A drug interaction prediction method based on DS evidence theory
By integrating multimodal drug data and DS evidence theory into a decision fusion approach, the problem of inaccurate drug interaction prediction in existing technologies is solved, achieving higher accuracy and reliability in drug interaction prediction and reducing the risk of side effects from drug combinations.
Patent Information
- Application Number
- CN202210837179.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-15
- Publication Date
- 2026-01-13
- Estimated Expiration
- 2042-07-15
AI Technical Summary
Existing methods for predicting drug interactions only focus on the structural characteristics of drugs, resulting in inaccurate and unreliable predictions that cannot effectively predict the potential side effects of drug combinations.
A method based on Dirichlet evidence theory is adopted to integrate multimodal drug data, including molecular structure, target and metabolic enzyme data. Prediction results are generated through deep learning network, and decision fusion is performed using Dirichlet distribution and Dirichlet evidence theory to improve the reliability and accuracy of prediction.
It improves the accuracy and reliability of drug interaction prediction, can clearly express the confidence level of the prediction, and reduces the risk of side effects from drug combinations.