Intelligent identification of kidney stone composition and individualized prevention and treatment plan generation system

Abstract

The present application relates to the technical field of medical diagnosis system, specifically to a kidney stone composition intelligent identification and individualized prevention and treatment scheme generation system, which comprises a stone type identification module, a composition correlation determination module, an acid-base state division module, a metabolic behavior ordering module and a path scheme construction module.In the present application, the linkage characteristics of the local structure and energy state of the stone surface are extracted through the corresponding relationship of the image texture and the spectral peak position in space, the continuous form of the composition change on the time axis is captured through the period comparison mode of the consistent trend direction, the structural change of the metabolic state in the stage is drawn through the collaborative control of the multiple biochemical trends of lactic acid and bicarbonate in the unified time scale, the mode track of the metabolic behavior is tracked through the difference of the behavior direction change frequency in the period, the path sequence of the metabolic correlation behavior is constructed through the step-by-step promotion of the trend relationship in the period, and the dynamic linkage recognition ability between the data and the stage adaptation ability of the individual intervention path are enhanced.

Description

Technical Field

[0001] This invention relates to the field of medical diagnostic system technology, and in particular to a system for intelligent identification of kidney stone components and generation of personalized prevention and treatment plans. Background Technology

[0002] The field of medical diagnostic systems technology involves procedural methods for collecting, identifying, analyzing, and intervening in human health conditions. Core aspects include medical image analysis, physiological signal acquisition, biochemical component detection, feature indicator extraction, pathological classification, and individual information integration. The methodological approach in this field typically relies on information collection methods to obtain patient data, combines knowledge base rules to complete symptom analysis and disease identification, and formulates intervention recommendations based on diagnostic conclusions, emphasizing the accuracy of data-driven judgments and the systematic nature of the processing flow. Among these, the traditional intelligent identification and personalized prevention and treatment plan generation system for kidney stones identifies the type of kidney stone by detecting the chemical composition of urine sediment or expelled stone samples. Based on the patient's daily water intake, diet, blood calcium levels, uric acid concentration, and other biochemical or lifestyle indicators, it formulates targeted stone prevention and dietary adjustment recommendations. Traditional methods typically use infrared spectroscopy to identify substances such as calcium oxalate, calcium phosphate, and uric acid, manually comparing the spectra for component judgment, and relying on existing clinical rules to classify and assess the patient's lifestyle to generate guidance.

[0003] Existing technologies rely on a single static image comparison method for component identification, lacking the ability to spatially identify the relationship between local image structure and spectral energy. The processing is based on indicator point values ​​rather than trend directions, making it difficult to reflect the dynamic evolution of metabolic processes. In biochemical indicator analysis, the temporal consistency between multiple trends is ignored, making it difficult to effectively capture changes in physiological state within a cycle. Behavioral intervention suggestions are generated based on rule template matching, lacking in-depth analysis of the differentiated structure of behavioral trajectories. When faced with significant individual cycle changes, it is difficult to form intervention rankings and path combinations that fit their metabolic characteristics, resulting in reduced timeliness and accuracy of diagnostic guidance. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a system for intelligent identification of kidney stone components and generation of individualized prevention and treatment plans.

[0005] On the one hand, it provides a system for intelligent identification of kidney stone components and generation of individualized prevention and treatment plans. This system includes: The stone type identification module acquires images and spectral data of the excrement, locates the edge and center regions of the crystals, identifies image segments that change synchronously in spatial location, and obtains the main type labeling results of the stones. Based on the main type of stone labeling results, the component association determination module calls the records of changes in free calcium and excreted calcium within the period, compares the trend direction in chronological order, analyzes the trend extension behavior, and obtains the component activity status determination results. Based on the results of the determination of the active state of the components, the acid-base state division module compares the direction of acid-base reaction with the trend of lactic acid and bicarbonate concentration changes, identifies the direction of the periodic trend, and obtains the structural results of the acid-base reaction segment. Based on the acid-base reaction segment structure results, the metabolic behavior ranking module obtains the direction of urinary calcium excretion, water drinking interval records and urination rhythm trajectory, compares the frequency of directional changes in behavior within the cycle, updates the behavioral trend changes, and obtains the metabolic behavior intervention ranking results. Based on the metabolic behavior intervention ranking results, the path scheme construction module extracts the trend direction of uric acid and urinary calcium, compares the directional relationship of each cycle, updates the cycle trend status, and obtains the individual prevention and control path segment combination results.

[0006] As a further aspect of the present invention, the main type of stone labeling results include crystal edge region identification, central region spectral feature extraction, texture change and peak position linkage region; the component activity state determination results include blood calcium trend direction consistent segment, urinary calcium trend direction consistent segment, trend structure extension segment, and direction coherence behavior feature; the acid-base reaction segment structure results include urinary acid-base reaction trend segment, lactic acid trend direction segment, bicarbonate trend direction segment, trend direction consistent continuous segment, and trend direction offset continuous segment; the metabolic behavior intervention ranking results include urinary calcium excretion change direction item, water drinking interval change item, urination rhythm change item, direction change frequency difference item, and direction trend tracking item; the individual prevention and treatment path segment combination results include uric acid trend matching period segment, urinary calcium trend matching period segment, trend direction consistent period combination, and trend direction inconsistent control period segment.

[0007] As a further aspect of the present invention, the record of changes in free calcium and excreted calcium refers to a data set that synchronously records and compares the concentrations of free calcium in human blood and excreted calcium in urine at the same time point within the same period. The direction of the acid-base reaction refers to the comparison between the trend of acid-base changes in urine and the trend of changes in lactic acid and bicarbonate concentrations within the target time period.

[0008] As a further aspect of the present invention, the urination rhythm trajectory refers to the changing trend of the daily urination frequency and time distribution of an individual within a cycle; The cycle trend status refers to the comparison of the trend directions of uric acid and urinary calcium within the metabolic cycle to determine the consistency of the trend in each cycle.

[0009] As a further aspect of the present invention, the stone type identification module includes: The image and spectrum acquisition submodule acquires the surface of the kidney stone excrement, collects surface images and spectral data of the corresponding areas, encodes image pixels, decomposes spectral channels, and associates image coordinates with spectral positions point-to-point to obtain corresponding image spectral groups. The regional feature localization submodule, based on the image spectrum correspondence group, calls the gray-level changes of the crystal edge and center region in the image, the differential region texture direction, and combines the spatial distribution characteristics of the corresponding spectral energy values ​​to extract positional data representing the texture change trend, thereby obtaining the texture spectrum corresponding point set; The type labeling submodule extracts the corresponding texture frequency and spectral main energy value based on the texture spectrum synchronization point set, calls the stone type feature data, compares the numerical similarity threshold, and obtains the stone main type labeling result.

[0010] As a further aspect of the present invention, the component association determination module includes: The type information docking submodule, based on the type information in the main type labeling result of the stone, calls the blood free calcium sequence and urine calcium excretion sequence within the same period, and obtains a dual-channel data group according to the data corresponding to the time point; The direction extraction submodule extracts the direction of change of blood calcium and urine calcium based on the dual-channel data set, compares the direction signs, identifies time periods with consistent directions, removes segments with reversed directions, and obtains a sequence of segments with consistent directions. The state feature extraction submodule calls the directional consistent segment sequence to extract the rate of change of blood calcium and urine calcium within the corresponding time period, monitors the fluctuation range between rate values ​​within continuous segments, and obtains the result of component activity status determination.

[0011] As a further aspect of the present invention, during the process of extracting the change direction of blood calcium and urine calcium, the numerical relationship between two adjacent time points in the blood free calcium sequence and the urine calcium excretion sequence is compared to analyze the change direction. During the process of removing segments with reversed direction, the time points of the direction change are identified along the sequence of change direction, and the data are separated by the time points. In the process of extracting the rate of change of blood calcium and urinary calcium within the corresponding time period, the blood calcium and urinary calcium data corresponding to the first and last time points in the time period with the same direction are extracted and arranged in chronological order; in the process of monitoring the fluctuation range between rate values, the trend of change of blood calcium and urinary calcium data in the time period with the same direction is compared segment by segment according to the chronological order to identify the data segments in which the change trend has different amplitudes.

[0012] As a further aspect of the present invention, the acid-base state division module includes: The time segment extraction submodule, based on the time period data in the component activity state determination result, calls the urine acid-base reaction direction change sequence within the corresponding time range, and simultaneously extracts the lactic acid concentration sequence and bicarbonate concentration sequence within the time period, aligns the data according to the time point, and obtains a three-sequence time point combination set. The trend direction comparison submodule extracts the trend direction of the lactate concentration sequence and the bicarbonate concentration sequence based on the three sequence time point combination set, compares the change direction of each time period, identifies the interval with consistent direction and the interval with offset direction, and obtains the trend comparison interval set. Based on the trend comparison interval set, the reaction structure separation submodule extracts the reaction direction change state according to the trend consistency segment and the offset segment, and classifies them into differentiated reaction segment types to obtain the acid-base reaction segment structure results.

[0013] As a further aspect of the present invention, the metabolic behavior ranking module includes: The behavioral data extraction submodule, based on the time period in the acid-base reaction segment structure result, obtains the urinary calcium excretion change direction, water drinking interval change data and urination rhythm trajectory information within the cycle, and extracts the directional change sequence of the behavioral item within the cycle according to the time point corresponding to the behavioral item, thus obtaining a behavioral direction time series set. The direction change tracking submodule, based on the behavior direction time series set, compares the change points of each type of behavior direction within the period, extracts the occurrence frequency of direction switching actions, compares the difference in change frequency of behavior items within the time period, and obtains behavior direction fluctuation feature groups. The behavior ranking output submodule tracks the directional continuity trajectory of behavior items within a period based on the behavior direction fluctuation feature group, and adjusts the position of the behavior before and after the intervention according to the distribution of the change frequency and directional trend change, so as to obtain the metabolic behavior intervention ranking result.

[0014] As a further aspect of the present invention, the path scheme construction module includes: The trend data extraction submodule obtains the uric acid change direction sequence and urinary calcium change direction sequence within the corresponding time period based on the behavior cycle in the ranking results of the metabolic behavior intervention, divides the trend data according to the cycle, and obtains the trend direction cycle sequence group. The cycle direction comparison submodule compares the uric acid direction and urinary calcium direction in the cycle point by point based on the trend direction cycle sequence group, distinguishes the trend direction of the cycle segment, and obtains the trend consistency partitioning result. The path segment merging submodule, based on the trend consistency partitioning results, connects the differentiated periodic segments in chronological order, connects the behavior sorting information with the trend direction status, extracts the intervention segment range in the continuous advancement path, and obtains the individual prevention and control path segment combination results.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the correlation between local structure and energy state on the surface of the stone is extracted by the spatial correspondence between image texture and spectral peak position. The continuous morphology of component changes on the time axis is captured by the periodic comparison method with consistent trend direction. The structural changes of metabolic state in stages are plotted by the synergistic comparison of multiple groups of biochemical trends of lactic acid and bicarbonate at a unified time scale. The difference in the frequency of behavioral direction changes within the cycle is used to track the pattern trajectory of metabolic behavior. The path sequence of metabolic related behaviors is constructed by the step-by-step advancement of trend relationship in the cycle, thereby enhancing the dynamic linkage recognition ability between data and the stage adaptation ability of individual intervention path. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a system block diagram of the present invention; Figure 3 This is a flowchart of the stone type identification module in this invention; Figure 4 This is a flowchart of the component correlation determination module in this invention; Figure 5 This is a flowchart of the acid-base state division module in this invention; Figure 6 This is a flowchart of the metabolic behavior ranking module in this invention; Figure 7 This is a flowchart of the path scheme construction module in this invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] This invention provides a system for intelligent identification of kidney stone components and generation of personalized prevention and treatment plans, such as... Figure 1-2 The diagram shown illustrates a system for intelligent identification and personalized prevention and treatment of kidney stones. This system includes: The stone type identification module acquires surface images and spectral data of kidney stone excrement, selects crystal edges and central regions in the image, compares the spatial positions of the same region image and peaks based on the energy changes of differential peak positions in the spectrum, identifies regions where image texture changes and peak position changes occur simultaneously, and obtains the main stone type labeling results. The component association determination module uses the type information in the main type of stone labeling results to call the blood free calcium change records and urine calcium excretion change records within the same period, aligns the data point by point on the time axis, analyzes whether the trend direction of each segment is consistent, extends the trend structure analysis in the segment with consistent direction, identifies the behavioral characteristics of coherent direction, and obtains the component activity status determination result. The acid-base state division module calls the urine acid-base reaction direction change records within the time range based on the time period corresponding to the component activity state determination results. At the same time, it analyzes the change trend of lactic acid concentration and the change trend of bicarbonate concentration, arranges the trends with the same time scale, compares the trend direction of the time period, distinguishes between continuous segments with consistent direction and continuous segments with shifted direction, and obtains the acid-base reaction segment structure results. The metabolic behavior ranking module is based on the results of acid-base reaction segment structure. It obtains the change direction of urinary calcium excretion, the change record of water drinking interval, and the change trajectory of urination rhythm within the corresponding time period. It compares the change direction of behavior within the cycle, checks the number of changes in direction within the cycle item by item, identifies the behavior items with different numbers of changes in direction, and follows the directional trend of behavior within the cycle to obtain the metabolic behavior intervention ranking results. The pathway construction module is based on the ranking results of metabolic behavior intervention. It obtains the trend direction of uric acid and the trend direction of urinary calcium within the behavior change cycle, compares the change direction of the trend in each cycle, continues the behavior analysis sequence in cycles with the same trend direction, and compares the trend separately in cycles with inconsistent directions. The cycle is advanced in chronological order to obtain the individual prevention and control pathway segment combination results.

[0024] The results of the main stone type labeling include crystal edge region identification, central region spectral feature extraction, texture change and peak position linkage region; the results of the component activity status determination include blood calcium trend direction consistent segment, urinary calcium trend direction consistent segment, trend structure extension segment, and direction coherence behavior characteristics; the results of the acid-base reaction segment structure include urinary acid-base reaction trend segment, lactic acid trend direction segment, bicarbonate trend direction segment, trend direction consistent continuous segment, and trend direction deviation continuous segment; the results of the metabolic behavior intervention ranking include urinary calcium excretion change direction item, water drinking interval change item, urination rhythm change item, direction change frequency difference item, and direction trend tracking item; the results of the individual prevention and treatment path segment combination include uric acid trend matching period segment, urinary calcium trend matching period segment, trend direction consistent period combination, and trend direction inconsistent control period segment.

[0025] Specifically, such as Figure 2 , 3 As shown, the stone type identification module includes: The image and spectrum acquisition submodule acquires the surface of the kidney stone excrement, collects surface images and spectral data of the corresponding areas, encodes image pixels, decomposes spectral channels, and associates image coordinates with spectral positions point-to-point to obtain corresponding image spectral groups. First, a high-definition imaging device was used to capture the entire excrement containing the kidney stone. The image resolution was set to 1024×1024 pixels. The viewing angle and imaging distance of the acquisition device were stabilized within a fixed range, for example, the imaging distance was kept at 15 cm, to ensure that the image covered the entire surface area of ​​the kidney stone. At the same time, the light source brightness was adjusted to a constant illuminance of 500 lux to eliminate the influence of shadows. Then, the spectral scanning range was set to the 400 to 1000 nm band, divided into 30 consecutive band intervals, each band width of 20 nm. A sampling point was marked every 20 pixels in the image. Then, at each sampling point, the spectral reflectance information of the area was acquired and recorded using a multi-channel spectral probe. The reflectance value for each band is recorded as follows: for example, the reflectance of the 10th band is 0.38. Then, the image pixel data is encoded in the form of row and column numbers. For example, the pixel in the 200th row and 300th column is encoded as P200_300. The spectral sampling points are numbered from S1 to Sn. A one-to-one mapping relationship between the P coordinate and the S coordinate is established through the image calibration matrix. Based on the pixel position of each sampling point in the image, the coordinate reprojection operation is performed and the error is corrected. The error is controlled within ±1 pixel. Finally, the binding of each pixel in the image to its corresponding spectral sampling point is realized, resulting in a combined data group including the image pixel value and its corresponding spectral reflectance of 30 bands, thus obtaining the image spectral correspondence group.

[0026] The regional feature localization submodule is based on the image spectral correspondence group. It calls the gray-level changes of the crystal edge and center region in the image, the differential region texture direction, and combines the corresponding spectral energy value change rate to extract the location data of synchronous fluctuations and obtain the texture spectral synchronization point set. First, several image patches covering the target crystalline region were selected from the image of kidney stone excrement. Each patch was 64×64 pixels in size. The radial variation curve of grayscale value from the edge to the center of each image patch was extracted. The grayscale value range of the edge region was recorded as 180 to 200, and the grayscale value range of the center region as 100 to 120. The radial grayscale gradient difference was calculated. If the difference exceeded a threshold of 60, it was marked as a significant grayscale transition region. Next, the grayscale change frequency of each scan line was counted along four directions in the same image patch: horizontal, vertical, upper left to lower right, and upper right to lower left. The main texture direction was determined based on a frequency difference greater than 8. Further, the spectral reflectance data corresponding to this region was obtained. The 30-band spectral reflectance associated with the pixels corresponding to significant gray-level gradient points in each image block is extracted. The reflectance change rate of these 30 sets of spectral data between adjacent bands is calculated. Each change rate is the absolute value of the difference in reflectance between adjacent bands. In the spectral energy change rate curve, if the change rate of three consecutive bands is greater than 0.06, the position is marked as an energy mutation point. Then, the image coordinate position difference between significant gray-level gradient points on the main axis of texture direction in the image block and energy mutation points is compared. When the difference is within 2 pixels, it is determined to be a synchronization fluctuation point. The synchronization tolerance is set to 2 pixels. Finally, the position points that meet the above synchronization fluctuation conditions are selected in all image blocks to obtain the texture spectral synchronization point set.

[0027] The type labeling submodule extracts the corresponding texture frequency and spectral principal energy value based on the texture spectrum synchronization point set, calls the stone type feature data, compares the numerical similarity threshold, and obtains the stone principal type labeling result; First, the texture frequency parameters and the spectral principal energy values ​​of each point in the synchronization point set are extracted. The texture frequency is obtained by statistically analyzing the periodic variation of gray values ​​along the main texture direction within a unit area. If there are 5 gray-level periodic fluctuations in the main texture direction within a 64×64 pixel block, the texture frequency is 5 times per block, which is converted to a frequency per pixel of 0.078. The spectral principal energy value is recorded by selecting the maximum reflectance value and its corresponding band position from the 30-band reflectance of each point. For example, if the reflectance of a point reaches 0.74 in band 18, it is the spectral principal energy value of that point. At the same time, the stone type feature data table is called, which contains the typical texture frequency range and spectral principal energy range for three stone types: calcium oxalate type, uric acid type, and magnesium ammonium phosphate type. Among them, the texture frequency of calcium oxalate type is between 0.05 and 0.08 per pixel, and the spectral principal energy is between 0.7 and 0.78. For the magnesium ammonium phosphate type, the texture frequency is between 0.02 and 0.04, and the spectral principal energy is between 0.5 and 0.65. For the magnesium ammonium phosphate type, the texture frequency is between 0.08 and 0.1, and the spectral principal energy is between 0.6 and 0.7. The texture frequency and spectral principal energy values ​​of each extracted synchronization point are compared with the corresponding data of the three types of stones. The differences are then normalized. For example, the extracted point frequency is 0.078, and the difference between it and the average texture frequency of the calcium oxalate type (0.065) is 0.013, with a normalized value of 0.2. The difference between the spectral energy value of 0.74 and the average principal energy of the calcium oxalate type (0.74) is 0, with a normalized value of 0. The average of the normalized differences of frequency and energy is calculated to obtain a comprehensive similarity of 0.1. The comprehensive similarity of all synchronization points is averaged across all types of stones. If the average similarity of a certain type is the smallest and less than 0.2, it is used as the judgment threshold, and then that type is labeled as the main type of stone. Finally, the main type labeling result of the stone is obtained.

[0028] Specifically, such as Figure 2 , 4 As shown, the component correlation determination module includes: The type information docking submodule, based on the type information in the main type labeling results of the stone, calls the blood free calcium sequence and urine calcium excretion sequence within the same period, and obtains a dual-channel data group according to the data corresponding to the time point; First, identify the stone type corresponding to the stone labeling results. For example, when the main type is calcium oxalate, extract the formation cycle information corresponding to this type. The formation cycle is uniformly set to 7 consecutive days according to the test records. By accessing the blood calcium test records within this cycle, obtain the blood free calcium concentration data at a fixed time point each day. For example, if the recording time is set to 8:00 AM every day, read the corresponding concentration data for 7 consecutive days from the database as 2.35, 2.32, 2.38, 2.34, 2.31, 2.36, and 2.33, in mmol / L. At the same time, call the urinary calcium excretion sequence and extract each urination at the same date and time point. The urine calcium concentration values ​​after urination, if recorded 15 minutes after urination, are recorded as 4.1, 4.3, 4.2, 4.0, 4.4, 4.1, and 4.2, in mmol per liter. Using a unified time label, the two sequences are indexed and aligned by date and time to generate a one-to-one corresponding dual-channel data set. For example, 8:00 AM on day 1 corresponds to a blood calcium concentration of 2.35 and a urine calcium concentration of 4.1, day 2 corresponds to 2.32 and 4.3, and so on. If some time points are missing data, the time pair is marked as empty and the pairing process is skipped. After pairing all non-empty data, the final dual-channel data set is obtained.

[0029] The direction extraction submodule is based on the dual-channel data set to extract the direction of change of blood calcium and urinary calcium. It compares the direction signs to identify time periods with consistent directions, removes segments with reversed directions, and obtains a sequence of segments with consistent directions. First, for each pair of blood calcium concentration and urine calcium concentration in the data set, calculate the change between adjacent time points. Define the sign of the difference between the current time point and the previous time point as the direction of change. If the current blood calcium value is higher than the previous day, the direction is defined as "+1", if it is lower than the previous day, it is defined as "-1", and if it is equal to the previous day, it is defined as "0". The same rule is used to mark the direction of change of the urine calcium sequence, forming two direction sequences. For example, the blood calcium concentration sequence is 2.35, 2.32, 2.38, 2.34, 2.31, and the direction sequence is "-1", "+1", "-1", "-1"; the urine calcium concentration sequence is 4.1, 4.3, 4.2, 4.0, 4.4, and the direction sequence is "+1", "-1", "-1", "+1". The two directional sequences are then compared daily to determine whether the blood calcium directional value and the urinary calcium directional value are the same at each time point: if both directional values ​​are "+1" or both are "-1", it is a point of directional consistency; if one is "+1" and the other is "-1", it is a point of directional reversal; if there is "0" or the directions are different, it is marked as inconsistent. After obtaining the alignment result sequence, the sequence is scanned sequentially from the beginning to identify consecutive points of directional consistency and mark them as a segment of directional consistency. When a point of directional reversal or inconsistency is encountered, the current segment is immediately terminated and a new judgment begins, thus forming multiple segments of directional consistency. The start and end times and length of each segment can be used for further analysis. During this process, points of directional reversal and their adjacent discontinuous regions are removed to ensure that only stable segments with high directional consistency are retained, ultimately resulting in a sequence of directional consistency segments.

[0030] The state feature extraction submodule calls the segment sequence with the same direction, extracts the rate of change of blood calcium and urine calcium in the corresponding time period, monitors the fluctuation range between rate values ​​in continuous segments, and obtains the result of component activity status determination. First, for each consecutive time point within a consistent directional segment, calculate the rate of change of blood calcium between adjacent points within that segment. The rate is defined as the current concentration minus the concentration at the previous time point divided by the time interval, which is uniformly set to 1 day. If the blood calcium concentrations from day 1 to day 4 are 2.30, 2.34, 2.31, and 2.36 respectively, then their adjacent rates are +0.04, -0.03, and +0.05 respectively. The urinary calcium sequence is processed in the same way. For example, if the urinary calcium concentrations are 4.1, 4.2, 4.4, and 4.3 respectively, the rates of change are +0.1, +0.2, and -0.1. Then, calculate the amplitude of fluctuation in the blood calcium and urinary calcium rate sequences respectively. The amplitude of fluctuation is defined as the difference between the maximum and minimum rates within the current segment. For the above blood calcium rate sequence, the maximum value is +0.05, and the minimum value is -0. The fluctuation range is 0.03, with a maximum value of +0.2 and a minimum value of -0.1 for the urinary calcium rate sequence, and a fluctuation range of 0.3. To avoid interference from extreme values, a standard threshold judgment range is set for the fluctuation range. For example, a fluctuation range of less than 0.04 for serum calcium is considered a stable range, between 0.04 and 0.08 is considered a moderate fluctuation range, and greater than 0.08 is considered a high fluctuation range. Similarly, a fluctuation range of less than 0.1 for urinary calcium is considered a stable range, between 0.1 and 0.25 is considered a moderate fluctuation range, and greater than 0.25 is considered a high fluctuation range. Based on the combination of the intervals in which serum calcium and urinary calcium are located, a state judgment rule is set. If both are high fluctuations, it is judged as a high-activity state; if one is high fluctuation and the other is moderate or stable, it is judged as a moderately active state; and if both are stable or below moderate, it is judged as a low-activity state. The final result of the component activity state judgment is obtained.

[0031] Specifically, such as Figure 2 , 5 As shown, the acid-base state classification module includes: The time segment extraction submodule, based on the time period data in the component activity state determination result, calls the urine acid-base reaction direction change sequence within the corresponding time range, and simultaneously extracts the lactic acid concentration sequence and bicarbonate concentration sequence within the time period, aligns the data according to the time point, and obtains a three-sequence time point combination set; First, the start and end times of each segment marked as highly active or moderately active are read from the judgment results. For example, segment 1 starts on day 2 and ends on day 5. Then, data is retrieved sequentially for each time period. When retrieving the urine acid-base reaction direction change sequence, the urine pH value within the corresponding time period is extracted first. The daily recording time is fixed at 8:00 AM. If the urine pH values ​​from day 2 to day 5 are 5.8, 6.0, 6.4, and 6.1 respectively, the reaction direction is alkalinity increase, alkalinity increase, and acidity increase. The direction is determined based on the pH value changes between adjacent days. The direction of pH increase is marked as "+1", decrease as "-1", and no change as "0", resulting in a direction sequence of "+1", "+1", and "-1". Next, the lactate concentration sequence is extracted simultaneously within the same time period. For example, the daily lactate values ​​are 1.1, 1.4, 1.3, and 1.2 mmol / L, and the bicarbonate concentrations are 24, 22, 23, and 25 mmol / L, respectively. These three data points are uniformly labeled with time indices T2 to T5 according to their respective recording dates. A three-column data correspondence table is constructed, representing the time point, lactate concentration, and bicarbonate concentration. Each row corresponds to a date. After aligning the data by time point, a three-sequence time point combination set is formed. For example, the data in row T2 is lactate 1.1, bicarbonate 24, pH "+1", T3 is 1.4, 22, "+1", and T4 is 1.3, 23, "-1". This process is repeated for each time point within the active period, and all time periods determined to be active are processed to finally obtain the three-sequence time point combination set.

[0032] The trend direction comparison submodule extracts the trend direction of the lactate concentration sequence and the bicarbonate concentration sequence based on the three sequence time point combination set, compares the change direction of each time period, identifies the interval with consistent direction and the interval with directional deviation, and obtains the trend comparison interval set. First, data points for every two consecutive days from the lactate and bicarbonate concentration sequences are sequentially read from the paired sets. The direction of the difference between each pair of adjacent values ​​is calculated chronologically. If the concentration at the later time point is higher than the previous time point, the change direction is marked as "+1"; if it is lower, it is marked as "-1"; and if they are equal, it is marked as "0". Taking days 1 to 4 as an example, the lactate concentration values ​​are 1.1, 1.4, 1.3, and 1.2 mmol / L, with a direction sequence of "+1", "-1", and "-1". The bicarbonate concentration values ​​are 24, 22, 23, and 25 mmol / L, with a direction sequence of "-1", "+1", and "+1". After generating the two trend direction sequences, the direction signs at corresponding positions are compared one by one according to the time points. If they are both "+", the direction is changed accordingly. If the sign is "1" or both are "-1", the direction of the time period is considered consistent. If the signs are opposite, the direction is considered to be offset. If one sign is "0", the time period is skipped and not included in the recognition area. The entire pairing sequence is traversed, and all consecutive time periods with consistent direction are merged into a consistent direction interval. Time periods with interrupted consistency are classified as offset direction intervals. For example, if T2 to T3 is lactic acid "+1" and bicarbonate "-1", it is an offset. If T3 to T4 is lactic acid "-1" and bicarbonate "+1", it is still an offset. If both are "-1" from T4 to T5, the segment is considered a consistent interval. Segments with a consistent direction interval length of 2 days or more are extracted separately as the core segment for trend comparison. Time segments with inconsistent direction signs are recorded as offset direction subsets. Finally, the set of trend comparison intervals is obtained.

[0033] The reaction structure separation submodule is based on the trend comparison interval set. It extracts the reaction direction change state according to the trend consistency segment and the offset segment, and classifies them into the differentiated reaction segment type to obtain the acid-base reaction segment structure result. First, the start and end times of each segment marked as either a consistent-direction segment or a deviating-direction segment in the trend comparison interval set are read and used as the time boundaries for subsequent classification processing. When processing consistent-direction segments, starting from the start time of the segment, the directional signs of the lactate and bicarbonate concentrations for each day within that segment are read sequentially. The directional sign is formed by the positive or negative difference between the values ​​of the previous day and the current day. If the lactate concentration of the following day is higher than that of the previous day, it is marked as "+1", otherwise "-1", and if they are equal, it is marked as "0". Taking the example segment T2 to T5, the lactate concentration sequence is 1.1, 1.4, 1.3, 1.2, and the directions are "+1" and "-1". The bicarbonate concentrations are 24, 22, 23, and 25, with directions of "-1", "+1", and "+1". The direction signs in a consistent direction segment must remain consistent for several consecutive days. Therefore, during the traversal, the equality condition is determined by comparing the direction signs of adjacent rows item by item. The equality condition is that the direction values ​​are exactly the same. If the direction is "-1" for three consecutive days, it is marked as a consistent reaction direction starting from the first day, and this segment is classified into the consistent reaction segment type. In the above process, for direction-offset segments, the lactic acid and bicarbonate direction values ​​are read point by point according to the range of the offset segment, and the symbol is executed in each pair of direction values. If the two signs are unequal or one is positive and the other negative, the time point is recorded as an offset reaction point, and the entire segment is classified into the offset reaction segment type. Simultaneously, the magnitude of directional change within the offset segment is further assessed. The magnitude of directional change is characterized by the difference in concentration between lactic acid and bicarbonate over two consecutive days. If the absolute value of the lactic acid difference is greater than 0.2 and the absolute value of the bicarbonate difference is greater than 2, the time point is marked as a high-amplitude offset point; otherwise, it is a low-amplitude offset point. For example, if lactic acid changes from 1.3 to 1.2 (a difference of 0.1) and bicarbonate changes from 23 to 25 (a difference of 2), the lactic acid difference is determined to be low-amplitude, and the bicarbonate difference to be high-amplitude. The value combination classifies this point into the mixed amplitude subclass within the offset segment. In the above classification process, a reaction direction benchmark value is set to determine whether the direction sign has a change meaning. This benchmark value is set at 0.05 mmol / L for lactic acid and 1 mmol / L for bicarbonate. The setting method refers to the detection resolution and daily average change, and is combined with the actual measured sequence. Taking the change of 0.3 from 1.1 to 1.4 for lactic acid as an example, the benchmark value of 0.05 is a reasonable range. Finally, after traversing all trend control intervals, the segments with consistent direction are classified into the consistent reaction structure, and the segments with offset direction are classified into the offset reaction structure, thus obtaining the acid-base reaction segment structure results.

[0034] Specifically, such as Figure 2 , 6 As shown, the metabolic behavior ranking module includes: The behavioral data extraction submodule, based on the time period in the acid-base reaction segment structure results, obtains the urinary calcium excretion change direction, water drinking interval change data and urination rhythm trajectory information within the cycle. According to the behavioral item corresponding to the time point, it extracts the directional change sequence of the behavioral item within the cycle to obtain the behavioral direction time series set. First, the start and end times of each segment are read, and this time period is used as the data extraction window. For example, if an acid-base reaction segment starts on day 3 and ends on day 6, then this window covers four time points: T3, T4, T5, and T6. Within this time range, the direction of urinary calcium excretion is first obtained by reading the daily average urinary calcium excretion value. If the value sequence is 4.1, 4.3, 4.2, and 4.0 mmol / L, then the difference sequence is +0.2, -0.1, and -0.2, with the direction markers being "+1", "-1", and "-1" respectively. Next, the data on changes in water intake intervals are read. This data records the number of hours between the first and last water intake each day, for example, 3.0, 2.5, 2.7, and 3.1 hours respectively. The difference sequence is -0.5, +0.2, and +0.4, with the direction sequence being "-1". The data is incremented by 1, 2, and 1. Then, the urination rhythm trajectory information is extracted. The urination rhythm is based on the number of urinations per day to represent the rhythm activity. If the record is 6 times, 5 times, 7 times, and 6 times, the change is -1, +2, and -1, respectively. The direction is marked as "-1", "+1", and "-1". These three behavioral items are aligned with the time intervals of the acid-base reaction. At each time point Tn, the three behavioral direction mark values ​​are summarized to form a dataset with time as the index and the three behavioral directions as the content. For example, the behavioral direction corresponding to T4 is urinary calcium "-1", water drinking interval "+1", and urination rhythm "+1". A behavioral direction time series is constructed. If the behavioral item data is missing, it is marked as invalid at that time point and the combination at that point is skipped. This process is repeated for all acid-base reaction intervals to finally obtain the behavioral direction time series set.

[0035] The direction change tracking submodule is based on the behavioral direction time series set. It compares the change points of each type of behavioral direction within the period, extracts the occurrence number of direction switching actions, compares the difference in change frequency of behavioral items within the time period, and obtains behavioral direction fluctuation feature groups. First, the behavior sequences are split into three categories based on the behavior item type: direction of urinary calcium excretion, direction of change in water drinking interval, and direction of urination rhythm. The direction value sequence of each category is traversed point by point, recording whether the direction value changes sign between adjacent time points. The direction value is defined as "+1", "-1", or "0". A sign change means that the direction value at the current time point has a different sign than the direction value at the previous time point and is not "0". For example, if "+1" changes to "-1" or "-1" changes to "+1", it is counted as one direction switch. If the urinary calcium direction is "+1" or "-1" for two consecutive days, the switch count is incremented by 1. If it is "-1" or "0" or "0" or "-1", it is not counted as a switch. After traversal, the number of switches for each category of behavior item within the cycle is counted, obtaining the number of urinary calcium behavior switches (n1), the number of water drinking interval switches (n2), and the number of urination rhythm switches (n3). The entire cycle is also recorded. The effective number of days T for each behavior item is used for subsequent frequency calculation. The number of switching times for each type is divided by T to obtain the change frequency. For example, if the cycle T is 7 days and n1 is 4 times, then the change frequency of urinary calcium direction is 0.57 times per day. The benchmark for evaluating the change frequency difference is set as a relative interval of 0.2 times per day. When the difference between the change frequencies of any two behaviors is greater than 0.2, it is marked as a significant difference; otherwise, it is considered close. If urinary calcium is 0.57 and the water drinking interval is 0.14, then the difference is 0.43, which is marked as a significant difference. At the same time, the behavior with the highest frequency among the three types is screened and recorded as the dominant fluctuating behavior item. The frequency differences are sorted to form a frequency ranking vector from dominant to secondary. If the urination rhythm frequency is 0.71, then the ranking is urination rhythm > urinary calcium > water drinking interval. This ranking vector is used to describe the main change trend of the behavior direction. Finally, the number of switching times, frequency values ​​and frequency ranking of each behavior item within the cycle are combined to obtain the behavior direction fluctuation feature group.

[0036] The behavior ranking output submodule tracks the directional continuity trajectory of behavior items within a cycle based on the behavior direction fluctuation feature group, and adjusts the position of behavior before and after intervention according to the distribution of change frequency and directional trend changes, so as to obtain the ranking result of metabolic behavior intervention. First, each type of behavior item in the feature group is sorted according to its frequency of directional change within the cycle. For example, if the urination rhythm frequency is 0.71 times per day, urinary calcium excretion is 0.57 times per day, and water drinking interval is 0.14 times per day, the preliminary sorting is urination rhythm > urinary calcium excretion > water drinking interval. Then, the directional continuity trajectory of each type of behavior item is extracted. The directional continuity trajectory is used to determine whether the behavior item maintains the same directional sign over multiple consecutive time points. Taking urination rhythm as an example, if its directional sequence in a 7-day cycle is "+1", "+1", "+1", "+1", "+1", "+1", it indicates that the behavioral direction is consistent throughout the entire cycle and can be identified as a stable continuation segment with a length of 7 days. However, if the directional sequence is "+1", "+1", "+1", "-1", "-1", "+1", "+1", it only maintains consistency in certain time periods, representing multiple discontinuous directional continuations, with the longest continuous segment being 3 days. Based on the number of stable durations and their longest duration, behavioral items can be classified: if there are 7 consecutive days with the same direction, it is classified as low-fluctuation, high-continuity; if the duration is short and the direction reverses multiple times, it is classified as high-fluctuation, low-continuity. Based on this type of difference, a priority rule for behavioral item intervention is established: if a behavioral item has a high frequency of change (>0.6) and the longest continuous duration is less than 3 days, it should be classified as a priority intervention item; conversely, if its frequency of change is low (<0.2) and there are more than 5 days of continuous duration with the same direction, it can be classified as a delayed intervention item. In this example, frequent changes in the direction of urination rhythm and weak continuity are classified as priority intervention; urinary calcium excretion shows some directional reversals but some segments remain consistent, classified as medium priority intervention; water drinking intervals are consistent in direction within a 7-day cycle and have extremely low fluctuations, classified as delayed intervention. Finally, the three types of behaviors are renumbered according to this logical model, resulting in the final ranking of metabolic behavioral interventions.

[0037] Specifically, such as Figure 2 , 7 As shown, the path scheme construction module includes: The trend data extraction submodule obtains the uric acid change direction sequence and urinary calcium change direction sequence within the corresponding time period based on the behavioral cycle in the ranking results of metabolic behavior intervention, divides the trend data according to the cycle, and obtains the trend direction cycle sequence group. First, based on the intervention cycles corresponding to various behaviors in the intervention ranking results, specific time periods were divided. For example, the cycle corresponding to urination rhythm behavior was days 1 to 5, the cycle corresponding to urinary calcium excretion behavior was days 2 to 6, and the cycle corresponding to water drinking interval behavior was days 3 to 7. The start and end times of each behavior cycle were read sequentially as the basis for trend data segmentation. Within each cycle, the uric acid concentration and urinary calcium concentration values ​​at the corresponding time points were extracted, and the direction of the difference between two adjacent time points was calculated. If the value of the next day was higher than that of the previous day, it was marked as "+1", lower as "-1", and equal as "0". Taking the urination rhythm cycle as an example, the uric acid concentrations from day 1 to 5 were 5.8, 6.1, 5.9, 5.7, and 5.6 mmol / L, corresponding to the direction sequence of "+1", "-1", "-1", and "-1", and the urinary calcium concentrations were 4.2, 4.1, 4.3, and 4.0 mmol / L. For a uric acid level of 4.1 mmol / L, the corresponding directional sequences are “-1”, “+1”, “-1”, and “+1”, forming two directional sequences within this cycle. Then, the urinary calcium excretion cycle is processed. If uric acid is 6.1, 5.9, 5.7, 5.8, or 6.0 mmol / L, the corresponding directions are “-1”, “-1”, “+1”, and “+1”; if urinary calcium is 4.1, 4.3, 4.0, 4.1, or 4.2 mmol / L, the corresponding directions are “+1”, “-1”, “+1”, and “+1”. This process is repeated until all behavioral cycles are covered. For each cycle, a set of uric acid and urinary calcium directional sequences is constructed. These are grouped and summarized according to cycle labels, numbered P1, P2, P3, etc. Each group contains uric acid and urinary calcium directional subsequences, corresponding to the intervention behavior items in the behavioral ranking results. Finally, the trend direction cycle sequence group is obtained.

[0038] The cycle direction comparison submodule compares the uric acid direction and urinary calcium direction in the cycle point by point based on the trend direction cycle sequence group, distinguishes the trend direction of the cycle segment, and obtains the trend consistency partitioning result. First, expand each data group sequentially according to its period label. Read the uric acid and urinary calcium direction subsequences for that period, and compare the direction signs at corresponding positions in the two sequences at the same time point. The direction sign value ranges from "+1", "-1", or "0". If the two direction signs at the same time point are completely consistent (both are "+1" or both are "-1"), then that time point is marked as a point of consistent direction. If one is "+1" and the other is "-1", then it is marked as a point of inconsistent direction. If one is "0", then that time point is excluded from the judgment. After traversing all comparable time points in the current period, count the number of consistent direction points n1 and the number of inconsistent direction points n2, and calculate the consistency ratio n1 / (n1+n2). Set the consistency judgment threshold to 0.6, that is, if a certain If the proportion of consistent points in a cycle is greater than or equal to 0.6, the cycle is determined to be a trend-consistent cycle. If it is less than 0.6, it is determined to be a trend-deviation cycle. For example, in cycle P1, there are 4 valid time points, of which 3 are consistent points in direction and 1 is inconsistent point in direction. The consistency ratio is 0.75, which is greater than 0.6, so it is determined to be a trend-consistent cycle. If there is 1 consistent point and 3 inconsistent points in cycle P2, the consistency ratio is 0.25, which is less than the threshold, so it is determined to be a trend-deviation cycle. After completing the traversal of all cycles and direction comparison, each cycle is classified into two categories, "Trend Consistent Segment" and "Trend Deviation Segment," according to the consistency results. The cycle number, start and end time points, number of valid comparisons, number of consistent points, number of inconsistent points, and consistency ratio are recorded to obtain the trend consistency partitioning results.

[0039] The path segment merging submodule connects differentiated periodic segments according to the trend consistency partitioning results and the time progression order, connects the behavior ranking information with the trend direction status, extracts the intervention segment range in the continuous advancement path, and obtains the individual prevention and control path segment combination results. First, all trend-consistent and trend-deviation segments are arranged chronologically. Then, they are rearranged in ascending order according to the start time of each cycle number, starting with the first cycle. The type marker, start and end time of the current cycle, and the type marker of the next cycle are sequentially read for continuity assessment. If the current cycle and the next cycle have different types, this is identified as a point of transition between different types, and this point is marked as a transition boundary. Simultaneously, the behavioral ranking information within the current cycle is extracted. This information comes from the dominant behavioral item corresponding to the cycle in the aforementioned behavioral intervention ranking results. For example, if the dominant behavior in cycle 3 is urination rhythm and in cycle 4 is water drinking interval, the change in the dominant behavioral item at the transition point between the two cycles is identified and treated as a path turning point. Then, the trend direction state of the current cycle is connected with the dominant behavioral item in the structure to construct path units. Each path unit includes cycle type, behavioral item name, and behavior. The sorting position and trend direction labels are recorded as follows: Path unit P3, type is consistent segment, main behavior is urination rhythm, sorted first, trend is stable upward direction. Then, multiple consecutive path units are spliced ​​to form a continuous advancement path. The splicing rule requires that there are no missing cycles in adjacent path units. If data is missing in a certain cycle, the splicing process is interrupted and the path construction starts again from the next complete cycle. Each path advancement segment must contain at least two path units and satisfy the requirement that the ranking information of the behavior items shows a significant change trend in the path, such as the sorting position changing from high to low, or the dominant behavior item changing from urinary calcium to drinking interval, etc. Then it can be determined that the path advancement segment has intervention significance. All such path segments are screened, and their covered time range and the combination of the main behaviors they contain are extracted. The path segment numbers are uniformly assigned, such as R1, R2, R3, and categorized for output. Finally, the individual prevention and control path segment combination results are obtained.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A system for intelligent identification of kidney stone components and generation of individualized prevention and treatment plans, characterized in that, The system includes: The stone type identification module acquires images and spectral data of the excrement, locates the edge and center regions of the crystals, identifies image segments that change synchronously in spatial location, and obtains the main type labeling results of the stones. Based on the main type of stone labeling results, the component association determination module calls the records of changes in free calcium and excreted calcium within the period, compares the trend direction in chronological order, analyzes the trend extension behavior, and obtains the component activity status determination results. Based on the results of the determination of the active state of the components, the acid-base state division module compares the direction of acid-base reaction with the trend of lactic acid and bicarbonate concentration changes, identifies the direction of the periodic trend, and obtains the structural results of the acid-base reaction segment. Based on the acid-base reaction segment structure results, the metabolic behavior ranking module obtains the direction of urinary calcium excretion, water drinking interval records and urination rhythm trajectory, compares the frequency of directional changes in behavior within the cycle, updates the behavioral trend changes, and obtains the metabolic behavior intervention ranking results. Based on the metabolic behavior intervention ranking results, the path scheme construction module extracts the trend direction of uric acid and urinary calcium, compares the directional relationship of each cycle, updates the cycle trend status, and obtains the individual prevention and control path segment combination results.

2. The intelligent identification and personalized prevention and treatment plan generation system for kidney stones according to claim 1, characterized in that, The main stone type labeling results include crystal edge region identification, central region spectral feature extraction, texture change and peak position linkage region; the component activity status determination results include blood calcium trend direction consistent segment, urinary calcium trend direction consistent segment, trend structure extension segment, and direction coherence behavior feature; the acid-base reaction segment structure results include urinary acid-base reaction trend segment, lactic acid trend direction segment, bicarbonate trend direction segment, trend direction consistent continuous segment, and trend direction offset continuous segment; the metabolic behavior intervention ranking results include urinary calcium excretion change direction item, water drinking interval change item, urination rhythm change item, direction change frequency difference item, and direction trend tracking item; the individual prevention and treatment path segment combination results include uric acid trend matching period segment, urinary calcium trend matching period segment, trend direction consistent period combination, and trend direction inconsistent control period segment.

3. The intelligent identification and personalized prevention and treatment system for kidney stones according to claim 1, characterized in that, The record of changes in free calcium and excreted calcium refers to a data set that synchronously records and compares the concentrations of free calcium in human blood and excreted calcium in urine at the same time point within the same period. The direction of the acid-base reaction refers to the comparison between the trend of acid-base changes in urine and the trend of changes in lactic acid and bicarbonate concentrations within the target time period.

4. The intelligent identification and personalized prevention and treatment system for kidney stones according to claim 1, characterized in that, The urination rhythm trajectory refers to the changing trend of the number of urinations per day and the time distribution within a period; The cycle trend status refers to the comparison of the trend directions of uric acid and urinary calcium within the metabolic cycle to determine the consistency of the trend in each cycle.

5. The intelligent identification and personalized prevention and treatment system for kidney stones according to claim 1, characterized in that, The stone type identification module includes: The image and spectrum acquisition submodule acquires the surface of the kidney stone excrement, collects surface images and spectral data of the corresponding areas, encodes image pixels, decomposes spectral channels, and associates image coordinates with spectral positions point-to-point to obtain corresponding image spectral groups. The regional feature localization submodule, based on the image spectrum correspondence group, calls the gray-level changes of the crystal edge and center region in the image, the differential region texture direction, and combines the spatial distribution characteristics of the corresponding spectral energy values ​​to extract positional data representing the texture change trend, thereby obtaining the texture spectrum corresponding point set; The type labeling submodule extracts the corresponding texture frequency and spectral main energy value based on the texture spectrum synchronization point set, calls the stone type feature data, compares the numerical similarity threshold, and obtains the stone main type labeling result.

6. The intelligent identification and personalized prevention and treatment system for kidney stones according to claim 1, characterized in that, The component association determination module includes: The type information docking submodule, based on the type information in the main type labeling result of the stone, calls the blood free calcium sequence and urine calcium excretion sequence within the same period, and obtains a dual-channel data group according to the data corresponding to the time point; The direction extraction submodule extracts the direction of change of blood calcium and urine calcium based on the dual-channel data set, compares the direction signs, identifies time periods with consistent directions, removes segments with reversed directions, and obtains a sequence of segments with consistent directions. The state feature extraction submodule calls the directional consistent segment sequence to extract the rate of change of blood calcium and urine calcium within the corresponding time period, monitors the fluctuation range between rate values ​​within continuous segments, and obtains the result of component activity status determination.

7. The intelligent identification and personalized prevention and treatment plan generation system for kidney stones according to claim 6, characterized in that, During the process of extracting the changes in blood calcium and urine calcium, the numerical relationship between two adjacent time points of the blood free calcium sequence and the urine calcium excretion sequence is compared to analyze the direction of change. During the process of removing segments with reversed direction, the time points of the direction change are identified along the sequence of change direction, and the data are separated by the time points. In the process of extracting the rate of change of blood calcium and urinary calcium within the corresponding time period, the blood calcium and urinary calcium data corresponding to the first and last time points in the time period with the same direction are extracted and arranged in chronological order; in the process of monitoring the fluctuation range between rate values, the trend of change of blood calcium and urinary calcium data in the time period with the same direction is compared segment by segment according to the chronological order to identify the data segments in which the change trend has different amplitudes.

8. The intelligent identification and personalized prevention and treatment plan generation system for kidney stones according to claim 1, characterized in that, The acid-base state classification module includes: The time segment extraction submodule, based on the time period data in the component activity state determination result, calls the urine acid-base reaction direction change sequence within the corresponding time range, and simultaneously extracts the lactic acid concentration sequence and bicarbonate concentration sequence within the time period, aligns the data according to the time point, and obtains a three-sequence time point combination set. The trend direction comparison submodule extracts the trend direction of the lactate concentration sequence and the bicarbonate concentration sequence based on the three sequence time point combination set, compares the change direction of each time period, identifies the interval with consistent direction and the interval with offset direction, and obtains the trend comparison interval set. Based on the trend comparison interval set, the reaction structure separation submodule extracts the reaction direction change state according to the trend consistency segment and the offset segment, and classifies them into differentiated reaction segment types to obtain the acid-base reaction segment structure results.

9. The intelligent identification and personalized prevention and treatment plan generation system for kidney stones according to claim 1, characterized in that, The metabolic behavior ranking module includes: The behavioral data extraction submodule, based on the time period in the acid-base reaction segment structure result, obtains the urinary calcium excretion change direction, water drinking interval change data and urination rhythm trajectory information within the cycle, and extracts the directional change sequence of the behavioral item within the cycle according to the time point corresponding to the behavioral item, thus obtaining a behavioral direction time series set. The direction change tracking submodule, based on the behavior direction time series set, compares the change points of each type of behavior direction within the period, extracts the occurrence frequency of direction switching actions, compares the difference in change frequency of behavior items within the time period, and obtains behavior direction fluctuation feature groups. The behavior ranking output submodule tracks the directional continuity trajectory of behavior items within a period based on the behavior direction fluctuation feature group, and adjusts the position of the behavior before and after the intervention according to the distribution of the change frequency and directional trend change, so as to obtain the metabolic behavior intervention ranking result.

10. The intelligent identification and personalized prevention and treatment plan generation system for kidney stones according to claim 1, characterized in that, The path scheme construction module includes: The trend data extraction submodule obtains the uric acid change direction sequence and urinary calcium change direction sequence within the corresponding time period based on the behavior cycle in the ranking results of the metabolic behavior intervention, divides the trend data according to the cycle, and obtains the trend direction cycle sequence group. The cycle direction comparison submodule compares the uric acid direction and urinary calcium direction in the cycle point by point based on the trend direction cycle sequence group, distinguishes the trend direction of the cycle segment, and obtains the trend consistency partitioning result. The path segment merging submodule, based on the trend consistency partitioning results, connects the differentiated periodic segments in chronological order, connects the behavior sorting information with the trend direction status, extracts the intervention segment range in the continuous advancement path, and obtains the individual prevention and control path segment combination results.

Images