AI-DRIVEN MEDICAL FUNDRAISING VERIFICATION SYSTEM TO DETECT AND PREVENT FRAUDULENT TREATMENT REQUESTS
- Version
- Download 12
- File Size 819.23 KB
- File Count 1
- Create Date 11 June 2025
- Last Updated 11 June 2025
AI-DRIVEN MEDICAL FUNDRAISING VERIFICATION SYSTEM TO DETECT AND PREVENT FRAUDULENT TREATMENT REQUESTS
Authors:
- Vasuki1, Dr. T. Amalraj Victoire2, I. Charanya3
1Associate Professor, Department of Computer Applications, Sri Manakula Vinayagar Engineering College (Autonomous), Puducherry 605107, India,
2Professor, Department of Computer Applications, Sri Manakula Vinayagar Engineering College (Autonomous), Puducherry 605107, India,
3Post Graduate student, Department of Computer Applications, Sri Manakula Vinayagar Engineering College (Autonomous),
Puducherry 605107, India charanyairisappin@gmail.com
ABSTRACT: Medical fundraising is a crucial source of financial aid for individuals requiring support for treatments, surgeries, or emergencies. Crowdfunding platforms and charitable organizations play a vital role in raising funds, but fraudulent activities such as fabricating treatment bills have become a significant challenge. Fraudulent requests erode donor trust and divert resources from genuine beneficiaries. Existing fraud detection methods, often manual or semi-automated, are inefficient, error-prone, and struggle to identify sophisticated fraud attempts.
This project introduces an AI-driven system to detect and prevent fraudulent medical fund requests. The system integrates YOLOv8 for detecting text in uploaded treatment bills and PaddleOCR for text extraction. Extracted details such as hospital names, patient data, and treatment costs—are validated against a trusted hospital dataset using the Fuzzy Matching Algorithm, ensuring accurate verification by identifying discrepancies between records. By automating text detection, recognition, and pattern matching, this system enhances verification accuracy, safeguards donor contributions, and fosters trust in medical crowdfunding.
Keywords: Fraud Detection, Medical Fundraising, YOLOv8, PaddleOCR, Fuzzy Matching Algorithm.
Download