How AI is Transforming Diagnosis in Precision Health

By Dr Mahsa Sheikh, Head of Research at REVIV

AI-supported diagnosis is one of the most exciting advances in modern healthcare. These systems do not replace doctors; they work alongside them, helping to interpret patient information, reduce missed diagnoses, and improve accuracy. By processing large amounts of data, from clinical notes and lab results to imaging and past history, AI can detect subtle patterns that might otherwise go unnoticed, providing a valuable second opinion that complements clinical expertise.

The idea is simple but powerful. Human clinicians bring years of training, deep contextual judgment, and the ability to connect with patients. Artificial intelligence, on the other hand, brings the capacity to instantly review vast datasets, draw comparisons across millions of cases, and flag rare or unexpected patterns. When combined, this partnership has the potential to reduce diagnostic errors, speed up decision-making, and give every patient a better chance of receiving the right treatment at the right time.

A particularly transformative development comes from Large Language Models (LLMs), advanced AI systems trained on extensive medical knowledge and real-world clinical cases. Unlike traditional diagnostic tools, LLMs can synthesise information from across a patient’s health record, symptoms, test results, imaging, and history, to create a more complete clinical picture. They can then suggest possible diagnoses, sometimes surfacing rare conditions or unusual presentations that may otherwise be missed. Some models, such as AMIE and MedFound, have been designed for safe, clinically grounded reasoning and have demonstrated improved diagnostic completeness and accuracy in controlled studies.

Evidence from clinical evaluations is encouraging. McDuff et al. (2025) found that adding an LLM to the diagnostic process improved the accuracy and completeness of differential diagnoses compared with standard resources. Wu et al. (2025) reported that in intensive care simulations, LLMs helped junior doctors reach correct top diagnoses more quickly. Studies in neurology and neurosurgery have shown that these systems can prompt clinicians to consider both common and rare conditions, adding value as a diagnostic support tool. Beyond hospitals, LLMs are being adapted for mental health, where they can analyse speech, writing, and wearable data for early detection of depression or cognitive decline. They are also finding applications in wellness and nutrition, where they provide personalised lifestyle guidance on diet, sleep, and stress.

At REVIV, our Life Sciences team is exploring how these technologies can strengthen our mission in precision health. In research, AI can help us analyse complex datasets from biomarkers, wearables, and clinical records, revealing subtle patterns that explain why some individuals respond more strongly to IV therapies than others. This insight can guide the development of more personalised and effective interventions. In our clinics, AI could support practitioners by broadening the diagnostic picture, suggesting nutritional or lifestyle contributors, and acting as a real-time safety net for clients with overlapping concerns such as fatigue, stress, or hydration imbalance.

Looking ahead, these tools could enable highly personalised wellness pathways. By combining genetic insights with continuous wearable data and clinical biomarkers, AI could help tailor IV formulas, nutrient strategies, and monitoring plans that adapt as a client’s physiology changes. The same models could also inform the development of future IV therapies by identifying which nutrient combinations or approaches deliver the greatest impact across diverse populations.

McDuff D, Schaekermann M, Tu T, et al. Towards Accurate Differential Diagnosis With Large Language Models. Nature. 2025;642(8067):451-457. doi:10.1038/s41586-025-08869-4.

Tu T, Schaekermann M, Palepu A, et al. Towards Conversational Diagnostic Artificial Intelligence. Nature. 2025;642(8067):442-450. doi:10.1038/s41586-025-08866-7.

Wu X, Huang Y, He Q. A Large Language Model Improves Clinicians’ Diagnostic Performance in Complex Critical Illness Cases. Critical Care. 2025;29(1):230. doi:10.1186/s13054-025-05468-7.

Kumar RP, Sivan V, Bachir H, et al. Can Artificial Intelligence Mitigate Missed Diagnoses by Generating Differential Diagnoses for Neurosurgeons? World Neurosurgery. 2024;187:e1083-e1088. doi:10.1016/j.wneu.2024.05.052.

Fonseca Â, Ferreira A, Ribeiro L, et al. Embracing the Future — Is Artificial Intelligence Already Better? Eur J Neurol. 2024;31(4):e16195. doi:10.1111/ene.16195.

Naeem A, Khan O, Baqir SM, et al. Language Artificial Intelligence Models as Pioneers in Diagnostic Medicine? J Clin Med. 2025;14(4):1131. doi:10.3390/jcm14041131.

Feldman MJ, Hoffer EP, Conley JJ, et al. Dedicated AI Expert System vs Generative AI With Large Language Model for Clinical Diagnoses. JAMA Netw Open. 2025;8(5):e2512994. doi:10.1001/jamanetworkopen.2025.12994.

Zöller N, Berger J, Lin I, et al. Human–AI Collectives Most Accurately Diagnose Clinical Vignettes. Proc Natl Acad Sci USA. 2025;122(24):e2426153122. doi:10.1073/pnas.2426153122.

Cruz-Gonzalez P, He AW, Lam EP, et al. Artificial Intelligence in Mental Health Care: A Systematic Review. Psychol Med. 2025;55:e18. doi:10.1017/S0033291724003295.

Abd-Alrazaq A, Alhuwail D, Schneider J, et al. The Performance of Artificial Intelligence-Driven Technologies in Diagnosing Mental Disorders: An Umbrella Review. NPJ Digit Med. 2022;5(1):87. doi:10.1038/s41746-022-00631-8