Hyperbaric oxygen in the artificial intelligence era: integration and innovation.

Area of Science:

• Biomedical Engineering
• Computational Medicine
• Medical Informatics

Background:

• Hyperbaric oxygen therapy (HBOT) is established for specific conditions but faces limitations in neurological and inflammatory applications due to inconsistent protocols and patient selection.
• The growing interest in HBOT for broader indications necessitates advancements beyond current empirical approaches.
• Artificial intelligence (AI) has demonstrated transformative potential in other clinical fields, suggesting its applicability to HBOT.

Purpose of the Study:

• To explore the integration of AI, including machine learning and digital twins, into hyperbaric oxygen therapy (HBOT) to enable precision medicine.
• To identify and categorize current and emerging applications of AI in HBOT.
• To outline future directions for AI-driven advancements in HBOT efficacy, safety, and mechanistic understanding.

Main Methods:

• A systematic literature search was conducted across PubMed, Scopus, Web of Science, and Google Scholar (January 2000-March 2025).
• Search terms combined "hyperbaric oxygen therapy" with AI-related keywords (e.g., "artificial intelligence", "machine learning", "digital twin").
• Eligible studies were peer-reviewed clinical, preclinical, or computational research linking HBOT with AI; a PRISMA-guided screening process was employed.

Main Results:

• Fifty-three eligible studies, with a significant portion from 2022-2025, highlighted four key AI application areas in HBOT.
• These areas include protocol optimization, biomarker-driven patient selection, real-time adaptive control via biosensors, and predictive safety analytics.
• Exemplars from various medical fields demonstrate AI's feasibility for decision support, federated learning, and digital-twin simulations in HBOT.

Conclusions:

• Integrating AI with HBOT can transition practice from standardized protocols to personalized, data-driven treatments.
• Key priorities for advancing AI in HBOT include hybrid trials, interoperable data registries, explainable AI models, and regulatory validation.
• AI holds the potential to significantly enhance the efficacy, safety, and efficiency of HBOT while deepening the understanding of its underlying mechanisms.