image images in the medical image data as an image frame group, and determining a frame difference result of each of the image frame groups;

[0044] An interval demarcation module, used for determining the active interval and the inactive interval of the patient based on the frame difference results of each of the image frame groups and the historical work and rest data;

[0045] A calculation module, configured to respectively calculate the image and audio abnormality values ​​corresponding to each of the inactive intervals, and extract the target inactive intervals whose image and audio abnormality values ​​meet a preset audio threshold range;

[0046] A prediction module, used to input the medical imaging data corresponding to the target inactive interval and the active interval into a pre-built medical care level prediction model to obtain a medical care level determination result of the patient;

[0047] A processing module is used to determine the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient based on the medical care level determination result.

[0048] The present invention has the following beneficial effects:

[0049] By obtaining the patient's medical imaging data and historical work and rest data, each two adjacent frames of image images in the medical imaging data are taken as an image frame group, and the frame difference results of each image frame group are determined. Based on the frame difference results of each image frame group and the historical work and rest data, the patient's active interval and inactive interval are determined, and the image and audio abnormal values ​​corresponding to each inactive interval are calculated respectively, and the target inactive interval whose image and audio abnormal values ​​meet the preset audio threshold range is extracted. The medical imaging data corresponding to the target inactive interval and the active interval are input into the pre-built medical care grade prediction model to obtain the patient's medical care grade determination result, and then based on the medical care grade determination result, the patient's corresponding medical matching recommendation data and neurosurgery nurse matching information are determined. Since the entire process can be automatically implemented without human intervention, the problem of subjective factors interfering with the nursing allocation results of the manual allocation plan is reduced, and the neurosurgery nurse resource allocation process is more accurate and more targeted, so as to better meet the real-time nursing needs of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions and advantages 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. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A flowchart of a method for processing neurosurgery nurse data based on intelligent matching provided by one embodiment of the present invention;

[0052] Figure 2 This is a schematic diagram of the fluctuation curve of the frame difference result under one situation;

[0053] Figure 3 It is a schematic diagram of the active probability curve and the inactive probability curve corresponding to the patient at different time periods under one situation;

[0054] Figure 4 A schematic diagram of the structure of a neurosurgery nurse data processing system based on intelligent matching provided by one embodiment of the present invention;

[0055] Figure 5 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a neurosurgery nurse data processing method and system based on intelligent matching proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0057] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0058] Combine the following Figures 1 to 5 The specific scheme of a neurosurgery nurse data processing method and system based on intelligent matching provided by the present invention is specifically described.

[0059] See also Figure 1 , which shows a flow chart of a neurosurgery nurse data processing method based on intelligent matching provided by an embodiment of the present invention.

[0060] like Figure 1 As shown, the neurosurgery nurse data processing method based on intelligent matching provided by the embodiment of the present invention, the execution subject can be an electronic device with data receiving and sending and data processing functions, and the method specifically includes the following steps:

[0061] Step 110: Obtain the patient's medical imaging data and historical work and rest data.

[0062] This embodiment aims to identify the patient's health status and recommend the corresponding nurse or medical care level by analyzing the patient's medical imaging data. Since the patient's recovery status and pain level at different stages have dynamic characteristics, in order to accurately control the patient's status and provide more personalized nursing needs, such as different medical methods under different recovery conditions or nurses with different professional needs, the purpose is to achieve the purpose of accurate and intelligent matching.

[0063] It should be noted that the patient's medical imaging data needs to be collected and analyzed in real time under the premise of complying with medical regulations and patient privacy. Specifically, the constantly updated medical imaging data can be obtained in real time from monitoring or medical recordings. For example, patient images and videos can be obtained from bedside cameras or other monitoring equipment, and further preprocessed to obtain medical imaging data. In practical applications, the patient's historical work and rest data can be extracted from the patient's work and rest record data of the on-duty physician.

[0064] In some embodiments, the historical work and rest data may specifically include the patient's rest periods and active periods in a historical period, such as the patient's daily rest periods and active periods in the previous 7 days.

[0065] Step 120: taking every two adjacent frames of image images in the medical image data as an image frame group, and determining a frame difference result of each image frame group.

[0066] It can be understood that the frame difference result can represent the inter-frame difference between two adjacent frames of image. The larger the frame difference result, the larger the inter-frame difference between the two adjacent frames of image.

[0067] Step 130: Determine the active interval and the inactive interval of the patient based on the frame difference results of each image frame group and the historical work and rest data.

[0068] In this embodiment, the active interval of the patient can represent the time period when the patient's behavior changes more, and the inactive interval can represent the time period when the patient's behavior changes less. For the inactive interval where the patient's behavior changes less, there is less data with reference value for the subsequent analysis and processing process, and it can be appropriately screened, thereby reducing the amount of data for subsequent analysis and processing and improving data processing efficiency.

[0069] Step 140: Calculate the image and audio abnormality values ​​corresponding to each inactive interval respectively, and extract the target inactive interval whose image and audio abnormality values ​​meet the preset audio threshold range.

[0070] It can be understood that in order to extract a part of data from the inactive interval that is useful for the subsequent analysis process, this embodiment uses the analysis of audio abnormalities as a basis for screening. Audio abnormalities such as patient breathing difficulties or groaning can be extracted through audio data, thereby extracting target inactive intervals with greater value for subsequent analysis links from multiple inactive intervals.

[0071] Step 150: inputting the medical imaging data corresponding to the target inactive interval and the active interval into a pre-built medical care level prediction model to obtain the medical care level determination result of the patient.

[0072] In this embodiment, the medical imaging data corresponding to the target inactive interval and the active interval are input into a pre-constructed medical care level prediction model, so that the medical care level can be determined with a smaller amount of data, thereby reducing the data processing amount of the medical care level prediction model and improving the data processing accuracy and efficiency of the data processing link.

[0073] Step 160: Based on the medical care level determination result, determine the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient.

[0074] The solution provided in this embodiment can configure corresponding medical matching recommendation data and neurosurgery nurse matching information for patients after the medical care level determination result is known, so as to realize the effective allocation of medical care resources in a targeted manner according to the patient's personal situation, thereby improving the accuracy of the nursing process.

[0075] In one embodiment, determining the frame difference result of each image frame group specifically includes:

[0076] In the first step, a first grayscale value of each key pixel point in each image frame group in a first image frame and a second grayscale value in a second image frame are extracted.

[0077] In the second step, the first grayscale value of each key pixel is subtracted from the second grayscale value to obtain the grayscale difference of each key pixel.

[0078] The third step is to sum the grayscale differences of each key pixel point to obtain the frame difference result of each image frame group.

[0079] Since patients have different rehabilitation needs under different rehabilitation states during the rehabilitation process, they will have different rehabilitation methods or different rehabilitation accompanying nurses, so as to provide patients with more personalized and comprehensive services through phased analysis. During the rehabilitation stage, the patient's non-verbal signals, such as pain level, are reflected in real-time medical imaging data, manifested as hyperactivity and other behaviors. This embodiment mainly obtains and identifies such behaviors to obtain the corresponding medical care level determination results. Therefore, for each image frame group, the larger the frame difference result, the closer the image image of the image frame group is to the key level medical care frame, that is, the larger the frame difference result, the more important the image frame group is to the medical care level prediction model.

[0080] In this embodiment, the frame difference result of each image frame group can be calculated by the following formula:

[0081]

[0082] Where Fre(t) represents the frame difference result, that is, the grayscale difference between the first and second adjacent image frames in the t-th image frame group at the same pixel position, g(i,j) t Represents the first grayscale value at pixel position i, j; g(i,j) t-1 represents the second grayscale value at pixel position i, j; q represents the total number of key pixels involved in the calculation.

[0083] In practical applications, the frame difference results usually show dynamic fluctuations. Figure 2 The frame difference result fluctuation curve in one situation is shown as an example.

[0084] The frame difference result is intended to quantify the characteristics of the patient's movement changes. In order to reduce the amount of calculation in the process of analyzing and identifying the patient's rehabilitation characteristics through the medical care level prediction model, and to increase the real-time and accuracy of network identification, this embodiment combines the frame difference result with the patient's historical work and rest data for analysis. The smaller the frame difference result or the time period close to 0, and the longer the state lasts, the higher the possibility of the patient's rest, and the lower the necessity of real-time monitoring of the data in this time period. In addition, if this time period belongs to the patient's historical rest period, such as at night, then the corresponding patient's inactivity is higher, and the necessity of network identification in this time period is lower. If it is the opposite of the above situation, the higher the patient's activity, then this time period belongs to the time period of the key medical care level frame, and the necessity of network identification is higher.

[0085] In one embodiment, based on the frame difference results of each image frame group and the historical work and rest data, the inactive interval of the patient is determined, specifically including:

[0086] In the first step, continuous time periods in which the frame difference results satisfy a preset first frame difference threshold range are extracted to obtain a plurality of first intervals to be verified.

[0087] In this embodiment, after obtaining each frame difference result, each frame difference result may be normalized first, and then interval screening may be performed. Specifically, the first frame difference threshold range may be an interval range close to 0, such as a numerical range of 0-0.2.

[0088] The second step is to determine the start time, end time and frame difference result average of each first interval to be verified, and calculate the time difference between the time period corresponding to each first interval to be verified and the rest period in the historical work and rest data.

[0089] The third step is to obtain the inactive probability of each first interval to be verified based on the starting time, ending time, frame difference result mean and time difference of each first interval to be verified; wherein the frame difference result mean is negatively correlated with the inactive probability.

[0090] In this embodiment, the inactivity probability can be calculated by the following formula:

[0091]

[0092] Where Una(k) represents the possibility that the kth first interval to be verified that meets the first frame difference threshold range is an inactive interval, that is, the inactive probability; represents the mean value of the frame difference results of the n image frame groups corresponding to the first interval to be verified; ΔT represents the time difference between this period and the rest period in the patient's historical work and rest data. For example, this period is 8:00, but the patient's rest period is generally from 10:00 pm to 7:00 am the next day, then the time difference at this time is 1h. When this period is within the patient's rest period, the difference time difference ΔT is taken as 0.1, so as to prevent the denominator from being 0; t e Indicates the starting time; t s Indicates the end time.

[0093] It can be seen from the above formula that the duration of the first interval to be verified (t e -t s ) is longer, the possibility that the first interval to be verified is a non-active interval is higher, that is, the greater the probability of inactivity; the smaller the mean value of the frame difference result of the first interval to be verified is, the possibility that the first interval to be verified is a non-active interval is higher, and the more it coincides with the patient's rest time, the higher the corresponding possibility.

[0094] In the fourth step, the first interval to be verified, in which the inactive probability meets the first probability threshold, is used as the inactive interval of the patient.

[0095] In practical applications, the first probability threshold can be reasonably set according to the actual scenario requirements. For example, the first probability threshold can be any value among 0.7, 0.8 or 0.9, so that the inactive interval with a higher inactive probability can be further screened out through the first probability threshold.

[0096] In one embodiment, based on the frame difference results of each image frame group and the historical work and rest data, the active interval of the patient is determined, specifically including:

[0097] In the first step, continuous time periods in which the frame difference results satisfy a preset second frame difference threshold range are extracted to obtain a plurality of second intervals to be verified.

[0098] In practical applications, the second frame difference threshold range may be a range close to 1, for example, the second frame difference threshold range may be a range of 0.8-1.

[0099] The second step is to determine the start time, end time and frame difference result average of each second interval to be verified, and calculate the time difference between the time period corresponding to each second interval to be verified and the rest period in the historical work and rest data.

[0100] The third step is to obtain the activity probability of each second interval to be verified based on the start time, end time, frame difference result mean and time difference of each second interval to be verified; wherein the frame difference result mean is positively correlated with the activity probability.

[0101] In the fourth step, the second interval to be verified, whose active probability meets the second probability threshold, is used as the active interval of the patient.

[0102] In practical applications, the second probability threshold can be reasonably set according to the accuracy requirement. For example, the second probability threshold can be any value among 0.7, 0.8 or 0.9.

[0103] In this embodiment, the activity probability of the second to-be-verified interval can be calculated by the following formula:

[0104]

[0105] Wherein, Act(u) represents the possibility that the u-th second interval to be verified that meets the second frame difference threshold range is an active interval, that is, the active probability; represents the mean value of the frame difference results of the n image frame groups corresponding to the second interval to be verified; ΔT represents the time difference between this period and the rest period in the patient's historical work and rest data. For example, this period is 8:00, but the patient's rest period is generally from 10:00 pm to 7:00 am the next day, then the time difference at this time is 1h. When this period is within the patient's rest period, the difference time difference ΔT is taken as 0.1, so as to prevent the denominator from being 0; t e Indicates the starting time; t s Indicates the end time.

[0106] It can be seen from the above formula that the mean value of the frame difference result of the second interval to be verified is proportional to the active probability. Since this part quantifies the possibility of activity of the patient, that is, the dynamic characteristics of the patient, the higher the mean value of the frame difference result of the second interval to be verified, the closer the patient is to the active state. Figure 3 The active probability curve and the inactive probability curve corresponding to the patient in different time periods are exemplarily shown.

[0107] In the active interval, it is necessary to identify and analyze the medical imaging data in this interval and determine the medical care level. This is because for patients in the rehabilitation stage, the time period corresponding to the active interval may contain more valuable non-verbal signals, such as pain level and emotions, which need to be identified and analyzed and then the medical care level is divided. This can reduce network energy consumption and computing power while increasing the real-time nature of the data analysis process and improving the accuracy of patient analysis.

[0108] In one embodiment, the image and audio abnormality values ​​corresponding to each inactive interval are calculated respectively, specifically including:

[0109] In the first step, for each inactive interval, the audio data in the medical image data within the inactive interval is extracted.

[0110] In the second step, based on the audio data, the average audio decibel value at each moment in the inactive interval is calculated, the audio decibel value at the key moment is extracted, and the audio signal amplitude difference at the key moment is determined.

[0111] In the third step, based on the audio decibel mean, the audio decibel value at the key moment, and the audio signal amplitude difference, the image audio anomaly value corresponding to each inactive interval is calculated.

[0112] It is understandable that the patient's medical care level is reflected in the degree of pain, emotions and other information manifested in behavioral characteristics, as well as the patient's audio information. However, when the patient is in an inactive interval, the patient is prone to diseases that are prone to occur in the rehabilitation stage. For example, patients in the rehabilitation stage are bedridden for a long time or have limited mobility, and their blood circulation is poor, which can easily lead to thrombosis. In addition, the difficulty breathing or groaning caused by such rehabilitation complications are real-time signals that need to be monitored. Therefore, this embodiment performs an audio abnormality analysis of the patient in the inactive interval, and determines whether further analysis is necessary through image and audio abnormalities.

[0113] In one embodiment, based on the audio decibel mean, the audio decibel value at the critical moment, and the audio signal amplitude difference, the image and audio abnormality value corresponding to each inactive interval is calculated, specifically including:

[0114] On the one hand, the audio decibel value at the critical moment is subtracted from the audio decibel mean and the absolute value is calculated to obtain the absolute value of the audio difference.

[0115] On the other hand, the absolute value of the audio difference is multiplied by the audio signal amplitude difference to obtain the image and audio abnormality value corresponding to each inactive interval.

[0116] In this embodiment, the image and audio abnormal value can be calculated by the following formula:

[0117]

[0118] Where Un(v) represents the image audio anomaly value of the vth inactive interval; dB(m) represents the audio decibel value at the key moment m in the interval; It represents the mean audio decibel value in this interval; ΔAm(m) represents the audio signal amplitude difference corresponding to the audio signal after short-term Fourier transform, that is, the frequency domain amplitude difference of the audio signal within two adjacent seconds.

[0119] This part aims to quantify the patient's recovery status in the inactive interval. By analyzing the audio data, the image and audio anomaly values ​​are obtained. The greater the difference in audio decibel values, the higher the image and audio anomaly values; and the lower the audio periodicity, that is, the higher the ΔAm(m), the higher the corresponding image and audio anomaly values. The audio periodicity aims to quantify the patient's breathing uniformity, that is, whether there is a decrease in audio periodicity due to breathing difficulties or the absence of a period.

[0120] After normalizing the audio outliers, the target inactive intervals whose image and audio outliers meet the preset audio threshold range can be extracted. The preset audio threshold range can be the interval range of 0.9 to 1, so that the target inactive intervals with higher image and audio outliers can be extracted. Subsequently, only the target inactive intervals in the inactive intervals are subjected to data analysis, and other inactive intervals are not analyzed and identified.

[0121] In one embodiment, the medical imaging data corresponding to the target inactive interval and the active interval are respectively input into a pre-built medical care level prediction model to obtain the medical care level determination result of the patient, specifically including:

[0122] First, the medical imaging data corresponding to the target inactive interval and active interval are respectively input into a pre-built medical care level prediction model to obtain the interval prediction result value output by the medical care level prediction model for each interval.

[0123] In this embodiment, each interval that requires data analysis corresponds to an interval prediction result value, that is, the input of the medical care level prediction model is the medical imaging data of an interval, and the output is the interval prediction result value.

[0124] In practical applications, the medical care level prediction model can adopt a neural network model. After training the neural network model with medical image samples and interval prediction result samples, a medical care level prediction model can be obtained. The medical care level prediction model has the function of patient feature recognition and can directly identify and predict the corresponding interval prediction result value based on the medical care image data.

[0125] Then, the predicted result values ​​of each interval are averaged to obtain the patient's medical care level judgment result.

[0126] This embodiment calculates the average of the prediction result values ​​in each interval, so that the final medical care level determination result can comprehensively represent the patient characteristics in each interval and has higher accuracy.

[0127] In one embodiment, based on the medical care level determination result, the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined, specifically including:

[0128] First, according to the medical care level determination result, the target level to which the patient belongs is determined from the preset patient levels.

[0129] Then, the pre-stored medical care configuration data corresponding to the target level is retrieved, and the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined based on the pre-stored medical care configuration data.

[0130] In some embodiments, the preset patient levels may specifically include a low risk level, a medium risk level, a high risk level, and a special needs level.

[0131] In actual applications, pre-stored medical care configuration data corresponding to different patient levels can be set in advance to establish a medical care configuration database. Later, after the target level of the patient is known, the pre-stored medical care configuration data corresponding to the target level can be directly retrieved from the medical care configuration database.

[0132] Among them, the pre-stored medical care configuration data includes patient status description content, medical guidance recommendation information, and neurosurgery nurse matching information.

[0133] Specifically, for the low-risk level, the patient status description is: the patient's status is stable, and the needs are mainly basic care and daily monitoring. The medical guidance recommendation information is: regular health checks and basic nursing. The matching information for neurosurgery nurses is: junior nurses or nurses with basic nursing skills.

[0134] For the medium-risk level, the patient status description is: the patient has a chronic disease that needs to be managed, or has moderate nursing needs. The medical guidance recommendation information is: chronic disease management, intermediate nursing skills, such as medication management and pain control. The matching information for neurosurgery nurses is: nurses with relevant disease management experience or intermediate nurses.

[0135] For the high-risk level, the patient status description is: the patient's status is unstable and may require intensive monitoring or specialized treatment. The medical guidance recommendation information is: advanced nursing skills, including emergency and complex disease management. The matching information for neurosurgery nurses is: advanced nurses or nurses with specific professional skills.

[0136] For the special needs level, the patient status description is: the patient needs special medical treatment and care, such as postoperative recovery or serious illness. The medical guidance recommendation information is: professional treatment and rehabilitation guidance. The matching information for neurosurgery nurses is: senior nurses or nurses with advanced professional qualifications, such as surgical nursing, rehabilitation nursing, etc.

[0137] After obtaining the patient's pre-stored medical care configuration data, the patient's status description content and medical guidance recommendation information can be used as medical matching recommendation data, and neurosurgery nurse matching information can be output, thereby providing powerful data guidance for the allocation of neurosurgery nurses.

[0138] In summary, the neurosurgery nurse data processing method based on intelligent matching provided by the embodiment of the present invention has at least the following advantages:

[0139] First, in order to identify and analyze the patient's recovery status, we chose to use a machine learning method to capture and analyze the patient's recovery status, accurately control the patient's status, and provide further medical care based on the patient's status.

[0140] Second, the patient's active and inactive intervals are determined based on the patient's historical work and rest data combined with the frame difference results. Only the data of the active intervals and some inactive intervals are identified and analyzed, which reduces the amount of calculation and increases accuracy and real-time performance.

[0141] Third, considering that there may be emergencies during the patient's inactive period, in order to avoid such situations, we choose to supplement the network recognition data based on the patient's image and audio difference values ​​to play a role in comprehensive monitoring.

[0142] Based on the same general inventive concept, the present invention also protects a neurosurgery nurse data processing system based on intelligent matching. The neurosurgery nurse data processing system based on intelligent matching provided by the present invention is described below. The neurosurgery nurse data processing system based on intelligent matching described below and the neurosurgery nurse data processing method based on intelligent matching described above can be referenced to each other.

[0143] See also Figure 4 , which shows a schematic diagram of the structure of a neurosurgery nurse data processing system based on intelligent matching provided by an embodiment of the present invention.

[0144] like Figure 4 As shown, the neurosurgery nurse data processing system based on intelligent matching provided by the embodiment of the present invention specifically includes:

[0145] An acquisition module 210 is used to acquire the patient's medical imaging data and historical work and rest data;

[0146] A frame difference determination module 220 is used to take each two adjacent frames of image images in the medical image data as an image frame group and determine a frame difference result of each image frame group;

[0147] An interval delineation module 230 is used to determine the active interval and the inactive interval of the patient based on the frame difference results of each image frame group and the historical work and rest data;

[0148] The calculation module 240 is used to calculate the image and audio abnormal values ​​corresponding to each inactive interval, and extract the target inactive interval whose image and audio abnormal values ​​meet the preset audio threshold range;

[0149] The prediction module 250 is used to input the medical image data corresponding to the target inactive interval and the active interval into a pre-built medical care level prediction model to obtain a medical care level determination result of the patient;

[0150] The processing module 260 is used to determine the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient based on the medical care level determination result.

[0151] Regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0152] Figure 5 Schematic diagram of the structure of an electronic device in an embodiment of the present invention.

[0153] like Figure 5 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the neurosurgery nurse data processing method based on intelligent matching provided in the above embodiments.

[0154] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0155] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the neurosurgery nurse data processing method based on intelligent matching provided in the above-mentioned embodiments.

[0156] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the neurosurgery nurse data processing method based on intelligent matching provided in the above-mentioned embodiments.

[0157] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0158] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0159] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0160] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A neurosurgery nurse data processing method based on intelligent matching, characterized in that: The method comprises: Obtain the patient's medical imaging data and historical daily routine data; Taking each two adjacent frames of image images in the medical image data as an image frame group, and determining a frame difference result of each of the image frame groups; Determine the active interval and the inactive interval of the patient based on the frame difference results of each of the image frame groups and the historical work and rest data; Calculating the image and audio abnormality values ​​corresponding to each of the inactive intervals respectively, and extracting the target inactive intervals whose image and audio abnormality values ​​meet the preset audio threshold range; Inputting the medical imaging data corresponding to the target inactive interval and the active interval into a pre-constructed medical care level prediction model to obtain a medical care level determination result of the patient; Based on the medical care level determination result, determining the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient; The determining of the active interval and the inactive interval of the patient based on the frame difference results of each of the image frame groups and the historical work and rest data includes: Extracting continuous time periods in which the frame difference result meets a preset first frame difference threshold range to obtain a plurality of first intervals to be verified; Determine the start time, end time and frame difference result average of each first interval to be verified, and calculate the time difference between the time period corresponding to each first interval to be verified and the rest period in the historical work and rest data; Based on the start time, end time, frame difference result mean and time difference of each first interval to be verified, the inactive probability of each first interval to be verified is obtained; wherein the frame difference result mean is negatively correlated with the inactive probability; Taking the first interval to be verified, in which the inactive probability meets the first probability threshold, as the inactive interval of the patient; Extracting continuous time periods in which the frame difference result meets a preset second frame difference threshold range to obtain a plurality of second intervals to be verified; Determine the start time, end time and frame difference result average of each second interval to be verified, and calculate the time difference between the time period corresponding to each second interval to be verified and the rest period in the historical work and rest data; Based on the start time, end time, frame difference result mean and time difference of each second interval to be verified, the activity probability of each second interval to be verified is obtained; wherein the frame difference result mean is positively correlated with the activity probability; The second interval to be verified, in which the activity probability meets the second probability threshold, is used as the active interval of the patient.

2. A neurosurgery nurse data processing method based on intelligent matching according to claim 1, characterized in that: Determining a frame difference result of each of the image frame groups includes: Extracting a first grayscale value of each key pixel point in each of the image frame groups in a first image frame and a second grayscale value in a second image frame; Subtract the first grayscale value from the second grayscale value of each key pixel to obtain a grayscale difference value of each key pixel; The grayscale differences of each key pixel point are summed to obtain a frame difference result of each image frame group.

3. A neurosurgery nurse data processing method based on intelligent matching according to claim 1, characterized in that: Calculating the image and audio abnormality values ​​corresponding to each of the inactive intervals respectively, including: For each of the inactive intervals, extracting audio data from the medical imaging data in the inactive interval; Based on the audio data, calculating the average audio decibel value at each moment in the inactive interval, extracting the audio decibel value at the key moment, and determining the audio signal amplitude difference at the key moment; Based on the audio decibel mean, the audio decibel value at the key moment, and the audio signal amplitude difference, an image and audio anomaly value corresponding to each of the inactive intervals is calculated.

4. A neurosurgery nurse data processing method based on intelligent matching according to claim 3, characterized in that: Based on the audio decibel mean, the audio decibel value at the key moment, and the audio signal amplitude difference, the image and audio anomaly value corresponding to each of the inactive intervals is calculated, including: Subtract the audio decibel value at the critical moment from the audio decibel mean and calculate the absolute value to obtain the audio difference absolute value; The absolute value of the audio difference is multiplied by the audio signal amplitude difference to obtain an image and audio anomaly value corresponding to each of the inactive intervals.

5. A neurosurgery nurse data processing method based on intelligent matching according to claim 1, characterized in that: The medical imaging data corresponding to the target inactive interval and the active interval are respectively input into a pre-constructed medical care level prediction model to obtain a medical care level determination result of the patient, including: Inputting the medical imaging data corresponding to the target inactive interval and the active interval into a pre-built medical care level prediction model respectively, and obtaining the interval prediction result value output by the medical care level prediction model for each interval; The predicted result values ​​of each interval are averaged to obtain the medical care level determination result of the patient.

6. A neurosurgery nurse data processing method based on intelligent matching according to claim 1, characterized in that: Based on the medical care level determination result, the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined, including: According to the medical care level determination result, determining the target level to which the patient belongs from the preset patient levels; The pre-stored medical care configuration data corresponding to the target level is retrieved, and the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient are determined according to the pre-stored medical care configuration data.

7. A neurosurgery nurse data processing method based on intelligent matching according to claim 6, characterized in that: The preset patient levels include low risk, medium risk, high risk and special needs level.

8. A neurosurgery nurse data processing system based on intelligent matching, characterized in that: The system is used to implement the steps of a neurosurgery nurse data processing method based on intelligent matching according to any one of claims 1 to 7, and the neurosurgery nurse data processing system based on intelligent matching includes: The acquisition module is used to obtain the patient's medical imaging data and historical work and rest data; A frame difference determination module, used for treating each two adjacent frames of image images in the medical image data as an image frame group, and determining a frame difference result of each of the image frame groups; An interval demarcation module, used for determining the active interval and the inactive interval of the patient based on the frame difference results of each of the image frame groups and the historical work and rest data; A calculation module, configured to respectively calculate the image and audio abnormality values ​​corresponding to each of the inactive intervals, and extract the target inactive intervals whose image and audio abnormality values ​​meet a preset audio threshold range; A prediction module, used to input the medical imaging data corresponding to the target inactive interval and the active interval into a pre-built medical care level prediction model to obtain a medical care level determination result of the patient; A processing module is used to determine the medical matching recommendation data and neurosurgery nurse matching information corresponding to the patient based on the medical care level determination result.

Images