An electronic medical record and medical record cataloging classification method, system and device

By employing technologies such as medical OCR, variational autoencoders, and medical knowledge graphs for the preprocessing and feature extraction of electronic medical records and case files, and combining BERT and ResNet models for multimodal feature fusion, a transfer learning classification model is constructed and its parameters are optimized. This solves the problems of accuracy and adaptability in cataloging and classifying multi-format medical records, and achieves efficient and stable medical record management and classification.

CN122173590APending Publication Date: 2026-06-09BEIJING YINGYAN CHUANGXIN TECH DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING YINGYAN CHUANGXIN TECH DEV CO LTD
Filing Date
2026-02-25
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing technologies for electronic medical records and medical record cataloging and classification suffer from problems such as low accuracy in extracting multi-format data, insufficient standardization of terminology, weak correlation of multimodal features, and poor model generalization, making it difficult to adapt to the classification needs of cross-institutions and new diseases.

Method used

Preprocessing is performed using a medical OCR model, a variational autoencoder anomaly detection algorithm, a medical knowledge graph word segmentation tool, and a Transformer bidirectional mapping terminology standardization model. Feature extraction and fusion are then performed using a BERT model, a ResNet convolutional neural network, and a multi-head attention mechanism to construct a classification model based on transfer learning. The model parameters are optimized using an uncertainty sampling and causal forest dynamic update framework.

Benefits of technology

It achieves efficient text extraction and anomaly cleaning of multi-format medical records, improves data standardization, enhances multimodal feature association capabilities, adapts to different medical institution scenarios, lowers the threshold for cross-scenario applications, reduces the workload of manual annotation, and improves classification accuracy and stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122173590A_ABST
    Figure CN122173590A_ABST
Patent Text Reader

Abstract

The application discloses an electronic medical record and a cataloging and classifying method, system and device thereof. The method comprises the following steps: obtaining standardized data by preprocessing multi-format medical record original data through a medical OCR model, a VAE anomaly detection algorithm, a medical knowledge graph word segmentation tool and a Transformer term standardization model; checking semantic rationality, extracting multi-dimensional features, and obtaining a comprehensive feature vector after strengthening and fusing; constructing a transfer learning classification model and training the model, inputting the feature vector to generate cataloging information; and finally, evaluating the classification result through an active learning mechanism, combining artificial labeling data and a causal forest dynamic updating framework, and incrementally learning and optimizing the model performance. The application solves the problems of low multi-format data extraction accuracy, insufficient term standardization and poor model generalization in the prior art.
Need to check novelty before this filing date? Find Prior Art