MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641107045 A) filed by Kkr & Ksr Institute Of Technology And Sciences on September 05, 2026, for An Integrated Multi-Model Approach For Automated Medicare Fraud Detection And Prevention.
Inventors include Mr. Prof. Repudi Ramesh; Mrs. Dr. Chittineni Aruna; Mrs. P. Kusuma; Karanam Purna Teja; Kavalakuntla Deva Sujan; Mohammad Gul E Adnan Sameer; and Kesana Rahul Ram.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: ABSTRACT [0011] With the rapid growth of healthcare systems and digital claim processing, a large amount of Medicare claim data is continuously generated and stored in healthcare databases. Many healthcare providers and users often submit identical or similar claims, leading to fraudulent activities and inefficient utilization of auditing resources. Fraud detection is an effective technique used to identify suspicious claims and reduce financial losses in healthcare systems. However, performing fraud detection in large-scale healthcare environments introduces serious challenges, especially when dealing with complex patterns and high-dimensional data. This invention proposes a hybrid healthcare fraud detection system with intelligent analysis capabilities for Medicare environments. The proposed system allows only valid and legitimate claims to be processed while ensuring that fraudulent claims are accurately identified. To achieve effective fraud detection, an autoencoder-based anomaly detection model is employed, enabling identification of unusual claim patterns without exposing underlying complexities. Additionally, machine learning techniques such as XGBoost are used to provide accurate classification based on extracted features. [0012] A key feature of the proposed system is explainable and dynamic fraud analysis. When claim characteristics change or suspicious patterns are detected, the system ensures that such claims are flagged and analyzed accordingly. This is achieved through feature importance evaluation and explainable artificial intelligence techniques such as SHAP, which prevent misclassification and provide transparency in decision-making. The system also protects against misleading patterns and ensures reliable prediction outcomes. Experimental evaluation is performed using healthcare claim datasets that simulate real Medicare scenarios with diverse and repetitive claim patterns. The results demonstrate that the proposed system effectively improves fraud detection accuracy while maintaining computational efficiency and scalability. Overall, the proposed approach provides a reliable, efficient, and interpretable solution for fraud detection in healthcare claim processing systems.
Disclaimer: Curated by HT Syndication.
