Scholarlink

Post-market surveillance (PMS) has become a critical aspect of medical device regulation, imposing significant financial and operational challenges on manufacturers and distributors. Despite the establishment of regulatory mechanisms, gaps in monitoring persist, leading to severe consequences for both healthcare systems and patient safety. With the increasing integration of artificial intelligence (AI) into healthcare technology, it is essential to explore how AI can transform traditional PMS processes by providing real-time, data-driven insights that ensure continuous compliance with safety and technical specifications. This paper presents a comprehensive analysis of various AI models applied to predictive PMS, focusing specifically on anesthesia machines and ventilators. It examines existing AI models alongside custom-made approaches, comparing their technical requirements, implementation feasibility, and long-term usability. Furthermore, the study establishes a framework for integrating AI-driven PMS into Inventory and Maintenance Management Information Systems (IMMIS), enhancing proactive failure detection and regulatory adherence.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

AI-Driven Post-Market Surveillance for Predicting Anesthesia Machine and Mechanical Ventilators Failures Based on Performance Data

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Nejra Merdovic,
  • Faruk Becirovic,
  • Lemana Spahic,
  • Ilma Gusinac,
  • Adna Softić,
  • Živorad Kovačević,
  • Lejla Gurbeta Pokvić,
  • Velid Dlakic

摘要

Post-market surveillance (PMS) has become a critical aspect of medical device regulation, imposing significant financial and operational challenges on manufacturers and distributors. Despite the establishment of regulatory mechanisms, gaps in monitoring persist, leading to severe consequences for both healthcare systems and patient safety. With the increasing integration of artificial intelligence (AI) into healthcare technology, it is essential to explore how AI can transform traditional PMS processes by providing real-time, data-driven insights that ensure continuous compliance with safety and technical specifications. This paper presents a comprehensive analysis of various AI models applied to predictive PMS, focusing specifically on anesthesia machines and ventilators. It examines existing AI models alongside custom-made approaches, comparing their technical requirements, implementation feasibility, and long-term usability. Furthermore, the study establishes a framework for integrating AI-driven PMS into Inventory and Maintenance Management Information Systems (IMMIS), enhancing proactive failure detection and regulatory adherence.