Related Experiment Video
Updated: Dec 6, 2025
04:04
Asthma Detection Research Based on Voice Signal Processing and Machine Learning
Published on: July 22, 2025
743
Medical Fraud and Abuse Detection System Based on Machine Learning.
Conghai Zhang1, Xinyao Xiao2, Chao Wu1
1School of Management, Zhejiang University, Hangzhou 310058, China.
Summary
This study introduces a novel neural network to detect medical fraud and abuse by scoring disease-drug relationships. The model significantly outperforms previous methods, aiding healthcare analysts in identifying anomalies.
Area of Science:
- Healthcare analytics
- Machine learning in medicine
- Medical fraud detection
Background:
- Approximately 10% of healthcare spending is lost to fraud and abuse.
- The complexity of drug-disease interactions complicates healthcare oversight.
Purpose of the Study:
- To develop a quantitative method for assessing disease-drug relationships.
- To implement an anomaly detection system for identifying medical fraud and abuse.
Main Methods:
- Proposed a neural network incorporating fully connected layers and sparse convolution.
- Utilized a focal-loss function to address data imbalance.
- Introduced a relative probability score for performance evaluation.
Main Results:
- The developed neural network demonstrated superior performance compared to existing models.
- The model effectively quantifies disease-drug relationships for anomaly detection.
Conclusions:
- The proposed model offers a significant advancement in detecting medical fraud and abuse.
- This approach can substantially reduce the workload for healthcare analysts.
