Exploring Patient Data Requirements in Training Effective AI Models for MRI-based Breast Cancer Classification

TL;DR

This study investigates the amount of MRI patient data needed for effective breast cancer AI models, finding that a relatively small dataset suffices when using foundation models, with larger datasets offering diminishing returns.

Contribution

It demonstrates that effective breast cancer detection models can be trained with limited data by leveraging foundation models, reducing data requirements for medical institutions.

Findings

Models trained with fewer than 50 patients perform competitively.

Beyond 50 patients, additional data has minimal impact on performance.

Simple ensemble methods further enhance model accuracy.

Abstract

The past decade has witnessed a substantial increase in the number of startups and companies offering AI-based solutions for clinical decision support in medical institutions. However, the critical nature of medical decision-making raises several concerns about relying on external software. Key issues include potential variations in image modalities and the medical devices used to obtain these images, potential legal issues, and adversarial attacks. Fortunately, the open-source nature of machine learning research has made foundation models publicly available and straightforward to use for medical applications. This accessibility allows medical institutions to train their own AI-based models, thereby mitigating the aforementioned concerns. Given this context, an important question arises: how much data do medical institutions need to train effective AI models? In this study, we explore this question in relation to breast cancer detection, a particularly contested area due to the prevalence of this disease, which affects approximately 1 in every 8 women. Through large-scale experiments on various patient sizes in the training set, we show that medical institutions do not need a decade's worth of MRI images to train an AI model that performs competitively with the state-of-the-art, provided the model leverages foundation models. Furthermore, we observe that for patient counts greater than 50, the number of patients in the training set has a negligible impact on the performance of models and that simple ensembles further improve the results without additional complexity.