Ai-accelerated modernization of legacy healthcare software

Abstract

Modernizing legacy healthcare software systems remains a critical challenge due to strict compliance requirements, operational inefficiencies, and patient safety concerns. This thesis introduces an AI-accelerated modernization framework that combines the industry-standard 6R methodology (Rehost, Refactor, Replatform, Rearchitect, Rebuild, Replace) with advanced AI tools to address both regulatory compliance and performance optimization. The framework follows a dual-path strategy: (i) frontend redevelopment using .NET Core 7 and React to enhance user experience and system responsiveness, and (ii) backend migration from Oracle 11g to Oracle 19c within an Oracle Linux containerized environment to strengthen data security, scalability, and availability. AI tools including ChatGPT, GitHub Copilot, Cursor.ai, and SonarQube were integrated across documentation, coding, testing, and compliance validation activities. Empirical evaluation demonstrates significant improvements: user interface load times reduced by 39%, report generation accelerated by 90%, unit test coverage increased from 58% to 92%, deployment cycles shortened by 94%, and manual coding effort reduced by 38%. Moreover, documentation and compliance workloads declined by 60–75%, while data integrity was preserved with near-zero data loss. These results validate the feasibility of AI-accelerated modernization in meeting regulatory standards such as EU MDR, HIPAA, and GDPR while enabling operational excellence. The framework establishes a practical blueprint for healthcare organizations seeking to modernize mission-critical systems while balancing automation, compliance, and long term sustainability.