Artificial Intelligence for Risk Prediction of Anemia Disease: Bioinformatics Scoping Review

Pena G.; Cornejo J.; Rodriguez S. · ISMSIT 2025 - 9th International Symposium on Multidisciplinary Studies and Innovative Technologies, Proceedings (2025) · DOI: 10.1109/ismsit67332.2025.11268265

Resumen

Anemia and other hematological disorders represent a major global health burden, affecting over 1.6 billion people, particularly women and children. These conditions are often underdiagnosed due to limited access to diagnostic resources and medical infrastructure. In recent times, Artificial Intelligence (AI) has become a valuable tool for the prediction and identification of such disorders using structured clinical data, hematological parameters, and medical imaging through techniques like Deep Learning and Machine Learning. This review systematically analyzes studies published between 2019 and 2025 that applied AI methods to predict or classify anemia and associated disorders. Research articles in English were retrieved from databases such as Scopus, PubMed, and ScienceDirect, with 25 relevant studies selected based on defined inclusion criteria using the PICOC framework. Findings show that AI models often achieve high performance, with accuracy frequently exceeding 80 percent, and strong sensitivity and specificity. Supervised learning approaches were the most applied and consistently effective. Despite these promising results, challenges remain in clinical validation, data variability, and model interpretability. This review highlights the growing role of AI in medical diagnostics and underscores the necessity for strong, transparent, and widely models for reliable use in the clinical environment.

Instituciones: Universidad Tecnológica del Perú