Guardat en:
| Autors principals: | , |
|---|---|
| Format: | Recurso digital |
| Idioma: | anglès |
| Publicat: |
Zenodo
2024
|
| Matèries: | |
| Accés en línia: | https://doi.org/10.5281/zenodo.15087787 |
| Etiquetes: |
Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
|
Taula de continguts:
- <div><strong>Introduction</strong>: The integrity of the medical equipment supply chain is essential for patient safety and regulatory compliance, particularly in high-demand environments like the United States. Rising incidents of counterfeit medical equipment have highlighted vulnerabilities within this chain, underscoring the need for a robust, data-driven approach to fraud detection.</div> <div><strong>Methodology</strong>: A diverse dataset was acquired, encompassing supplier transaction records, IoT-generated environmental metrics, and digital footprint data. Machine learning models, including Isolation Forests, support vector machines, and recurrent neural networks, were employed to identify anomalies across multiple dimensions. Privacy-preserving techniques like homomorphic encryption and federated learning were integrated to comply with data protection standards.</div> <div><strong>Results</strong>: The framework achieved high detection accuracy, with a significant reduction in false positives across a wide range of transaction and environmental anomalies. Real-time IoT monitoring enabled prompt detection of tampering and environmental fluctuations, enhancing the algorithm’s fraud detection capabilities.</div> <div><strong>Conclusion</strong>: This framework provides a scalable, compliance-focused solution for securing the medical equipment supply chain. Its relevance to the U.S. healthcare industry is underscored by its ability to ensure the authenticity of medical devices, ultimately supporting patient safety and regulatory mandates.</div>