Large language models streamline automated machine learning for clinical studies

A knowledge gap exists between machine learning (ML) developers and clinicians, hindering ML's use in clinical data analysis. This study investigated ChatGPT Advanced Data Analysis (ADA), an extension of GPT-4, to bridge this gap. ChatGPT ADA autonomously developed state-of-the-art ML models for predicting clinical outcomes using real-world clinical datasets from large trials. A head-to-head comparison with manually crafted models showed no significant performance differences, with ChatGPT ADA models often outperforming their counterparts. ChatGPT ADA offers a promising avenue to democratize ML in medicine but should enhance, not replace, specialized training and resources.

Soroosh Tayebi Arasteh, Tianyu Han, Mahshad Lotfinia, Christiane Kuhl, Jakob Nikolas Kather, Daniel Truhn, Sven Nebelung