Portable integrated health parameter monitor device suitable for dialysis patients in multi-ethnic regions of frontier Xinjiang

Abstract

The invention relates to the technical field of medical health, and discloses a portable integrated health parameter monitor device suitable for dialysis patients in multi-ethnic regions of frontier Xinjiang. Comprising an input sampling module, an index processing module, a national feature embedding and index weight adjusting module, a total risk scoring and grading module and a risk interpretation and national language broadcasting module. The device is combined with portable detection hardware, and an integrated device with the following capabilities is constructed: personalized risk judgment can be performed according to a national differentiation index reference range; the method can be operated locally with low computing power in a weak network environment of frontier Xinjiang; a grading system with a transparent structure and an interpretable logic is provided; and national language voice broadcast and visual interaction output are supported, so that the detection accessibility, risk identification timeliness and intervention guidance individualization level of dialysis crowds in national regions are improved, and the influence of remote medical dependence and cultural obstacles on dialysis management quality is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical health, in particular to a portable health parameter integrated monitor device suitable for dialysis patients in border multi-ethnic areas. BACKGROUND

[0002] In the current management of chronic kidney disease maintenance hemodialysis (MHD), patients need to regularly monitor key indicators such as serum parathyroid hormone (iPTH), blood calcium, blood phosphorus, body weight changes, and blood pressure to assess calcium and phosphorus metabolism, dialysis adequacy, and complication risk. However, in China's border areas, due to insufficient medical resources, poor transportation conditions, diverse ethnic languages, and other factors, the coverage rate of grassroots follow-up is low, and detection means are scarce. Some patients are in a state of long-term risk assessment and effective intervention, which seriously affects the effectiveness of dialysis treatment and quality of life.

[0003] Existing health monitoring systems rely on centralized computing platforms and fixed threshold rules, making it difficult to address the significant differences in physiological characteristics, language communication, and cultural understanding among multi-ethnic populations. In addition, while machine learning models have some predictive ability, they often require large amounts of training data and strong computing power support, lack local adaptability and interpretability, and are not suitable for dialysis follow-up scenarios in weak network and low resource environments in border areas. SUMMARY

[0004] To overcome the deficiencies of the prior art, the present application provides a portable health parameter integrated monitor device suitable for dialysis patients in border multi-ethnic areas, which has the advantages of dynamic scoring and graded result broadcasting, solving the above technical problems.

[0005] TECHNICAL SOLUTION To achieve the above purpose, the present application provides the following technical solution: a portable health parameter integrated monitor device suitable for dialysis patients in border multi-ethnic areas, comprising an input sampling module, an index processing module, a national characteristic embedding and index weight adjustment module, a total risk score and grade division module, and a risk explanation and national language broadcast module; The input sampling module is used to collect key health indicators of patients, and the index processing module is used to normalize and standardize the key health indicators of patients to obtain the score of each key health indicator; The national characteristic embedding and index weight adjustment module sets corresponding national embedding vectors based on different nationalities and constructs the importance weight of each key health indicator of the detected patient in combination with the national sensitivity vector of each key health indicator; The total risk score and grade division module comprehensively calculates the total risk score based on the importance of the key health indicators of the detected patient and the score of the key health indicators, and gives the corresponding grade, which is output through the risk explanation and national language broadcast module.

[0006] As a preferred technical solution of the present invention, the input sampling module collects key health indicators of patients, including blood calcium, blood phosphorus, iPTH, blood pressure, weight change rate, calcium-phosphorus product and corresponding ethnic identity information; The indicator processing module includes distribution analysis of samples from various ethnic groups, and uses the quantile method to obtain the corresponding boundary values. The specific expression is as follows: in They represent ethnic groups index The 25th, 50th, and 75th percentiles, superscript Indicates the corresponding ethnic category (e.g., Han, Zhuang, Tibetan, etc.), subscript This refers to specific test indicators (such as blood calcium, blood phosphorus, blood pressure, etc.), set , representing ethnicity In terms of indicators The set of quantile thresholds; The expression for the score of each key health indicator is as follows: in, Indicates the detection indicators The score, Indicates the detection indicators The detected value.

[0007] As a preferred embodiment of the present invention, the ethnic feature embedding and indicator weight adjustment module sets corresponding ethnic embedding vectors based on different ethnic groups, and constructs the importance weights of key health indicators for each patient by combining the ethnic sensitivity vector of each key health indicator. The specific expression is as follows: in, Indicates the basic weight. Represents the ethnic embedding vector. A vector representing the ethnic sensitivity of key health indicators.

[0008] As a preferred technical solution of the present invention, the ethnic embedding vector Specifically, the acquisition is as follows: set up 4D features: Body Mass Index, risk of inherited metabolic diseases, common diseases, dietary type, frequency of physical activity, and frequency of smoking and drinking. Each feature is Z-score standardized. A national-feature matrix is ​​constructed as the national embedding vector. , They represent ethnic groups 1~ Uyghur ethnic characteristics.

[0009] As a preferred technical solution of the present invention, the ethnic sensitivity vector The specific steps to obtain it are as follows: The ElasticNet regression method is used to calculate the ethnic embedding vector. , where is the input variable and the index score is the output variable, and the ethnic sensitivity vector is obtained through training; , Indicates 1~ Dimensional indicators The sensitivity of ethnic characteristics is assessed. If the absolute value of the sensitivity of a certain dimension is lower than the threshold, it is considered that the indicator is not sensitive and the feature is removed.

[0010] As a preferred embodiment of the present invention, the total risk score and grading module calculates the total risk score based on the importance of the patient's key health indicators and the scores of the key health indicators. The specific expression is as follows: in, This represents the total number of health indicators included in the calculation. For the first The scores of each indicator These are the corresponding weighting coefficients.

[0011] Compared with existing technologies, this invention provides a portable integrated health parameter monitoring device suitable for dialysis patients in multi-ethnic border regions, which has the following beneficial effects: This invention collects key health indicators from patients, normalizes and standardizes these indicators to obtain scores for each key health indicator, and sets corresponding ethnic embedding vectors based on different ethnic groups. Weights are obtained by weighting and comprehensively evaluating these ethnic embedding vectors and the base vectors, thus achieving personalized risk assessment by integrating ethnically differentiated indicator reference ranges, achieving local adaptability and interpretability. Subsequently, the invention combines the ethnic sensitivity vector of each key health indicator to construct the importance weight of each patient's key health indicator, comprehensively calculates the total risk score, and provides a corresponding classification. This improves the accessibility of testing, the timeliness of risk identification, and the individualization of intervention guidance for dialysis populations in ethnic minority areas, reducing the impact of reliance on telemedicine and cultural barriers on the quality of dialysis management. Attached Figure Description

[0012] Fig. 1 This is a schematic diagram of the device of the present invention; Fig. 2 This is a schematic diagram of the E-RSG model structure of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Please see Figs. 1-2 A portable health parameter integrated monitoring device suitable for dialysis patients in border multi-ethnic areas includes an input sampling module, an indicator processing module, an ethnic characteristic embedding and indicator weight adjustment module, a total risk score and level classification module, and a risk interpretation and ethnic language broadcasting module. The input sampling module is used to collect key health indicators of patients, and the indicator processing module is used to normalize and standardize the key health indicators of patients to obtain the score of each key health indicator. The ethnic feature embedding and indicator weight adjustment module sets corresponding ethnic embedding vectors based on different ethnic groups, and constructs the importance weight of key health indicators for each patient by combining the ethnic sensitivity vector of each key health indicator. The overall risk score and classification module calculates the overall risk score based on the importance of the patient's key health indicators and the scores of the key health indicators, and gives the corresponding classification, which is output through the risk interpretation and ethnic language broadcasting module.

[0015] The input sampling module collects key health indicators of patients, including blood calcium, blood phosphorus, iPTH, blood pressure, weight change rate, calcium-phosphorus product, and corresponding ethnic identity information; The indicator processing module includes distribution analysis of samples from various ethnic groups, using the quantile method to obtain corresponding boundary values. The specific expression is as follows: in They represent ethnic groups index The 25th, 50th, and 75th percentiles; superscript Indicates the corresponding ethnic category (such as Han, Zhuang, Tibetan, etc.); Subscript Indicates specific test indicators (such as blood calcium, blood phosphorus, blood pressure, etc.); Therefore, set , representing ethnicity In terms of indicators Given the set of quantile thresholds, the expression for the score of each key health indicator is as follows: in, Indicates the detection indicators The score, Indicates the detection indicators The detected value.

[0016] The ethnic feature embedding and indicator weight adjustment module sets corresponding ethnic embedding vectors based on different ethnic groups, and constructs the importance weights of key health indicators for each patient by combining the ethnic sensitivity vector of each key health indicator. The specific expression is as follows: in, The basic weighting is represented by the following steps: 1. Literature and Guideline Review: Collect information from publicly available medical databases, national health guidelines, academic journals, and WHO standards, ranking the importance of this type of indicator (such as dialysis-related indicators) and suggesting weight ranges; 2. Expert Evaluation (Delphi Method): Organize 5–10 experts in the field to conduct a Delphi survey, performing at least two rounds of anonymous feedback and weighted scoring; the coefficient of variation (CV) and coefficient of conformity (W) after each round of evaluation are used to assess consistency; the average weighting result is the weighting factor. The candidate set is further tested for consistency (e.g., Kendall's W) to determine the final value. 3. Result fusion: The expert scores and literature recommendations are fused using the entropy method or principal component analysis; the fusion results are then linearly normalized to ensure statistical stability and cross-population generality. Represents the ethnic embedding vector. A vector representing the ethnic sensitivity of key health indicators.

[0017] Ethnic Embedding Vector Specifically, the acquisition is as follows: set up 4D features: Body Mass Index, risk of inherited metabolic diseases, common diseases, dietary type, frequency of physical activity, and frequency of smoking and drinking. Each feature is Z-score standardized. A national-feature matrix is ​​constructed as the national embedding vector. , They represent ethnic groups 1~ Uyghur ethnic characteristics.

[0018] Ethnic Sensitivity Vector The specific steps to obtain it are as follows: This application employs the ElasticNet regression method, using ethnic embedding vectors. , where is the input variable and the index score is the output variable, the ethnic sensitivity vector is obtained through training. , Indicates 1~ Dimensional indicators The sensitivity of ethnic characteristics is assessed. If the absolute value of the sensitivity of a certain dimension is below a threshold, it is considered insensitive to that indicator, and the feature is removed. Specifically... Sample data preparation: Collect real medical data samples from multi-ethnic populations (such as dialysis records, biochemical indicators, and health questionnaires), and set labels such as "whether in the risk range"; Regression modeling and feature selection: This application adopts the ElasticNet regression method, with the ethnic embedding vector as the input variable. The output variable is the indicator score, and the training yields an ethnic sensitivity vector. ,in Indicators For the first The sensitivity of individual ethnic characteristics is assessed. The results are ranked by the importance of coefficients; if the absolute value of the sensitivity of a certain dimension is below a threshold (e.g., 0.05), it is considered insensitive and the feature is removed. Representing the dimensions of ethnic characteristics For indicators The score has a positive reinforcing effect; if This indicates that the feature affects the index. The score has a negative weakening effect.

[0019] The overall risk score and grading module calculates the overall risk score based on the importance of the patient's key health indicators and the scores of those indicators. The specific expression is as follows: in, This represents the total number of health indicators included in the calculation. For the first The scores of each indicator These are the corresponding weighting coefficients.

[0020] Among them, the E-RSG model integrates patients' ethnic background information with dialysis-related physiological indicators to construct ethnic-specific risk thresholds and scoring rules, thereby achieving health risk assessment without training and with low computational power. The overall risk score is graded according to the following rules: Normal level: If If l < 6, output a green notification. Warning level: If 6≤ If the value is less than 11, a yellow warning will be displayed. High risk level: If If the value is ≥11, a red warning will be displayed. The system automatically matches the semantic interpretation template library according to the risk level, and generates broadcast content and text prompts in ethnic languages; Example 1 This example is for a Mandarin-speaking user - a male Yi patient. The input parameters are shown in Table 2 below: Table 2 Ethnic Embedding: Yi ethnic group embedding vector: Threshold settings (Yi ethnic group) are shown in Table 3 below: Table 3 Scoring Calculation (Based on the formula): See Table 4 below: Table 4 Weight calculation See Table 5 below: Table 5 Taking iPTH as an example, calculation ; The same logic applies to the other indicators, yielding their respective weights. ; Overall score calculation: Output level: High risk (red), broadcast content: "Your parathyroid hormone levels are severely elevated, and your blood phosphorus levels are also high. We recommend immediate re-examination and adjustment of your diet and dialysis frequency." Example 2: Zhuang ethnic minority female patient (low risk) iPTH=420pg / mL, Ca=2.25, P=1.7, Ca×P=38, SBP=130, ΔW=0.5% After scoring, the total score was 5.6, which was determined to be a green normal level. Output: "Your indicators are generally good; please continue with daily monitoring." Example 3: Tibetan male, weight data missing. The model can still score even when some inputs are missing. Missing values ​​are handled by default using a mean or zero-weight strategy, and the total score can still be obtained. The risk level is yellow. See Table 6 below for a detailed explanation of the parameters: Table 6 The present invention also includes a specific implementation device, including a housing 11, a handle 10 connected to the top of the housing 11, a screen 12 (display and touch module) embedded in the front of the housing 11, a built-in thermally sensitive printing port 15 provided on the side of the housing 11, a detection slot 17 and other interfaces and an identity recognition module 16 respectively provided on the bottom of the front of the housing 11, a main body detection chamber 14 provided inside the housing 11, and a multi-language voice indication area 13 provided inside the housing 11; Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A portable integrated health parameter monitoring device suitable for dialysis patients in multi-ethnic border regions, characterized in that: It includes an input sampling module, an indicator processing module, an ethnic feature embedding and indicator weight adjustment module, a total risk score and level classification module, and a risk interpretation and ethnic language broadcasting module; The input sampling module is used to collect key health indicators of patients, and the indicator processing module is used to normalize and standardize the key health indicators of patients to obtain a score for each key health indicator. The ethnic feature embedding and indicator weight adjustment module sets corresponding ethnic embedding vectors based on different ethnic groups, and constructs the importance weight of key health indicators for each patient by combining the ethnic sensitivity vector of each key health indicator. The overall risk score and classification module calculates the overall risk score based on the importance of the patient's key health indicators and the scores of the key health indicators, and gives the corresponding classification, which is output through the risk interpretation and ethnic language broadcasting module.

2. The portable integrated health parameter monitoring device for dialysis patients in multi-ethnic border areas according to claim 1, characterized in that: The input sampling module collects key health indicators of patients, including blood calcium, blood phosphorus, iPTH, blood pressure, weight change rate, calcium-phosphorus product, and corresponding ethnic identity information. The indicator processing module includes distribution analysis of samples from various ethnic groups, and uses the quantile method to obtain the corresponding boundary values. The specific expression is as follows: in They represent ethnic groups index The 25th, 50th, and 75th percentiles Indicates the corresponding ethnic category. Indicates the detection indicators, Representing ethnicity In terms of indicators The set of quantile thresholds; The expression for the score of each key health indicator is as follows: in, Indicates the detection indicators The score, Indicates the detection indicators The detected value.

3. The portable integrated health parameter monitoring device for dialysis patients in multi-ethnic border areas according to claim 2, characterized in that: The ethnic feature embedding and indicator weight adjustment module sets corresponding ethnic embedding vectors based on different ethnic groups, and constructs the importance weights of key health indicators for each patient by combining the ethnic sensitivity vector of each key health indicator. The specific expression is as follows: in, Indicates the basic weight. Represents the ethnic embedding vector. A vector representing the ethnic sensitivity of key health indicators.

4. The portable integrated health parameter monitoring device for dialysis patients in multi-ethnic border areas according to claim 3, characterized in that: The ethnic embedding vector Specifically, the acquisition is as follows: set up Dimensional features: Body mass index, risk of inherited metabolic diseases, common diseases, dietary type, frequency of physical activity, and frequency of smoking and drinking were all evaluated using Z-score standardization. An ethnic-feature matrix was constructed as the ethnic embedding vector. , They represent ethnic groups 1~ Uyghur ethnic characteristics.

5. The portable integrated health parameter monitoring device for dialysis patients in multi-ethnic border areas according to claim 3, characterized in that: The ethnic sensitivity vector The specific steps to obtain it are as follows: The ElasticNet regression method is used to calculate the ethnic embedding vector. , where is the input variable and the index score is the output variable, the ethnic sensitivity vector is obtained through training. , Indicates 1~ Dimensional indicators The sensitivity of ethnic characteristics is assessed. If the absolute value of the sensitivity of a certain dimension is lower than the threshold, it is considered that the indicator is not sensitive and the feature is removed.

6. The portable integrated health parameter monitoring device for dialysis patients in multi-ethnic border areas according to claim 3, characterized in that: The overall risk score and grading module calculates the overall risk score based on the importance of the patient's key health indicators and the scores of those indicators. The specific expression is as follows: in, This represents the total number of health indicators included in the calculation. For the first The scores of each indicator These are the corresponding weighting coefficients. This indicates a summation.

Images