Artificial intelligence in HIV research: a structured review and task-oriented clinical framework

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Clinical Decision Support Systems

Background:

• Human Immunodeficiency Virus (HIV) remains a global health concern, despite advances in Antiretroviral Treatment (ART).
• The increasing volume of diverse data in HIV management has spurred the adoption of Artificial Intelligence (AI).
• AI offers potential for developing decision-making tools to support healthcare professionals in HIV care.

Purpose of the Study:

• To propose a task-oriented framework for applying AI in HIV management.
• To connect specific clinical tasks within HIV care with suitable AI methodologies.
• To guide clinicians and researchers in leveraging AI for improved patient outcomes.

Main Methods:

• Conducted a structured review of over 50 studies from 2017 to November 2025.
• Analyzed literature based on clinical application scope, data types, data sources, and AI techniques (statistical models, neural networks, natural language processing).
• Developed a framework associating HIV clinical tasks with appropriate AI approaches.

Main Results:

• The study presents a framework that maps HIV clinical tasks to suitable AI approaches.
• Recommendations for algorithms and techniques are provided for various HIV management scenarios.
• The framework emphasizes AI's role in different disease phases, considering data availability.

Conclusions:

• Artificial Intelligence is a crucial tool for advancing HIV management strategies.
• The proposed framework offers a structured foundation for future AI research in HIV.
• Continuous updates to the framework are essential to adapt to evolving medical challenges and AI advancements.