Multi-task meta-attention network for traditional Chinese medicine diagnostic recommendation

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Biomedical Data Science

Background:

• Increasing medical data and technological advancements necessitate personalized assisted diagnosis.
• Implementing recommendation systems in healthcare faces significant challenges, requiring further research.
• The growing need for accurate diagnostic support in smart healthcare environments.

Purpose of the Study:

• To explore the application of recommendation technology in smart healthcare.
• To design a deep learning model integrating medical knowledge for enhanced diagnostic support.
• To improve the accuracy and efficiency of clinical decision-making through AI.

Main Methods:

• Developed a tailored feature engineering process for medical data, including preparation, selection, and transformation.
• Designed a knowledge-matching deep learning model for medical data analysis and prediction.
• Leveraged nonlinear fitting and feature learning capabilities for enhanced evaluation metrics.

Main Results:

• The proposed deep learning model achieved an average improvement of +2.7% in Hits@10 compared to baseline models.
• Demonstrated effective processing of medical data for accurate predictions in Traditional Chinese Medicine (TCM) clinical recommendations.
• Provided valuable insights supporting clinical decision-making.

Conclusions:

• The deep learning model offers significant support for advancing smart medical technology.
• The model shows strong potential for medical data analysis and clinical decision-making.
• Contributes to enhanced healthcare quality, efficiency, and the overall advancement of the medical field.