AI Programmer Cooperation in Health Software Development

This article presents a case study on the impact of generative artificial intelligence on software development, focusing on the transformation of a mobile application for cognitive testing in patients with Parkinson’s disease. Originally developed as an undergraduate thesis project, the tool was adopted and reengineered by a single developer with no prior experience in mobile development. Through the use of ChatGPT and prompt engineering techniques, the application was refactored and optimized. The human-AI collaboration enabled the implementation of design patterns, improved code maintainability, and significantly reduced its structural complexity. A quantitative analysis revealed a 60.9% reduction in lines of code per unit of complexity and a 63.9% reduction in cyclomatic complexity compared to the original version. Furthermore, the application was used and validated in real clinical settings by the Fundación Valle de Lili, including a formal validation of the MoCA test, reinforcing its medical utility. The findings suggest that integrating AI assistants into the software development lifecycle can accelerate progress, enhance product quality, and support adoption in critical environments, provided it is paired with rigorous human oversight and sound engineering practices.