Sample size determination for machine learning in medical research

Machine learning (ML) methods are being increasingly used across various domains of medicine research. However, despite advancements in the use of ML in medicine, clear and definitive guidelines for determining sample sizes in medical ML research are lacking. This article proposes a method for determining sample sizes for medical research utilizing ML methods, beginning with the determination of the testing set sample size, followed with the determination of the training set and total sample sizes. Introduction Machine learning (ML) methods are being increasingly used in medical research, spanning various domains of medicine from oncology, orthopaedics, ophthalmology and general practice (Sirocchi et al., 2024). However, despite this advancement in medical research, currently there are no clear and definitive guidelines for determining sample sizes when using ML methods in the medical domain.