Overfitting in Artificial Intelligence Applications in Healthcare: A Scoping Review

Read full text Summarise with VIRAT
Authors: Ana Carolina Dos Santos, Isabelle Cristinne Pinto Costa, Yasmim Ribeiro Fracaroli, Sara Luiza Custódio de Souza, Francini Castilha do Nascimento
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Artificial intelligence (AI) has become a central tool in transforming healthcare systems, with applications ranging from image-based diagnosis to precision medicine. However, technical challenges persist, including overfitting—a phenomenon in which models learn patterns that are too specific to the training data, resulting in poor generalization and potentially compromising safety and effectiveness in real clinical settings. Given the relevance of this problem and its potential impact on diagnostic, prognostic, and therapeutic decision-making, this scoping review aims to map the existing literature on overfitting in AI applications within healthcare, identifying its causes, consequences, and reported mitigation strategies. The review will follow the methodology of the Joanna Briggs Institute (JBI) and will be reported according to the PRISMA-ScR extension. Evidence selection will be guided by the PCC framework (Population, Concept, Context), and the data will be analyzed descriptively and thematically, enabling the synthesis of model types, areas of application, manifestations of overfitting, and prevention methods used.

Fields of Study & Subjects

Artificial Intelligence in Healthcare and Education Artificial Intelligence in Healthcare Machine Learning in Healthcare

Publication details

Journal / source
OSF Preprints (OSF Preprints)
Publication date
12 August 2026
Language
EN
Document type
Preprint