African maternal nutrition health dynamic assessment and intervention system
The dynamic assessment and intervention system for maternal nutrition and health in Africa, generated through multi-source data fusion and offline intervention packages, solves the problems of data fragmentation and high costs. It enables dynamic nutritional assessment and low-cost nutritional intervention, adapts to the complex environment in Africa, and improves the accuracy of assessment and the timeliness of response.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WENZHOU MEDICAL UNIV
- Filing Date
- 2026-02-04
- Publication Date
- 2026-06-02
AI Technical Summary
Nutritional health assessments for pregnant and postpartum women in Africa suffer from problems such as fragmented data, delayed assessments, high costs, and incompatibility with local environments. Existing technologies cannot achieve dynamic responses and low-cost nutritional interventions.
A dynamic assessment and intervention system for maternal nutrition and health in Africa was designed. Through multi-source data fusion, weight correction, and offline intervention package generation, the system enables dynamic assessment and low-cost intervention of maternal nutrition and health. The system includes a data acquisition module, a data preprocessing and weight correction module, a nutritional risk assessment module, and an offline intervention package generation module. It integrates clinical physiological data, traditional Chinese medicine physical signs data, local dietary data, and environmental risk data to generate nutritional intervention strategies that can be executed offline.
It enables unified evaluation of multi-source data, improves the accuracy and timeliness of evaluation, reduces costs, adapts to the complex environment in Africa, enhances the robustness of the system, and supports nutrition guidance in environments with weak or no networks.
Smart Images
Figure CN122135890A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of maternal health management technology, and more specifically, relates to a dynamic assessment and intervention system for the nutritional health of pregnant and postpartum women in Africa. Background Technology
[0002] In parts of Africa, pregnant women commonly face problems such as malnutrition, insufficient medical resources, and complex environmental risks. Current technologies suffer from data fragmentation and low adaptability. Clinical indicators (such as Hb levels), dietary records (using local ingredients like cassava leaves), and environmental factors (such as rainy season floods) for African pregnant women are scattered across different systems, and internationally accepted models do not include pregnancy-specific parameters in Africa (such as iron absorption disorders in malaria-prone areas). Intervention is also delayed, with traditional nutritional assessments taking longer than 3 months, failing to dynamically respond to sudden changes during pregnancy (such as acute nutritional deficiencies caused by morning sickness). Furthermore, current solutions rely on smart wearable devices (costing over $50 USD), resulting in a coverage rate of less than 15% in Africa. Summary of the Invention
[0003] The technical problem to be solved by this invention is to provide a dynamic assessment and intervention system for maternal nutrition and health in Africa, which can realize the fusion of multi-source health-related data on maternal nutrition and health, dynamic nutritional risk assessment, and low-cost intervention implementation.
[0004] The present invention provides a dynamic assessment and intervention system for maternal nutrition and health in Africa, comprising: The data acquisition module is used to collect multi-source data from pregnant and postpartum women. The data preprocessing and weight correction module is used to standardize multi-source health-related data and adjust the weights of different data sources based on regional environmental risk parameters. The nutrition risk assessment module is used to generate maternal nutrition risk scores based on processed and corrected data. The dynamic decision-making module is used to generate corresponding nutritional intervention strategies based on nutritional risk scores. The offline intervention package generation module is used to convert nutritional intervention strategies into intervention instructions that can be executed offline and output them under conditions of limited network or terminal performance.
[0005] As a further improvement to the present invention, multi-source health-related data includes: Clinical physiological data, including hemoglobin levels, weight information, and gestational age information; Traditional Chinese medicine physical signs and characteristics data, including tongue diagnosis image feature parameters and pulse rhythm feature parameters, are used for auxiliary assessment of nutritional status; Data on African local diets and communities was obtained through community questionnaires, including at least information on the frequency of consumption of local ingredients; Environmental and regional risk data, including rainfall data, flood warning data, and regional safety risk index data.
[0006] As a further improvement of the present invention, the hemoglobin value in the clinical physiological data is collected by a portable detection device.
[0007] As a further improvement of the present invention, the tongue diagnosis image data is collected by a tongue diagnosis instrument or obtained by image recognition of a paper tongue coating colorimetric card; the pulse rhythm data is collected by a pulse sensor to obtain the pulse frequency per unit time and obtain the pulse rhythm data.
[0008] As a further improvement of the present invention, the RGB values of the tongue coating are acquired through tongue diagnosis images, and different RGB values represent different physical signs.
[0009] As a further improvement to the present invention, the environmental and regional risk data are obtained respectively through the following methods: Rainfall Warning API: Connects to the Africa Meteorological Service data stream to obtain real-time rainfall data for the next 7 days; Conflict Area Index: Integrates the ACLED armed conflict database to calculate regional risk values based on conflict frequency.
[0010] As a further improvement of the present invention, the dynamic decision-making module includes a medicinal herb efficacy mapping unit, which is used to map preset medicinal herb efficacy rules to nutritional supplementation solutions available in Africa.
[0011] As a further improvement of the present invention, the dynamic decision-making module also includes an environmental adaptation unit, which adjusts the nutrient priority based on rainfall data.
[0012] As a further improvement of the present invention, the offline intervention package generated by the offline intervention package generation module includes at least nutrition protocol coding and medical resource coding; The coding rules for offline intervention packages include a header identifier, a nutrition plan code, a medical resource code, and a checksum. Offline intervention packages are sent to the target terminal via SMS or short-range wireless communication.
[0013] As a further improvement of the present invention, the offline intervention package generation module is equipped with adaptive generation logic: When in a high-power environment: Send the complete code and automatically redirect to the navigation interface; When in a low-power environment: the text and images are plain text commands; In the absence of network access: Distribute to community nodes via Bluetooth wireless mesh network.
[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: The system achieves multi-source data fusion assessment, incorporating clinical indicators, vital signs, local diets, and environmental risks into a unified assessment model to avoid data fragmentation; through multi-source health data fusion and dynamic weight correction mechanisms, it improves the accuracy of maternal nutrition assessment in the complex environment of Africa. To improve regional adaptability, the assessment and intervention results are made more consistent with the actual conditions in different regions of Africa through weight correction mechanisms and local food mapping rules, thereby improving the feasibility of intervention programs under local resource conditions. Improve response timeliness, support real-time assessment based on dynamic data, and reduce intervention delays caused by the excessively long cycle of traditional nutrition assessment; It reduces implementation costs, does not rely on high-cost smart wearable devices, is suitable for areas in Africa with scarce medical resources, and has good promotional value and social benefits; Enhance system robustness so that nutritional guidance can still be provided through offline intervention packages even in environments with weak or no network coverage. Attached Figure Description
[0015] Figure 1 This is a schematic diagram of the various modules of the system of the present invention; Figure 2 This is a schematic diagram of the data acquisition module of the present invention; Figure 3 This is a schematic diagram of the dynamic decision-making module of the present invention. Detailed Implementation
[0016] The present invention will be further described in detail below with reference to the embodiments. The following embodiments are explanations of the present invention, but the present invention is not limited to the following embodiments.
[0017] Specific Implementation Example 1: Please refer to... Figures 1-3 A dynamic assessment and intervention system for maternal nutrition and health in Africa includes a data acquisition module, a data preprocessing and weight correction module, a nutritional risk assessment module, a dynamic decision-making module, and an offline intervention package generation module. The data preprocessing and weight correction module is connected to the data acquisition module. It preprocesses the data collected by the data acquisition module and corrects the weights of different data sources. The nutrition risk assessment module is connected to the data preprocessing and weight correction module. It generates a nutrition risk score based on the processed data and the weighted data. The dynamic decision-making module is connected to the nutrition risk assessment module. It generates a nutrition intervention plan based on the nutrition risk score. Under conditions of limited network or simplified equipment, it automatically generates an offline executable nutrition intervention instruction package through the offline intervention package generation module and sends it to the target person, so as to realize rapid assessment, real-time adjustment and low-cost intervention of pregnancy nutrition status.
[0018] The data acquisition module is used to acquire multi-source data related to pregnant and postpartum women, including clinical physiological data, traditional Chinese medicine physical signs data, local African diet and community data, and environmental and regional risk data. Clinical physiological data, including hemoglobin (Hb) levels, weight, and gestational age, are used to reflect the basic physiological state of pregnant women.
[0019] Hemoglobin levels can be collected using portable testing devices (with an error of <3%); the frequency of morning sickness in pregnant women can be collected using the KylinCare APP.
[0020] Traditional Chinese medicine physical signs data, including tongue diagnosis image data and pulse rhythm data.
[0021] Tongue diagnosis image data can be collected through a tongue diagnosis instrument or obtained by image recognition of a paper tongue coating color chart, such as tongue coating thickness and color; the RGB values of the tongue coating are collected through the tongue diagnosis image, and different RGB values represent different physical signs, for example: #D32F2F represents blood deficiency.
[0022] Pulse rhythm data is collected by a pulse sensor (pulse meter) to obtain the pulse frequency per unit time and obtain pulse rhythm data. For example, a slow pulse ≤ 4 beats / second indicates insufficient qi and blood.
[0023] The system is equipped with a tongue diagnostic instrument interface, which can be connected to a tongue diagnostic instrument and / or a pulse diagnostic instrument via a USB / Bluetooth connection.
[0024] Data on local African diets and communities, including information on the frequency of food intake collected through community questionnaires, such as the intake of local ingredients like cassava leaves and moringa leaves.
[0025] Environmental and regional risk data, including rainfall data, flood warning data, and conflict area index data, can be obtained through external data interfaces. Among them, rainfall and flood warning data are used to reflect the degree of climate anomalies in the target area, and conflict area index is used to reflect the level of regional security risk.
[0026] Environmental and regional risk data can be obtained through the following methods: Rainy Season Warning API: Connects to the Africa Meteorological Service data stream to obtain real-time rainfall forecasts for the next 7 days (threshold > 50mm triggers a warning); Conflict Area Index: Integrates the ACLED armed conflict database and calculates the regional risk value (0.2-1.0) based on conflict frequency.
[0027] The data preprocessing and weight correction module is used to normalize the collected multi-source data and assign different weights to different data sources based on regional environmental factors.
[0028] Regional environmental factors refer to the rainfall or flood risk index and conflict area risk index in the region where the pregnant woman is located, as well as other environmental and regional risk data.
[0029] Calculation formula: Weight of a data source = Base weight × π(1 - Risk coefficient_i) Example: Community questionnaire weight = 0.3 (1 - Conflict area coefficient; if the conflict area coefficient is 0.5 and the rainfall risk coefficient is 0.3, the final weight may be the base weight × (1 - 0.5) × (1 - 0.3). Weight adjustment aims to reduce the contribution of unreliable data in high-risk environments.
[0030] The nutritional risk assessment module is used to quantitatively assess the current nutritional status of pregnant and postpartum women based on the collected and processed data. For example, it calculates the blood and qi index based on hemoglobin levels and tongue diagnosis scores, and calculates a comprehensive nutritional risk score based on data such as iron absorption rate in malaria-prone areas and local food ingredients.
[0031] The dynamic decision-making module is used to generate nutritional intervention strategies based on nutritional risk assessment results, including medicinal efficacy mapping decisions and environmental adaptation decisions. Herbal efficacy mapping decision: Based on preset ontology rules, the system establishes herbal efficacy equivalence rules, mapping traditional Chinese medicine blood-tonifying herbs to ingredients that can be obtained locally in Africa; for example, triggering a blood and qi regulation plan based on tongue coating data.
[0032] The principle of equivalence of medicinal efficacy applies, for example: using Moringa leaves (20mg iron + 500IU vitamin A) instead of Angelica sinensis to replenish blood, and using baobab tree instead of Astragalus membranaceus.
[0033] Examples of partial mappings in the Traditional Chinese Medicine-African Herbal Medicine Mapping Table: Traditional Chinese Medicine Efficacy African alternative medicines Equivalent dose ratio Replenish Qi and Blood Moringa leaves + baobab fruit 1：1.2 Nourishing Yin and Moistening Dryness Shea butter 0.8g replaces 1g Regarding its "qi and blood replenishing" effect, the African alternative herbal formula is "moringa leaves + baobab fruit," with an equivalent dosage ratio of "1:1.2." Here, "1:1.2" refers to the overall ratio of the traditional Chinese medicine herb to the African alternative, not the individual proportions of moringa leaves and baobab fruit. 1:1.2 means that if the standard dosage of a traditional Chinese medicine qi and blood replenishing herb (such as angelica) is 1 unit, then the moringa leaf + baobab fruit combination requires 1.2 units to achieve an equivalent substitution. The approximate ratio of moringa leaves to baobab fruit is 1:1 (this may vary depending on individual differences in the herbs). "Replenishing Qi and Blood" is a broad effect that may encompass a variety of Chinese medicinal herbs (such as Angelica sinensis, Astragalus membranaceus, Codonopsis pilosula, etc.). During the mapping process, the system will select the closest African ingredient based on preset rules, rather than a single fixed herb. Nourishing Yin and Moistening Dryness does not specify any particular Chinese medicinal herbs, but common Chinese medicinal herbs include Adenophora stricta and Ophiopogon japonicus, with shea butter as a substitute ingredient.
[0034] The ontology rules include a local food database, which collects data on relevant local foods, such as teff and cassava leaves. Tetrandric acid is highly nutritious, rich in amino acids, protein, various trace elements, and dietary fiber; its calcium content is higher than that of milk, and its iron content is twice that of wheat. Cassava leaves are rich in protein, minerals, and vitamins.
[0035] Environmental adaptation decision-making: Adjust nutritional priorities based on rainfall data; for example, automatically increase calcium supplementation priority by 2 levels during the rainy season and flood season.
[0036] The offline intervention package generation module selects the corresponding instruction generation strategy based on the network status and battery status of the terminal device, and sends it to the target person via SMS. The offline intervention package selects an adaptive generation logic based on environmental parameters such as network status and battery status (e.g., sending a complete encoded command when the battery is high and sending a plain text command when the battery is low), and then sends the intervention command package to the target terminal through the optimal offline channel (e.g., SMS, Bluetooth Mesh network).
[0037] When network unavailability is detected, prioritize generating text-based intervention instructions that can be transmitted via SMS or short-range communication.
[0038] The offline intervention package includes nutrition program codes, medical resource codes, and check codes.
[0039] Offline intervention package coding rules: Header identifier: KYLIN-EMG (Emergency Intervention Identifier); Nutritional regimen codes: Example: ZN2 - Mid-pregnancy birth tablets (Z represents mid-pregnancy) M1 - 1 tablet daily (M2 is 2 tablets); Medical resource code: PHC09 (9th Community Health Center); Checksum: CRC8 checksum (e.g., 0x2F).
[0040] Complete example analysis of offline intervention package, KYLIN-EMG: ZN2-M1|PHC09|0x2F means: KYLIN-EMG: ZN2-M1: In case of emergency, take the prenatal vitamin D supplement during the second trimester immediately (1 tablet daily). PHC09: Head to Community Health Center No. 9 (GPS coordinates are pre-stored in the app); 0x2F: Verification code verification command.
[0041] The offline intervention package generation module is equipped with adaptive generation logic: When in a high-power environment: Send the complete code and automatically redirect to the navigation interface; When in a low-power environment: The image shows the plain text command "SOS: Take the purple pill → Find PHC09"; In the absence of a network: Distribute to community nodes via Bluetooth Mesh network.
Claims
1. A dynamic assessment and intervention system for maternal nutrition and health in Africa, characterized in that: include: The data acquisition module is used to collect multi-source data from pregnant and postpartum women. The data preprocessing and weight correction module is used to standardize multi-source health-related data and adjust the weights of different data sources based on regional environmental risk parameters. The nutrition risk assessment module is used to generate maternal nutrition risk scores based on processed and corrected data. The dynamic decision-making module is used to generate corresponding nutritional intervention strategies based on nutritional risk scores. The offline intervention package generation module is used to convert nutritional intervention strategies into intervention instructions that can be executed offline and output them under conditions of limited network or terminal performance.
2. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: Multi-source health-related data includes: Clinical physiological data, including hemoglobin levels, weight information, and gestational age information; Traditional Chinese medicine physical signs and characteristics data, including tongue diagnosis image feature parameters and pulse rhythm feature parameters, are used for auxiliary assessment of nutritional status; Data on African local diets and communities was obtained through community questionnaires, including at least information on the frequency of consumption of local ingredients; Environmental and regional risk data, including rainfall data, flood warning data, and regional safety risk index data.
3. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: Hemoglobin levels in clinical physiological data were collected using portable testing devices.
4. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: Tongue diagnosis image data is collected by a tongue diagnosis instrument or obtained by image recognition of a paper tongue coating color chart; pulse rhythm data is collected by a pulse sensor to obtain the pulse frequency per unit time and obtain pulse rhythm data.
5. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 4, characterized in that: The RGB values of the tongue coating are collected through tongue diagnosis images, and different RGB values represent different physical signs.
6. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: Environmental and regional risk data were obtained through the following methods: Rainfall Warning API: Connects to the Africa Meteorological Service data stream to obtain real-time rainfall data for the next 7 days; Conflict Area Index: Integrates the ACLED armed conflict database to calculate regional risk values based on conflict frequency.
7. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: The dynamic decision-making module includes a medicinal herb efficacy mapping unit, which maps preset medicinal herb efficacy rules to nutritional supplementation solutions available in Africa.
8. The dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: The dynamic decision-making module also includes an environment adaptation unit, which adjusts nutrient priority based on rainfall data.
9. A dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: The offline intervention package generated by the offline intervention package generation module includes at least nutrition protocol codes and medical resource codes; The coding rules for offline intervention packages include a header identifier, a nutrition plan code, a medical resource code, and a checksum. Offline intervention packages are sent to the target terminal via SMS or short-range wireless communication.
10. A dynamic assessment and intervention system for maternal nutrition and health in Africa according to claim 1, characterized in that: The offline intervention package generation module is equipped with adaptive generation logic: When in a high-power environment: Send the complete code and automatically redirect to the navigation interface; When in a low-power environment: the text and images are plain text commands; In the absence of network access: Distribute to community nodes via Bluetooth wireless mesh network.