The Applications of Machine Learning to Prognostic Modeling for Human Immunodeficiency Virus: A Protocol for a Scoping Review.

Background and Aims: As of late 2023, an estimated 39.9 million people are living with HIV, placing strain on healthcare systems. Machine learning (ML), a branch of artificial intelligence, enables systems to improve performance through data-driven learning without explicit programming. HIV prognosis is influenced by clinical, epidemiological, and psychosocial factors, and ML algorithms have the potential to integrate these determinants efficiently. This can provide valuable insights into disease progression and risk assessment in terms of viral load, CD4 cell count, treatment initiation, treatment adherence, hospitalization, acquired immunodeficiency syndrome diagnosis, quality of life and mental health. This protocol outlines the existing applications of ML to prognostic modeling in the context of HIV, highlighting how ML can equip physicians with rapid and accurate predictions of disease progression, thereby informing treatment decisions such as clinical prescriptions and social support plans, and optimizing patient outcomes. Methods: The protocol follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) framework. A search strategy has been developed for Medline (PubMed) and will be adapted for searches in Embase, Web of Science, Scopus, IEEE Xplore, and ACM Digital Library. The study selection and data extraction will be conducted in duplicate. The methods for the scoping review are prespecified to ensure transparency. Discussion: The proposed scoping review will identify effective model types, data inputs, and applications of ML in the context of HIV prognosis. While ML has been integrated into various aspects of HIV research, few studies have focused on predicting prognosis. This review aims to synthesize current uses of ML in prognostic modeling and highlight gaps within the existing technology. The findings from this review will support the development of future ML models that can inform clinical decision-making, and, in turn, optimize patient care, improve resource allocation, and enhance public health responses to the ongoing HIV epidemic.