A medical health data collaborative processing system and method based on traditional Chinese and western medicine feature evolution for lung cancer whole cycle

Abstract

The present application relates to the technical field of health care health science, and discloses a medical health data collaborative processing system and method based on feature evolution of traditional Chinese medicine and western medicine for lung cancer whole cycle, which comprises the following steps: synchronously acquiring traditional Chinese medicine phenotype feature data, spatial position information and western medicine objective detection data of a target object at multiple time sequence follow-up nodes; constructing space-time covariant features based on the spatial position information; generating standardized traditional Chinese medicine syndrome labels by using a knowledge base; constructing a structural equation model, taking the traditional Chinese medicine phenotype features and the western medicine objective detection data as parallel observation indexes, calculating factor load, and generating a traditional Chinese medicine and western medicine fusion feature vector; activating a dynamic evolution network, inputting the fusion vector and the space-time covariant into a pre-trained state transition model, calculating a syndrome state transition probability matrix, generating a risk prediction index, and triggering a hierarchical early warning. The present application realizes a paradigm shift from static evaluation to dynamic evolution law description, and provides an objective and quantitative evidence-based framework for individualized diagnosis and treatment of lung cancer.

Description

Technical Field

[0001] This invention relates to the field of medical and health sciences, and in particular to a collaborative processing system and method for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine throughout the entire lung cancer cycle. Background Technology

[0002] Lung cancer is one of the leading causes of cancer-related deaths worldwide, and surgical resection is the primary treatment for early-stage non-small cell lung cancer. During postoperative recovery, symptom management and dynamic identification and intervention of TCM syndromes are crucial for improving quality of life and prolonging survival.

[0003] Current methods for assessing postoperative symptoms in lung cancer often rely on static scales (such as the EORTC QLQ-LC13) or electronic patient-reported outcomes (ePRO) tools. These methods suffer from low data collection frequency and information lag, making it difficult to reflect the dynamic evolution of symptoms and TCM syndromes over time. Furthermore, the determination of TCM syndromes largely depends on physician experience, lacking objective and standardized quantitative evidence, leading to poor diagnostic consistency among different physicians. Some studies have attempted to collect tongue, facial, and pulse information using digital devices, but these remain at the level of feature extraction at a single time point and have not established structured mathematical correlations with Western medical indicators.

[0004] Furthermore, existing methods lack a clear hierarchical mapping mechanism between symptoms and syndromes. While traditional clustering or factor analysis can identify symptom clusters, they fail to systematically couple these clusters with TCM syndromes, and they do not consider the transition patterns of syndrome states over time. Clinically, patients may experience syndrome evolution from "Qi deficiency" to "Yin deficiency" and then to "phlegm and blood stasis" after surgery, but existing static models cannot characterize this process or quantify the transition probabilities between different syndrome states, thus hindering accurate dynamic syndrome differentiation and prognostic early warning.

[0005] Therefore, there is an urgent need for a system and method that can integrate multimodal data from traditional Chinese and Western medicine, construct a symptom-syndrome hierarchy, and model dynamic evolution patterns to support individualized diagnosis and treatment and risk management for lung cancer patients. Summary of the Invention

[0006] This invention provides a collaborative processing system and method for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine throughout the entire lung cancer cycle, in order to overcome the shortcomings of existing technologies.

[0007] This invention provides a collaborative medical and health data processing system based on the evolution of traditional Chinese and Western medicine characteristics throughout the entire lung cancer cycle. The system includes an interactive data acquisition terminal, a geospatial positioning and sensing component, a medical device communication interface, and a central control processor. The interactive acquisition terminal is configured to collect TCM phenotypic feature data of the target object at multiple time-series follow-up nodes throughout the entire lung cancer cycle; The geospatial positioning sensing component is configured to synchronously collect spatial location information of the target object at each time-series follow-up node; The medical device communication interface is configured to obtain objective Western medicine test data of the target object at the corresponding time-series follow-up node from an external medical information system; The central control processor is configured to receive TCM phenotypic feature data collected by the interactive acquisition terminal, spatial location information collected by the geospatial positioning sensing component, and Western medicine objective test data obtained by the medical device communication interface, and execute the following logic: Based on spatial location information, the spatiotemporal covariate features of the corresponding time-series follow-up nodes are constructed by calling the external environment database. The TCM lung cancer syndrome diagnosis knowledge base is used to perform rule matching on TCM phenotypic feature data to generate standardized TCM syndrome labels for each time-series follow-up node; the knowledge base is constructed based on authoritative expert consensus. By combining the standardized TCM syndrome labels of each time-series follow-up node, a structural equation model of TCM and Western medicine joint representation is constructed for each time-series follow-up node. The observation indicators of TCM phenotypic feature data and Western medicine objective detection data are used as the observation indicators of TCM phenotypic dimension and Western medicine pathological dimension, respectively. By calculating the factor loading of each observation indicator, the mathematical correlation between parallel observation indicators is quantified, and the TCM and Western medicine integrated feature vector of each time-series follow-up node is obtained. Activate the full-cycle dynamic evolution network, extract the TCM-Western medicine integration feature vector of the target object at the current time-series follow-up node and historical time-series follow-up nodes, and combine the spatiotemporal covariate features and time intervals of all time-series follow-up nodes. As covariates, they are input into the pre-trained state transition model to calculate the potential syndrome state transition probability matrix under different time windows; Based on the probability matrix of potential syndrome state transition under different time windows and the integrated Chinese and Western medicine feature vector of the current time-series follow-up node, a risk prediction index representing the evolution trajectory of lung cancer throughout the entire cycle is generated. Based on the risk prediction index, the interactive acquisition terminal is controlled to execute the corresponding hierarchical early warning and intervention response interaction for the entire cycle stage.

[0008] According to the present invention, a medical and health data collaborative processing system based on the evolution of Chinese and Western medicine characteristics for the entire lung cancer cycle is provided. The Chinese medicine phenotypic characteristic data includes subjective symptom scale data and / or tongue and pulse sign data acquired through image sensors.

[0009] According to the present invention, a collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle is provided. The objective detection data of Western medicine includes any one of the following or any combination thereof: imaging biomarker data (such as nodule diameter, spiculation sign, lobulation sign, ground glass density component ratio), pathological data (pathological laboratory indicators (such as histological classification, immunohistochemical results)), laboratory test data (serum tumor markers, inflammatory factors), functional status data (including vital signs and physical performance status (ECOG) score), treatment intervention data (records of drugs and regimens for surgery, radiotherapy and chemotherapy, targeted and immunotherapy), and gene phenotype data (driver gene mutation status (EGFR / ALK, etc.)).

[0010] According to the present invention, a collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle is provided. The system, based on spatial location information, invokes an external environment database to construct spatiotemporal covariate features corresponding to time-series follow-up nodes, including: Based on spatial location information, the external environment database is accessed to obtain data on the regional environmental influencing factors of the target object at each time-series follow-up node; By combining spatial location information and regional environmental influencing factor data, spatiotemporal covariate characteristics of each time series follow-up node are constructed.

[0011] According to the present invention, a medical and health data collaborative processing system based on the evolution of Chinese and Western medicine characteristics for the entire lung cancer cycle is provided. The central control processor is further configured to execute logic: preprocessing Chinese medicine phenotypic characteristic data and / or Western medicine objective detection data, wherein the preprocessing includes standardization (e.g., Min-Max normalization (mapping the original value to [0,1]), Z-score standardization (for continuous Western medicine indicators), and inverse indicator processing (e.g., albumin: the higher the value, the better)).

[0012] In one implementation, the expression for processing the reverse indicator is:

[0013] In the formula, Indicates the first The standardized values ​​of each indicator (range [0,1], the larger the value, the lower the albumin (the worse the condition)). Indicates the first The clinically reasonable minimum value of each indicator. Indicates the first The original values ​​of each indicator Indicates the first The clinically reasonable maximum value of each indicator.

[0014] According to the present invention, a collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle is provided. The expression of the second-order confirmatory factor analysis structural equation model is as follows: First-order measurement model:

[0015]

[0016] Second-order structure model:

[0017]

[0018] In the formula, Indicates the first Standardized values ​​of a TCM phenotypic observation index Indicates the first One TCM indicator for the first-order latent variable "TCM phenotypic dimension" Factor loadings; Indicates the first Standardized values ​​of objective test indicators in Western medicine. Indicates the first A Western medicine indicator for the first-order latent variable "Western medicine pathology dimension" Factor loadings; This represents the second-order latent variable "the state of integration of traditional Chinese and Western medicine pathology"; The second-order factor loadings reflect the contribution strength of the TCM phenotypic dimension and the Western medicine pathological dimension to the fusion state. This represents the error term. The larger the absolute value of the factor loading, the stronger the correlation between the indicator and the corresponding latent variable.

[0019] According to the present invention, a collaborative processing system for medical and health data based on the evolution of traditional Chinese medicine (TCM) and Western medicine characteristics for the entire lung cancer cycle is provided. This system, by combining standardized TCM syndrome labels at each time-series follow-up node, constructs a structural equation model for the joint representation of TCM and Western medicine at each time-series follow-up node. The system uses the observation indicators of TCM phenotypic characteristic data and Western medicine objective detection data as observation indicators of corresponding first-order latent variables. By calculating the factor loadings of each parallel observation indicator, the system quantifies the mathematical correlation between parallel observation indicators to obtain the TCM-Western medicine integrated feature vector for each time-series follow-up node. The system further includes: The factor loadings are standardized, and the expression for the standardized factor loadings is as follows:

[0020] In the formula, Indicates the first Each explicit observation indicator affects latent variables. The standardized factor loadings (range [-1, 1], the closer the absolute value is to 1, the stronger the correlation). Indicates the first Each explicit observation indicator affects latent variables. Factor loadings, Representing latent variables standard deviation Indicates the first The standard deviation of the standardized values ​​of an explicit observation indicator (usually fixed at 1).

[0021] According to the present invention, a collaborative processing system for medical and health data based on the evolution of traditional Chinese medicine (TCM) and Western medicine characteristics for the entire lung cancer cycle is provided. This system combines standardized TCM syndrome labels from each time-series follow-up node to construct a structural equation model representing the combined TCM and Western medicine characteristics for each time-series follow-up node. The system uses observation indicators from TCM phenotypic feature data and Western medicine objective detection data as parallel observation indicators. By calculating the factor loadings of each parallel observation indicator, the system quantifies the mathematical correlation between the parallel observation indicators, thereby obtaining the TCM-Western medicine integrated feature vector for each time-series follow-up node, including: Based on TCM phenotypic characteristic data, factor score regression was used to obtain the TCM phenotypic dimension scores of the target subjects at each time-series follow-up node. ; Based on objective Western medicine test data, factor score regression was used to obtain the Western medicine pathology dimension scores of the target subjects at each time-series follow-up node. ; By combining the scores of TCM phenotypic dimension, Western medicine pathological dimension, and factor loadings, the TCM-Western medicine integration feature values ​​of each time-series follow-up node are calculated as the TCM-Western medicine integration feature vectors of each time-series follow-up node.

[0022] According to the present invention, a collaborative processing system for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics throughout the entire lung cancer cycle provides the following expression for calculating the score of the traditional Chinese medicine phenotypic dimension:

[0023] In the formula, Indicates the scores of the TCM phenotypic dimensions (range and Consistent, approximately 0-1). Indicates the first Standardized values ​​of TCM phenotypic feature data Indicates the first First-order factor loadings of TCM phenotypic feature data This represents the total number of phenotypic characteristics in Traditional Chinese Medicine. This represents a group of phenotypic characteristics in Traditional Chinese Medicine (TCM), including indicators such as tongue appearance, pulse appearance, and symptoms collected by the four diagnostic methods of TCM. The expression for calculating the score of the Western medicine pathology dimension is:

[0024] In the formula, Indicates the score of the Western medicine pathology dimension (range and Consistent, approximately 0-1). Indicates the first Standardized values ​​of objective test data from Western medicine. Indicates the first The first-order factor loading of objective test data from Western medicine. This represents the total number of Western medicine test indicators. This refers to the objective test data set of Western medicine, including indicator data obtained from Western medical equipment or testing, such as imaging, laboratory tests, and pathology. The first expression for calculating the characteristic value of the integration of traditional Chinese and Western medicine is:

[0025] To ensure In The system normalizes the interval as follows:

[0026] In the formula, This indicates the score of the TCM phenotypic dimension. This indicates the score in the Western medicine pathology dimension. This represents the second-order factor loading of the TCM phenotypic dimension score on the state of integration between TCM and Western medicine. This represents the second-order factor loading of the Western medicine pathology dimension score on the integration status of traditional Chinese and Western medicine.

[0027] In structural equation modeling, second-order factor loadings and with first-order factor loading These are parameters that are simultaneously estimated when fitting a second-order structural equation model: first-order factor loadings. The second-order factor loadings reflect the contribution strength of each observed indicator (such as tongue color and CEA) to its respective first-order latent variable (traditional Chinese medicine phenotypic dimension and Western medicine pathological dimension). , This reflects the contribution strength of the two first-order latent variables to the top-level second-order latent variable (the fusion status of traditional Chinese and Western medicine pathology). To obtain unique and comparable parameter estimates, a measurement scale needs to be set for the second-order latent variables during model estimation. This is achieved by fixing its variance to 1—equivalent to standardizing the fusion status scores of all patients to a distribution with a mean of 0 and a standard deviation of 1, thereby eliminating the arbitrariness of the latent variable scale. and This can be directly interpreted as the standardized effect size of the change in the fusion state of TCM and Western medicine pathology for every one standard deviation change in either the TCM phenotypic dimension or the Western medicine pathology dimension. Using this fixed-variance identification condition, structural equation modeling software (such as AMOS and Mplus) can simultaneously and iteratively solve all factor loadings using the training dataset, ultimately outputting the standardized... , The values ​​can be directly compared; the larger the value, the greater the contribution of that dimension to the final fusion state.

[0028] According to the present invention, a collaborative processing system for medical and health data based on the evolution of traditional Chinese medicine (TCM) and Western medicine characteristics for the entire lung cancer cycle is provided. This system combines standardized TCM syndrome labels from each time-series follow-up node to construct a structural equation model representing the combined TCM and Western medicine characteristics for each time-series follow-up node. The system uses observation indicators from TCM phenotypic feature data and Western medicine objective detection data as parallel observation indicators. By calculating the factor loadings of each parallel observation indicator, the system quantifies the mathematical correlation between the parallel observation indicators, thereby obtaining the TCM-Western medicine integrated feature vector for each time-series follow-up node, including: Based on TCM phenotypic data, Western medicine objective test data, and factor loadings, the TCM-Western medicine integration feature values ​​for each time-series follow-up node were calculated, serving as the TCM-Western medicine integration feature vectors for each time-series follow-up node. The second expression for calculating the TCM-Western medicine integration feature values ​​is as follows:

[0029] In the formula, This represents the characteristic value of the integration of traditional Chinese and Western medicine. Indicates the first Standardized values ​​of TCM phenotypic feature data Indicates the first First-order factor loadings of TCM phenotypic feature data This represents the total number of phenotypic characteristics in Traditional Chinese Medicine. This represents the phenotypic characteristics group in Traditional Chinese Medicine. Indicates the first Standardized values ​​of objective test data from Western medicine. Indicates the first The first-order factor loading of objective test data from Western medicine. This represents the total number of Western medicine test indicators. This indicates the objective testing data group of Western medicine. Indicates the first The final fusion weights of the TCM phenotypic feature data Indicates the first The final fusion weight of the objective test data from Western medicine.

[0030] According to the present invention, a collaborative processing system for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle is provided. The expression for calculating the probability matrix of syndrome state transition is as follows: Main model (including covariates and time intervals):

[0031] in, This includes the spatiotemporal covariate characteristics and the integration of traditional Chinese and Western medicine features of the current time-series follow-up nodes. , The time interval between adjacent follow-up nodes; The regression coefficients are obtained by using maximum likelihood estimation from the training data.

[0032] (Explanation of the auxiliary smoothing term) When the training data is sparse, the system can combine Laplace smoothing estimation based on observation frequency as a prior:

[0033] In the formula, Represents the slave state based on frequency Transition to state The probability, This indicates the state of TCM syndromes during training. Transition to state Number of observations Indicates the TCM syndrome state during training. Total number of occurrences Represents the Laplace smoothing parameter (in this invention, it is taken as...) ), This represents the total number of TCM syndrome states defined in the system.

[0034] According to the present invention, a collaborative processing system for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics for the entire life cycle of lung cancer is provided. The system generates a risk prediction index characterizing the entire life cycle evolution trajectory of lung cancer based on the probability matrix of syndrome state transitions under different time windows and the integrated traditional Chinese and Western medicine feature vector of the current time-series follow-up node. The system includes: Risk prediction index calculation formula (based on transition probability matrix) Function: Based on the current TCM syndrome status and a predefined "high-risk syndrome set," the system calculates the probability that a patient will transition to a high-risk state in the next follow-up period, and maps this probability to a risk prediction index of [0, 100]. This index directly quantifies the likelihood of a patient evolving into an adverse syndrome in the short term and is the core basis for the system to achieve "early warning."

[0035] Step 1: Define the high-risk syndrome set Based on the consensus of TCM clinical experts on lung cancer, a high-risk syndrome set was established. The high-risk syndrome set is dynamically defined according to different disease stages (including screening period, treatment period and follow-up period); Step 2: Obtain the current TCM syndrome status: in time The patient's current condition can be obtained through TCM syndrome identification. .

[0036] Step 3: Calculate the underlying risk based on transition probabilities: Extract the current state from the potential symptom state transition probability matrix. Transfer to high-risk groups The sum of probabilities of any state in the equation:

[0037] in, Only includes spatiotemporal covariate features of the current time-series follow-up node (excluding) This ensures that the risk of syndrome evolution is independent of the current fusion characteristics.

[0038] Step 4: Obtain the current integrated traditional Chinese and Western medicine feature value: Obtain the integrated traditional Chinese and Western medicine feature value of the current time series follow-up node. (already normalized to) (Interval) directly used as:

[0039] Step 5: Calculate the final risk prediction index This index reflects both the evolution trend of TCM syndromes (based on the probability of transition) and the current multimodal pathological fusion status (based on the fusion feature value), with a value range of [0, 100].

[0040] Using a weighted linear combination:

[0041] in: The risk of syndrome evolution defined in step 3 (independent of the current situation) ) The normalized characteristic value of the integration of traditional Chinese and Western medicine (equal to) ) Weights (default) );in For the preset weights, satisfy ,default It can also be determined by optimization based on training data.

[0042] : Output risk prediction index, range [0,100] when High-risk warnings are triggered at certain times; when At that time, it was considered a medium-risk period; when The risk level is currently low.

[0043] According to the present invention, a collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle is provided. The system, which controls the interactive data acquisition terminal to execute hierarchical early warning and intervention response interactions corresponding to each stage of the entire cycle based on a risk prediction index, includes: Risk levels are classified based on risk prediction indices, and the corresponding tiered early warning and intervention response interactions are executed by the interactive data collection terminal according to the classified risk levels throughout the entire life cycle.

[0044] In one implementation, the interactive data acquisition terminal can use a tablet computer as its core. The tablet computer features a touchscreen, a 5G / Wi-Fi 6 communication module, and a high-precision GPS / BeiDou positioning and timestamp unit, facilitating mobile use in wards, outpatient clinics, or patients' homes and accurately recording each time-series follow-up node. To collect TCM phenotypic characteristic data, the tablet can connect to an external hyperspectral camera (equipped with a ring light and chin rest) via USB-C or Bluetooth for standardized tongue and facial color imaging. It integrates a high signal-to-noise ratio MEMS microphone array for voice breath and cough sound acquisition and obtains pulse waveforms through a wristband-type array pressure sensor (mimicking the cun, guan, and chi positions). Consultation data is directly entered through a structured questionnaire on the touchscreen, and a voice input module can be added to assist elderly patients. The entire device uses a sterilizable medical-grade casing and has built-in encrypted storage to ensure data security and consistency during long-term follow-up.

[0045] In one implementation, the geospatial positioning and sensing component can employ a multi-mode fusion positioning scheme to support the synchronous acquisition of spatial location information in different outdoor and indoor scenarios. In outdoor scenarios, a GPS / BeiDou dual-mode positioning module (supporting real-time dynamic differential technology with meter-level or sub-meter-level accuracy) is used. In indoor scenarios, a Bluetooth beacon array or ultra-wideband (UWB) base station system is integrated, combined with the inertial measurement unit (IMU) built into the tablet for dead reckoning and point supplementation. The geospatial positioning and sensing component is strictly synchronized with the timestamp of the interactive acquisition terminal to ensure that the spatial location information of each time-series follow-up node is aligned with the data from traditional Chinese medicine and Western medicine on the same timeline. This component can be independently packaged as a small wearable beacon or embedded inside the interactive acquisition terminal, and the location data is transmitted to the central control processor in real time via serial port or Bluetooth.

[0046] In one implementation, the medical device communication interface can employ multi-protocol gateway hardware, integrating physical layer support for common medical data exchange protocols such as HL7, FHIR, DICOM, and Modbus. It includes an RJ45 Ethernet port, serial ports (RS232 / RS485), a USB host, and a Bluetooth / BLE module for connecting to hospital information systems (such as HIS, LIS, and PACS) and bedside laboratory equipment. This interface hardware can be designed as a standalone embedded gateway box (ARM architecture, running lightweight Linux, with a built-in SSL / TLS encryption chip). It automatically retrieves objective Western medicine test data (such as blood routine, CT image parameters, tumor markers, etc.) of the target object at the corresponding time-series follow-up node from the electronic medical record system or laboratory database via the hospital's internal network. Simultaneously, to support off-site follow-up scenarios, the gateway can support 4G / 5G backhaul and establish a VPN tunnel with the hospital's front-end server to ensure data transmission security and real-time performance.

[0047] In one implementation, the central control processor can utilize a high-performance edge computing server as its hardware platform, configured with a multi-core CPU (such as a high-performance SoC based on Intel Xeon or ARM architecture), large-capacity memory (≥32GB), and high-speed SSD storage. An optional GPU (such as an NVIDIA Tesla or Jetson series) can be added to accelerate structural equation model calculations and state transition model inference tasks. This processor can be deployed on a regional medical collaboration platform or a cloud server, receiving TCM phenotypic feature data from the interactive acquisition terminal, location information from the geospatial positioning sensing component, and objective Western medicine test data forwarded from the medical device communication interface via a network switch. At the hardware level, it can include an independent Hardware Security Module (HSM) for key management and data encryption, while also featuring redundant power supplies and RAID disk arrays to ensure the continuity and reliability of collaborative data processing throughout the entire lung cancer lifecycle. All model calculations (symptom rule matching, structural equation modeling, state transition probability calculation, and risk prediction) are completed within the processor, and the final generated warning and intervention commands are transmitted back to the interactive acquisition terminal for execution via Wi-Fi / 5G network.

[0048] This invention also provides a collaborative processing method for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics, applicable to the system described in any of the above claims, for the entire lung cancer cycle, comprising: At multiple time-series follow-up points throughout the entire lung cancer cycle, TCM phenotypic characteristics data, spatial location information, and Western medicine objective test data of the target subjects were acquired simultaneously. Based on spatial location information, the spatiotemporal covariate features of the corresponding time-series follow-up nodes are constructed by calling the external environment database. The TCM lung cancer syndrome diagnosis knowledge base was used to perform rule matching on TCM phenotypic feature data to generate standardized TCM syndrome labels for each time-series follow-up node; Structural equation modeling of integrated traditional Chinese and Western medicine representation was constructed for each time-series follow-up node. Based on the standardized TCM syndrome labels of each time-series follow-up node, a structural equation modeling of integrated traditional Chinese and Western medicine representation was constructed for each time-series follow-up node. The observation indicators of TCM phenotypic feature data and Western medicine objective detection data were used as parallel observation indicators. By calculating the factor loadings of each parallel observation indicator, the mathematical correlation between the parallel observation indicators was quantified, and the integrated traditional Chinese and Western medicine feature vector of each time-series follow-up node was obtained. Activate the full-cycle dynamic evolution network, extract the TCM-Western medicine integration feature vector of the target object at the current time-series follow-up node and historical time-series follow-up nodes, and combine the spatiotemporal covariate features and time intervals of all time-series follow-up nodes. As covariates, they are input into the pre-trained state transition model to calculate the potential syndrome state transition probability matrix under different time windows; Based on the probability matrix of potential syndrome state transitions under different time windows, a risk prediction index representing the entire evolution trajectory of lung cancer is generated. Control commands are then generated based on the risk prediction index and sent to the front-end terminal to trigger hierarchical early warning signals and interface rendering interactions for preoperative screening or postoperative monitoring stages.

[0049] The present invention also provides an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the computer program to implement any of the above-described methods for collaborative processing of medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle.

[0050] The present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements any of the above-described methods for collaborative processing of medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle.

[0051] The present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute any of the above-described methods for collaborative processing of medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine for the entire lung cancer cycle.

[0052] The present invention provides a collaborative processing system and method for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine throughout the entire lung cancer cycle, which can bring at least the following beneficial effects: This invention synchronously acquires TCM phenotypic data and Western medicine objective test data at multiple time-series follow-up nodes throughout the entire lung cancer cycle via an interactive acquisition terminal. Using structural equation modeling, it measures the latent variable of "the integration state of TCM and Western medicine pathology" using observation indicators from both TCM phenotypic data and Western medicine objective test data as parallel observation indicators. By calculating the factor loadings of each observation indicator, the correlation weights between parallel observation indicators are quantified. This mechanism allows for the precise numerical expression of the contribution weight of Western medicine micro-indications to TCM macro-syndromes, establishing a clear mathematical correlation between "disease" and "syndrome." This provides a unified and objective quantitative basis for diagnosis for physicians at different levels and with varying experience, thereby promoting the standardization of TCM syndrome diagnosis.

[0053] This invention activates a full-cycle dynamic evolution network, extracts integrated traditional Chinese and Western medicine feature vectors from multiple time-series nodes, and inputs spatiotemporal covariate features combining spatial location information and external environmental data as key constraints into a pre-trained state transition model. The system can quantify and analyze the probability of transitions between different syndrome states, such as "Qi deficiency" to "Yin deficiency" or "phlegm and blood stasis," by incorporating the climate and environmental factors of the patient's location, generating a risk prediction index characterizing the full-cycle evolution trajectory of lung cancer. This invention enables a paradigm shift from static symptom assessment to dynamic evolution pattern characterization, accurately predicting the patient's syndrome progression trend and providing a crucial time window for early clinical intervention.

[0054] This invention provides a complete closed-loop system for collaborative data processing. First, through an interactive data acquisition terminal and a communication interface with medical devices, multimodal data, including subjective symptom scales, tongue and pulse examination findings, and Western medical laboratory tests, are uniformly aggregated. Second, based on a "symptom-symptom cluster-syndrome" framework and combined with integrated traditional Chinese and Western medicine feature vectors, a systematic coupling and quantitative representation of the patient's recovery status is achieved. Finally, based on a risk prediction index generated by a dynamic evolution model, the system can automatically control the interactive data acquisition terminal to execute tiered early warning and intervention response interactions. This directly empowers the outpatient management of lung cancer post-operative patients, implementing precise risk stratification management based on individualized evolution trajectories. This not only improves patients' quality of life and recovery outcomes but also provides a reliable technical platform for conducting large-scale, standardized integrated traditional Chinese and Western medicine clinical research and accumulating real-world data. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0056] Figure 1This invention provides a flowchart illustrating a collaborative processing method for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine throughout the entire lung cancer cycle.

[0057] Figure 2 This invention provides a schematic diagram of a collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine, which is designed for the entire life cycle of lung cancer. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, embodiments of this invention, and should not be construed as limiting the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0059] Figure 1 This invention provides a flowchart illustrating a collaborative processing method for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine throughout the entire lung cancer cycle. Figure 2 This invention provides a schematic diagram of a collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine, targeting the entire life cycle of lung cancer. The executing entity of this collaborative processing system can be any applicable terminal-side device or network-side device, such as a health data collaborative processing device.

[0060] The present invention will be illustrated by several specific embodiments provided below.

[0061] Example 1: Patient's initial consultation and record creation after entering the system for the first time This embodiment provides a collaborative processing method for medical and health data based on the evolution of TCM and Western medicine characteristics throughout the entire lung cancer cycle. It is used for patient identity registration, intelligent consultation, TCM phenotypic feature data collection, Western medicine objective test data access, syndrome identification, joint modeling, and risk reasoning when the patient first enters the system.

[0062] In this embodiment, the collaborative processing system for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics throughout the entire lung cancer cycle includes an interactive data acquisition terminal, a geospatial positioning and sensing component, a medical device communication interface, a central control processor, and a medical workstation. The interactive data acquisition terminal serves as the interface for consultation, data collection, early warning, and follow-up visit reminders; the geospatial positioning and sensing component acquires the patient's spatial location information; the medical device communication interface connects to the hospital information system; the central control processor is configured to receive traditional Chinese medicine phenotypic feature data acquired by the interactive data acquisition terminal, spatial location information acquired by the geospatial positioning and sensing component, and objective Western medicine test data acquired by the medical device communication interface, and performs multi-source data fusion, feature modeling, and risk inference; the medical workstation receives the analysis results and performs manual intervention.

[0063] When a patient first enters the system due to symptoms such as cough, chest tightness, shortness of breath, hemoptysis, fatigue, or abnormal lung imaging during a physical examination, they can initiate an access request through a hospital outpatient terminal, mobile terminal, mini-program, or in-hospital self-service device. The system first performs identity verification and file creation, guiding the patient to sign an electronic informed consent form and enter their identity information, contact information, age, gender, occupation, smoking history, family history, and past medical history. Simultaneously, based on authorization, the system automatically connects to the hospital information system to retrieve past examination records, generating an initial patient file and a unique patient identifier. After file creation, the system uses an interactive data collection terminal to collect structured data according to a preset consultation strategy, focusing on lung cancer-related high-risk factors, typical symptoms, accompanying symptoms, and lifestyle. The consultation content includes at least the nature of the cough, sputum production, sputum color, sputum volume, whether there is blood in the sputum, the degree of chest pain, chest tightness, shortness of breath, exercise tolerance, degree of fatigue, fever, night sweats, dry mouth and thirst, loss of appetite, weight changes, sleep quality, bowel and bladder function, and emotional state. During the consultation process, the system can collect information through various methods such as natural language interaction, button selection, scale scoring, and voice input, and perform semantic analysis on the natural language input by the patient.

[0064] After completing the first round of consultation, the interactive data acquisition terminal guides the patient to collect TCM phenotypic features, including tongue image acquisition, facial image acquisition, and pulse signal acquisition. The collected tongue and pulse data are then preprocessed and converted into standardized feature data that can be used for syndrome identification.

[0065] Subsequently, the medical device communication interface automatically retrieves relevant objective Western medicine test data from the hospital information system, laboratory system, imaging system, pathology system, and follow-up system. The objective Western medicine test data includes chest CT or PET-CT imaging features, lung function results, blood routine, liver and kidney function, inflammatory markers, blood gas analysis, serum tumor markers, pathological diagnosis results, driver gene detection results, and previous treatment records.

[0066] The central control processor standardizes the collected TCM phenotypic feature data and uses the TCM lung cancer syndrome diagnosis knowledge base for rule-based reasoning to map the patient's current state to standardized TCM syndrome labels, such as phlegm-heat obstructing the lungs syndrome, qi and yin deficiency syndrome, lung yin deficiency syndrome, or phlegm-blood stasis syndrome. Subsequently, the system constructs a structural equation model. Further, the central control processor combines geospatial positioning information to construct a spatiotemporal covariate feature set, which is then input into a full-cycle dynamic evolution network and a pre-trained state transition model to generate a risk prediction index.

[0067] The system can output high-risk alerts to patient terminals and simultaneously push them to medical workstations; the system outputs follow-up visit reminders, lifestyle intervention suggestions, and traditional Chinese medicine conditioning suggestions; the system maintains a regular follow-up schedule and automatically triggers follow-up tasks in the next cycle. Through the above process, the system realizes the establishment of files, consultation, data collection, analysis, and risk output for patients after their first access to the system.

[0068] Example 2: Patients enter the long-term follow-up phase after treatment This embodiment provides a collaborative health data processing workflow suitable for lung cancer patients entering the long-term follow-up phase after treatment. This workflow is used to continuously collect traditional Chinese and Western medicine information, analyze status evolution, predict recurrence risk, and manage follow-up visits after patients complete surgery, radiotherapy, chemotherapy, targeted therapy, or immunotherapy.

[0069] Once a patient completes their primary treatment and enters the rehabilitation observation period, the system switches the patient from treatment management status to long-term follow-up status based on the treatment completion indicator or follow-up instructions issued by the doctor. The central control processor automatically adjusts the follow-up template, data collection frequency, and risk assessment rules according to this status, and assigns a follow-up period to the patient based on their treatment method, pathological stage, genetic status, past complications, and recurrence risk level.

[0070] Once the follow-up phase begins, the central control processor automatically sends follow-up task reminders to the interactive data acquisition terminal according to a preset time cycle. Patients then follow the terminal's guidance to complete the entry of follow-up content for this cycle. This content includes changes in cough, sputum volume, chest tightness, shortness of breath, fatigue level, appetite, weight, sleep quality, night sweats, hemoptysis, fever, constipation or diarrhea, emotional state, and daily activity tolerance.

[0071] During periodic follow-ups, patients repeatedly collect images of their tongue, face, and pulse using an interactive data acquisition terminal. The central control processor standardizes the collected tongue images and pulse signals, extracting features such as tongue color, tongue coating, tongue body, pulse rate, pulse shape, and pulse strength. These features, along with the patient's current chief complaint, form a TCM phenotypic feature vector at each follow-up time point. If the patient is in a stable phase, it may be identified as lung and spleen qi deficiency syndrome or lung yin deficiency syndrome. If the patient presents with dry mouth, dry throat, red tongue with little coating, and a thready and rapid pulse, it may indicate an increasing trend of yin fluid depletion. If the patient exhibits increased sputum, worsening chest tightness, greasy tongue coating, and a slippery pulse, it may indicate an increasing trend of phlegm-dampness accumulation or phlegm-heat obstruction.

[0072] After the patient completes periodic follow-up examinations, the medical device communication interface obtains corresponding objective Western medicine test data from the hospital information system and related examination systems. The objective Western medicine test data includes chest imaging results, tumor markers, blood routine, liver and kidney function, inflammatory indicators, oxygenation status, lung function results, and records of post-treatment complications.

[0073] Subsequently, the central control processor inputs the integrated TCM and Western medicine feature vectors from multiple follow-up nodes into the full-cycle dynamic evolution network. It then constructs a spatiotemporal covariate feature set by combining the patient's historical treatment path, current spatial location, and external environmental information. This set is then input into a pre-trained state transition model for risk inference, calculating the conditional probability of the patient transitioning from the current follow-up state to the next risk state. Finally, it constructs a syndrome state transition probability matrix to identify whether the patient is trending toward a state with high recurrence risk, high metastasis risk, or complication risk.

[0074] Based on the risk prediction index and state transition results, the central control processor automatically generates follow-up responses at different levels. When the risk prediction index is low, the system maintains the routine follow-up plan; when the risk prediction index is moderate, the system pushes lifestyle adjustment suggestions, traditional Chinese medicine conditioning suggestions, and observation prompts to the patient; when the risk prediction index is high, the system outputs a follow-up reminder to the patient's terminal and simultaneously pushes it to the medical staff workstation, where medical staff decide whether to expedite the outpatient follow-up, conduct supplementary examinations, or adjust the subsequent management plan based on the patient's condition. Through this process, the system forms a closed-loop management system encompassing follow-up reminders, symptom collection, syndrome analysis, data fusion, risk prediction, tiered early warning, follow-up triggering, and result feedback.

[0075] Example 3: Patient enters the perioperative management stage This embodiment provides a collaborative health data processing workflow suitable for perioperative management of lung cancer patients. This workflow is used for preoperative patient assessment and postoperative recovery monitoring.

[0076] When a patient is scheduled for surgery, the system switches them to perioperative management mode. The central control processor, based on the perioperative management template, prioritizes collecting information on preoperative symptoms, basic vital signs, tongue and pulse status, and functional status. It also synchronously retrieves preoperative chest CT imaging features, lung function, coagulation function, complete blood count, serum tumor markers, pathological classification, and driver gene detection results from the hospital information system.

[0077] During the postoperative recovery phase, the system continues to acquire information on body temperature, blood oxygen saturation, pain score, inflammatory markers, drainage status, and postoperative complications. At the same time, the interactive acquisition terminal collects data on TCM phenotypic characteristics such as dyspnea, fatigue, appetite, sleep, and changes in tongue and pulse.

[0078] The system further combines spatiotemporal covariates and pre-trained state transition models to calculate the probability of postoperative recovery transitioning to infection, effusion, qi deficiency and blood stasis, or delayed recovery. The central control processor sends alarms to the patient terminal and medical staff workstation, prompting medical staff to strengthen monitoring, adjust nursing plans, or take early action.

[0079] Example 4: Lightweight follow-up visits in home and grassroots settings outside hospitals This embodiment provides a lightweight data collection and collaborative processing workflow suitable for outpatient home follow-up and primary screening scenarios.

[0080] In this embodiment, the interactive data acquisition terminal is deployed on the patient's mobile phone or tablet and communicates with the wearable device. The wearable device includes a wrist-type vital signs monitoring device, a finger-clip blood oxygen acquisition device, a portable heart rate monitoring module, a sleep monitoring module, and an activity monitoring module. An image sensor can be integrated into the front-facing camera of the mobile terminal or an external tongue image acquisition accessory to acquire tongue and facial images; pulse data can be acquired by the pulse wave acquisition module in the wearable device or an external pulse diagnosis device.

[0081] In scenarios where patients are at home or outside the hospital, the system sends data collection task reminders to patients according to a preset cycle. Guided by the interactive data collection terminal, patients complete symptom scale completion, tongue image capture, wearable physiological parameter collection, pulse image collection, location information acquisition, and synchronization of recent follow-up medical data. The central control processor performs time synchronization and format standardization processing on data from different sources, and then performs dimensionality reduction, rule matching, and syndrome label generation on TCM phenotypic features.

[0082] Since some data in this embodiment comes from wearable devices, and the data is characterized by high frequency, continuity, and low workload, the central control processor can use a sliding time window method to extract phased statistical features, including average heart rate, lowest nocturnal blood oxygen saturation, duration of continuous hypoxia, and degree of activity reduction. When the system identifies that a patient has recently experienced persistent nocturnal cough, decreased blood oxygen saturation, reduced activity, accompanied by a red tongue with little coating and a thready and rapid pulse, it can determine that the patient's TCM syndrome is shifting towards Qi and Yin deficiency. When an increase in inflammatory markers or changes in lesions are detected in follow-up imaging, the system increases the recurrence risk index and outputs warning information and follow-up consultation suggestions to the terminal.

[0083] Through this embodiment, the system can achieve continuous off-site monitoring, remote data collection and risk identification without relying on a large number of specialized equipment, and is suitable for patient home management, primary care screening and pre-visit early warning.

[0084] Example 5: A Distributed Implementation of Collaborative Management in Multi-Center Hospitals This embodiment provides a distributed data access and collaborative processing workflow suitable for multi-center hospital collaborative management, so as to realize unified access, collaborative modeling and risk sharing of lung cancer full-cycle data among different medical institutions.

[0085] In this implementation, the central control processor can be deployed on a regional medical collaboration platform or a cloud server, and the communication interfaces of medical devices from multiple medical institutions are connected to the information systems of their respective hospitals. These institutions include tertiary hospitals, oncology hospitals, primary healthcare institutions, and rehabilitation management institutions. Interactive data acquisition terminals in each institution are used to collect TCM phenotypic characteristic signals and follow-up information of patients at different diagnostic and treatment stages; geospatial positioning sensing components are used to record the spatial location information of patients in different hospitals, departments, or residential areas; and medical device communication interfaces are used to obtain medical data at corresponding stages from the institutions' HIS, LIS, PACS, pathology systems, radiotherapy systems, and follow-up systems.

[0086] During patient referrals, follow-up examinations, or rehabilitation management across institutions, the system uses a unified identity identifier to link patient data, ensuring that data collected by different medical institutions can be mapped to the same target object. For example, if a patient completes preoperative assessment at Hospital A, undergoes surgery at Hospital B, and receives rehabilitation follow-up at Hospital C, the system can access the examination results, pathology results, treatment records, and follow-up records generated by the three hospitals at different stages and integrate them according to time sequence.

[0087] Upon receiving data from multiple institutions, the central control processor first performs field standardization, unit normalization, and time-series alignment on the data from different sources to eliminate differences in testing equipment, recording formats, and indicator standards among different hospitals. Subsequently, the system standardizes the TCM phenotypic feature signals collected from various institutions and performs consistency correction on the objective testing data from Western medicine. In multi-center application scenarios, the system can also improve the model's generalization ability and stability under cross-institutional conditions by considering the data quality, collection completeness, and sample stability of different institutions. The central control processor generates a patient risk prediction index based on the integrated TCM and Western medicine feature vector from multiple centers, combined with spatiotemporal covariate feature sets and a pre-trained state transition model.

[0088] When a patient exhibits a high-risk trend at any follow-up node of an access institution, the system can simultaneously send early warning information to the medical and nursing workstations of the patient's current institution and the previous management institution. When a patient needs to be transferred to a higher-level hospital for further treatment, the system can automatically generate referral assistance prompts and data summary information so that each institution can quickly understand the patient's condition evolution. For patients with long-term follow-up, the system can maintain continuous records of syndrome evolution and risk trajectory across different institutions, avoiding data interruptions caused by hospital transfers, hospital changes, or cross-regional medical treatment.

[0089] Through this embodiment, the system can achieve unified access and collaborative processing of cross-institutional, multi-source, and heterogeneous medical data, enhancing the continuity, integrity, and traceability of lung cancer full-cycle management.

[0090] Example 6: I. Patient Basic Information and Enrollment Process Patient Li, male, 58 years old, entered the system for the first time due to "coughing and blood-tinged sputum for 2 weeks". The system performed the following steps: Step 1: Informed Consent and Filing Mr. Li signed an electronic informed consent form through the hospital terminal, and his personal information, smoking history (30 years, 20 cigarettes / day), family history (father's lung cancer), and past medical history (hypertension for 5 years) were entered. The system connects to the HIS to obtain outpatient medical records and preliminary examination results, generating a unique patient identifier.

[0091] Step 2: Intelligent Consultation Interactive data acquisition terminals (such as a four-diagnosis robot) collect symptoms (Likert 5: 0-4): cough 3 / 4, yellow sputum 2 / 4, chest tightness 2 / 4, shortness of breath 2 / 4, fatigue 2 / 4, poor sleep 3 / 4, decreased appetite 2 / 4. Symptoms are mapped to standardized codes.

[0092] Step 3: Traditional Chinese Medicine Phenotypic Data Collection Mr. Li used an interactive data acquisition terminal to collect images of his tongue (dark red tongue color, yellow and greasy tongue coating), facial features, and pulse (slippery and rapid). The central processing unit extracted the following features: dark red tongue color (example normalized code 0.72), yellow and greasy tongue coating (example normalized code 0.85), and slippery and rapid pulse (example normalized code 0.78). In practical applications, the normalized codes for each feature need to be obtained based on the training dataset through factor analysis or optimal scaling calibration (0-1 scale). Step 4: Western Medicine Data Integration Medical device communication interface retrieved hospital data: CT showed a 3.8 cm nodule in the upper lobe of the right lung (lobulated, spiculated, with visceral pleural invasion), blood tests: CEA 15.3 ng / mL, NLR 4.8, albumin 32.1 g / L, clinical stage IIB (cT2N1M0).

[0093] Step 5: Identification of Traditional Chinese Medicine Syndromes The central control processor calls upon the TCM lung cancer syndrome diagnosis knowledge base, performs rule-based reasoning based on structured diagnostic methods and symptom data, and outputs the auxiliary diagnosis result of "phlegm-heat obstructing the lungs syndrome". After being reviewed and confirmed by the physician, it is included in the patient's file.

[0094] Step 6: Structural Equation Model Construction and Feature Vector Fusion Generation The central control processor constructs a second-order structural equation model, using the symptoms of step 2 (cough, sputum color, chest tightness) and the TCM phenotypic characteristics of step 3 (tongue color, tongue coating, pulse) as the first set of observation indicators, and the objective Western medicine test data of step 4 (tumor diameter, CEA, NLR, albumin, stage) as the second set of observation indicators, to jointly measure the latent variable "fusion status of TCM and Western medicine pathology". The calculated fusion characteristic value is fusion=0.71 (0-1 scale).

[0095] Step 7: Risk Prediction and Initial Intervention The central control processor inputs the fusion and spatiotemporal covariates (winter, air quality index 125) into the full-cycle dynamic evolution network, outputting a risk prediction index of 71 (range 0-100). This is classified as medium risk (thresholds: <50 low risk, 50-84 medium risk, ≥85 high risk). The interactive data acquisition terminal pushes the following auxiliary suggestions to the medical workstation, which are implemented after confirmation by the attending physician: Lifestyle intervention: mandatory smoking cessation guidance, a light diet, and moderate aerobic exercise; Traditional Chinese medicine conditioning suggestions: a herbal tea for clearing heat and resolving phlegm, promoting lung function and relieving cough (Trichosanthes kirilowii 10g, Fritillaria thunbergii 6g, Houttuynia cordata 10g); Outdoor activities: reduce outdoor activities when AQI>150, and wear an N95 mask if necessary; avoid walking near main roads during peak traffic hours (7-9 am and 5-7 pm); Residential reminder: if the patient lives in an industrial area or near a main road, it is recommended to increase indoor greenery (such as pothos and snake plant) and regularly clean with wet cleaning. Daily self-test: Measure resting heart rate and blood oxygen saturation (SpO2) upon waking, record cough frequency and sputum color changes, and submit to the system weekly; Follow-up reminder: Reminds patients that close follow-up is required after surgery according to guidelines, and a specific plan will be formulated after the pathological and genetic testing results are clear.

[0096] Step 8: Long-term follow-up process after treatment Mr. Li underwent thoracoscopic right upper lobectomy and lymph node dissection. Postoperative pathology showed 1 / 6 positive lymph nodes and EGFR exon 19 deletion mutation. He received 4 cycles of adjuvant chemotherapy (pemetrexed + cisplatin) postoperatively, followed by adjuvant osimertinib 80 mg once daily (planned for 3 years). Based on the NCCN Non-Small Cell Lung Cancer Guidelines (v5.2026), patient pathological stage, genetic status, and initial risk stratification, the system recommended a follow-up plan every 3 months to the attending physician. After physician confirmation, a personalized follow-up plan was generated. The system operation and clinical decisions at each follow-up node are shown in the table below.

[0097]

[0098] Step 9: Dynamic Modeling and Early Warning At node V3 (9 months post-surgery), the central control processor identified a multi-dimensional abnormal trend: "newly developed small nodules + progressively elevated tumor markers + worsening inflammatory indicators + transformation of TCM syndrome to phlegm-stasis syndrome." Through a full-cycle dynamic evolution network, it issued a high-risk warning signal two months in advance. Based on the TCM syndrome evolution knowledge base, the interactive data acquisition terminal indicated a high risk of the patient evolving from "phlegm-heat obstructing the lungs" to "blood stasis and toxin accumulation," consistent with the subsequent pathological confirmation of recurrence. Upon receiving the warning, the medical team initiated the examination process ahead of schedule, confirming recurrence before the patient experienced significant worsening of clinical symptoms and adjusting the treatment plan accordingly.

[0099] This invention provides a collaborative processing system and method for medical and health data based on the evolution of TCM and Western medicine characteristics throughout the entire lung cancer cycle. Through an interactive acquisition terminal, it simultaneously acquires TCM phenotypic characteristic data and Western medicine objective detection data at multiple time-series follow-up nodes throughout the lung cancer cycle. Using structural equation modeling, it uses observation indicators from both TCM phenotypic characteristic data and Western medicine objective detection data as parallel observation indicators to jointly measure the latent variable of "the integration state of TCM and Western medicine pathology." By calculating factor loadings, it quantifies the correlation weights between parallel observation indicators (TCM four diagnostic methods and Western medicine micro-indicators), thereby establishing a clear "disease-syndrome" mathematical correlation. This provides unified and objective quantitative diagnostic evidence for physicians at different levels, promoting the standardization of TCM syndrome diagnosis. Simultaneously, it activates the dynamics throughout the entire cycle... The evolution network incorporates spatiotemporal covariate features combining spatial location information and external environmental data as constraints input to a pre-trained state transition model. It quantifies the transition probabilities between different syndrome states and generates a risk prediction index, achieving a paradigm shift from static assessment to dynamic evolutionary pattern characterization, providing a critical time window for early clinical intervention. Furthermore, this invention constructs a complete data collaborative processing closed loop that unifies the aggregation of multimodal data, achieves quantitative state representation based on a "symptom-symptom cluster-syndrome" framework, and automatically triggers stratified early warning and intervention responses based on the risk prediction index. This directly empowers precise risk stratification in the outpatient management of lung cancer post-operative patients, effectively improving patients' quality of life and rehabilitation outcomes, and providing a reliable technical platform for large-scale integrated traditional Chinese and Western medicine clinical research.

[0100] The present invention provides an electronic device that may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The processor can invoke logical instructions in the memory to execute the steps of the aforementioned collaborative processing method for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics throughout the entire lung cancer cycle.

[0101] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to perform the steps of the medical and health data collaborative processing method based on the evolution of Chinese and Western medicine characteristics for the entire lung cancer cycle as described above.

[0103] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described collaborative processing method for medical and health data based on the evolution of Chinese and Western medicine characteristics throughout the entire lung cancer cycle.

[0104] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0105] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A collaborative processing system for medical and health data based on the evolution of characteristics of traditional Chinese and Western medicine, addressing the entire life cycle of lung cancer, characterized in that: It includes an interactive data acquisition terminal, a geospatial positioning and sensing component, a medical device communication interface, and a central control processor, among which, The interactive acquisition terminal is configured to collect TCM phenotypic feature data of the target object at multiple time-series follow-up nodes throughout the entire lung cancer cycle; The geospatial positioning sensing component is configured to synchronously collect spatial location information of the target object at each time-series follow-up node; The medical device communication interface is configured to obtain objective Western medicine test data of the target object at the corresponding time-series follow-up node from an external medical information system; The central control processor is configured to receive TCM phenotypic feature data collected by the interactive acquisition terminal, spatial location information collected by the geospatial positioning sensing component, and Western medicine objective test data obtained by the medical device communication interface, and execute the following logic: Based on spatial location information, the spatiotemporal covariate features of the corresponding time-series follow-up nodes are constructed by calling the external environment database. The TCM lung cancer syndrome diagnosis knowledge base was used to perform rule matching on TCM phenotypic feature data to generate standardized TCM syndrome labels for each time-series follow-up node; By combining the standardized TCM syndrome labels of each time-series follow-up node, a structural equation model of TCM and Western medicine integration is constructed for each time-series follow-up node. The observation indicators of TCM phenotypic feature data and Western medicine objective detection data are used as parallel observation indicators. By calculating the factor loading of each parallel observation indicator, the mathematical correlation between the parallel observation indicators is quantified, and the TCM and Western medicine integration feature vector of each time-series follow-up node is obtained. Activate the full-cycle dynamic evolution network, extract the integrated Chinese and Western medicine feature vectors of the target object at the current time-series follow-up node and the historical time-series follow-up node, and input the spatiotemporal covariate features of all time-series follow-up nodes as constraints into the pre-trained state transition model to calculate the syndrome state transition probability matrix under different time windows. Based on the probability matrix of syndrome state transition under different time windows and the integrated Chinese and Western medicine feature vector of the current time-series follow-up node, a risk prediction index representing the evolution trajectory of lung cancer throughout the entire cycle is generated. Based on the risk prediction index, the interactive acquisition terminal is controlled to execute the corresponding hierarchical early warning and intervention response interaction for the entire cycle stage.

2. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 1, is characterized in that... Traditional Chinese medicine phenotypic data include subjective symptom scale data and / or tongue and pulse sign data acquired via image sensors; and / or, Western medicine objective test data include any one or any combination of the following: imaging biomarker data, pathological data, laboratory test data, functional status data, treatment intervention data, and gene phenotypic data.

3. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 1, is characterized in that... The process of constructing spatiotemporal covariate features for corresponding time-series follow-up nodes by calling an external environment database based on spatial location information includes: Based on spatial location information, the external environment database is accessed to obtain data on the regional environmental influencing factors of the target object at each time-series follow-up node; By combining spatial location information and regional environmental influencing factor data, spatiotemporal covariate characteristics of each time series follow-up node are constructed.

4. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 3, is characterized in that... The structural equation model is a second-order confirmatory factor analysis model, and its expression includes: First-order measurement model: Second-order structure model: In the formula, This represents the standardized values ​​of the TCM phenotypic observation indicators. For TCM indicators, the first-order latent variable "TCM phenotypic dimension" is used. Factor loadings; This represents the standardized value of an objective test indicator in Western medicine. For Western medicine indicators, the first-order latent variable "Western medicine pathology dimension" is used. Factor loadings; The second-order latent variable is "the state of integration of traditional Chinese and Western medicine pathology"; Second-order factor loading; This is the error term.

5. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 4, is characterized in that... The method combines standardized TCM syndrome labels from each time-series follow-up node to construct a structural equation model for the integrated TCM and Western medicine representation of each time-series follow-up node. Observational indicators of TCM phenotypic feature data and Western medicine objective detection data are used as parallel observation indicators. By calculating the factor loadings of each parallel observation indicator, the mathematical correlation between the parallel observation indicators is quantified, resulting in the TCM-Western medicine integrated feature vector for each time-series follow-up node, including: Based on TCM phenotypic characteristic data, factor score regression was used to obtain the TCM phenotypic dimension scores of the target subjects at each time-series follow-up node. ; Based on objective Western medicine test data, factor score regression was used to obtain the Western medicine pathology dimension scores of the target subjects at each time-series follow-up node. ; By combining the scores of TCM phenotypic dimension, Western medicine pathological dimension, and factor loadings, the TCM-Western medicine integration feature values ​​of each time-series follow-up node are calculated as the TCM-Western medicine integration feature vectors of each time-series follow-up node.

6. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 5, is characterized in that... The expression for calculating the scores of the TCM phenotypic dimensions is: In the formula, This indicates the score of the TCM phenotypic dimension. Indicates the first Standardized values ​​of TCM phenotypic feature data Indicates the first First-order factor loadings of TCM phenotypic feature data This represents the total number of phenotypic characteristics in Traditional Chinese Medicine. This represents the phenotypic characteristics group in Traditional Chinese Medicine; The expression for calculating the score of the Western medicine pathology dimension is: In the formula, This indicates the score in the Western medicine pathology dimension. Indicates the first Standardized values ​​of objective test data from Western medicine. Indicates the first The first-order factor loading of objective test data from Western medicine. This represents the total number of Western medicine test indicators. This represents the objective testing data set from Western medicine. The first expression for calculating the characteristic value of the integration of traditional Chinese and Western medicine is: To ensure In The system normalizes the interval as follows: In the formula, This represents the characteristic value of the integration of traditional Chinese and Western medicine. This indicates the score of the TCM phenotypic dimension. This indicates the score in the Western medicine pathology dimension. This represents the second-order factor loading of the TCM phenotypic dimension score on the state of integration between TCM and Western medicine. This represents the second-order factor loading of the Western medicine pathology dimension score on the integration status of traditional Chinese and Western medicine.

7. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 6, is characterized in that... The pre-trained state transition model uses a logistic regression form that includes covariates and time intervals. The main model expression for calculating the syndrome state transition probability matrix is ​​as follows: in, Includes the spatiotemporal covariate features of the current time-series follow-up node, but does not include the integrated traditional Chinese and Western medicine feature values. ; The time interval between adjacent follow-up nodes; These are regression coefficients, estimated from the training data; As an auxiliary smoothing term, the system also employs frequency-based Laplace smoothing estimation: In the formula, This represents the prior transition probability based on the frequency of the training set. Indicates the state in the training set Transition to state Number of observations Representing state Total number of occurrences Represents the Laplace smoothing parameter. This represents the total number of TCM syndrome state types defined in the system; this prior is used for regularization of the main model parameter estimation and is not used as the final output alone.

8. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in claim 7, is characterized in that... The risk prediction index characterizing the entire lifecycle evolution trajectory of lung cancer is generated based on the potential syndrome state transition probability matrix under different time windows and the integrated traditional Chinese and Western medicine feature vector of the current time-series follow-up node, including: Based on the potential syndrome state transition probability matrix under different time windows and the integrated traditional Chinese and Western medicine feature vector of the current time-series follow-up node, the risk prediction index is calculated using the risk prediction index calculation formula, where the risk prediction index calculation formula is: In the formula, Indicates starting from the current state A step to move to a high-risk syndrome set The high-risk syndrome set is the sum of the probabilities of any state. Based on the dynamic definition of the entire lung cancer cycle stages; ,and Normalized to the [0,1] interval; Indicates the preset weight, satisfying Furthermore, it can be optimized and determined based on training data; This represents the risk prediction index, ranging from [0, 100].

9. The medical and health data collaborative processing system based on the evolution of traditional Chinese and Western medicine characteristics for the entire lung cancer cycle, as described in any one of claims 1-8, is characterized in that... The step of controlling the interactive data acquisition terminal based on the risk prediction index to execute hierarchical early warning and intervention response interactions corresponding to the full life cycle stage includes: The system classifies risk levels based on the risk prediction index and controls the interactive data collection terminal to execute tiered early warning and intervention response interactions for the corresponding full-cycle stage according to the risk level classification. At the same time, the system feeds back the data collected during subsequent follow-ups to the central control processor for incremental updates of the factor loadings of the structural equation model, the parameters of the pre-trained state transition model, and the risk prediction weights, forming a full-cycle closed-loop dynamic optimization.

10. A collaborative processing method for medical and health data based on the evolution of traditional Chinese and Western medicine characteristics, applied to the system described in any one of claims 1-9, characterized in that, include: At multiple time-series follow-up points throughout the entire lung cancer cycle, TCM phenotypic characteristics data, spatial location information, and Western medicine objective test data of the target subjects were acquired simultaneously. Based on spatial location information, the spatiotemporal covariate features of the corresponding time-series follow-up nodes are constructed by calling the external environment database. The TCM lung cancer syndrome diagnosis knowledge base was used to perform rule matching on TCM phenotypic feature data to generate standardized TCM syndrome labels for each time-series follow-up node; By combining the standardized TCM syndrome labels of each time-series follow-up node, a structural equation model of TCM and Western medicine integration is constructed for each time-series follow-up node. The observation indicators of TCM phenotypic feature data and Western medicine objective detection data are used as parallel observation indicators. By calculating the factor loading of each parallel observation indicator, the mathematical correlation between the parallel observation indicators is quantified, and the TCM and Western medicine integration feature vector of each time-series follow-up node is obtained. Activate the full-cycle dynamic evolution network, extract the integrated Chinese and Western medicine feature vectors of the target object at the current time-series follow-up node and the historical time-series follow-up node, and input the spatiotemporal covariate features of all time-series follow-up nodes as constraints into the pre-trained state transition model to calculate the potential syndrome state transition probability matrix under different time windows. Based on the probability matrix of syndrome state transition under different time windows, a risk prediction index representing the evolution trajectory of lung cancer throughout the entire cycle is generated. Control commands are then generated based on the risk prediction index and sent to the front-end terminal to trigger hierarchical early warning signals and interface rendering interactions for preoperative screening or postoperative monitoring stages.

Images