From Genomic Data to Predictive Models: A Systematic Review of Machine Learning for Antimicrobial Resistance Prediction

Africa and globally, antimicrobial resistance (AMR) is a significant health problem and genomic data together with machine learning approaches is proving to be a tool which has potential to predict resistance phenotypes. The aim of this systematic review is to thoroughly investigate the use of machine learning (ML) and deep learning (DL) techniques in the prediction of antimicrobial resistance (AMR) using genomic data. The review will aim at summarizing evidence regarding the genomic data sets used, bacterial species, antimicrobial agents, feature extraction and selection methods, ML algorithms, validation methods, performance evaluation metrics, methodologies for model interpretability and challenges with model development and clinical application. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines will be followed in the review, while the review protocol will be registered in advance in the Open Science Framework (OSF) to ensure transparency and reproducibility.