Standard-Compliant Defect Management with Data-Driven Analysis in Medical Software Development

Abstract

The aim of this study was to define a standard-compliant defect management approach, structure defect data, and apply statistical and analytical methods in the context of low-risk medical software development. The research question addressed how such a system can be implemented and how defect data can support quality improvement and decision-making.

The study applies a Design Science Research (DSR) approach combined with a case study. Defect data from Azure DevOps was analysed using statistical and analytical methods, including cumulative defect S-curve and Pareto analysis to identify patterns in defect occurrence and process performance. The scope of the study covers the definition of a defect life cycle, data collection and structuring, and analysis. The study is limited to a single low-risk medical software project.

The results of this study demonstrate that a standard-compliant defect management system can be implemented using commercial tools and integrated into existing development workflows. The collected defect data can be utilised to derive multiple metrics and perform trend analyses, enabling the identification of high-risk areas and supporting data-driven decision-making throughout the entire software life cycle, from development to post-market use.

Keywords: IEC 62304, ISO 13485, ISO 14971, defect management, defect life cycle, defect data analysis, medical software, defect trends, FDA, MDR, S-curve, Pareto