Related Experiment Videos
Artificial Intelligence in Clinical Decision-Making: A Systematic Review of Diagnostic Accuracy, Predictive
Natasha Chari1, Ahmed Abdelateef Ahmed Abdelmageed2, Divyank Subhedar3
1Urology, Chelsea and Westminster NHS Foundation Trust, London, GBR.
Cureus
|July 31, 2026
Summary
Artificial intelligence (AI) shows strong diagnostic and predictive accuracy in healthcare, comparable or superior to clinicians in areas like radiology. Further prospective studies are needed before widespread clinical use.
Area of Science:
- Medical Informatics
- Clinical Decision Support Systems
Background:
- Artificial intelligence (AI) is increasingly integrated into clinical decision-making.
- AI aims to enhance diagnostic accuracy, predictive performance, and treatment planning.
Purpose of the Study:
- To systematically review the accuracy and clinical outcomes of AI systems versus standard clinical practice.
- To evaluate AI's performance across various medical specialties.
Main Methods:
- A comprehensive literature search was performed across major databases (PubMed, Embase, Scopus, Cochrane Library).
- Six studies (approx. 1.1 million patients) were included, adhering to PRISMA 2020 guidelines.
- Narrative synthesis and risk of bias assessment (QUADAS-2, ROBINS-I) were conducted due to heterogeneity.
Main Results:
- AI models exhibited strong performance, with AUCs from 0.85 to 0.96, sensitivity up to 97%, and specificity up to 93%.
- AI performance was comparable or superior to clinicians in radiology and dermatology.
- Predictive models in intensive care units (ICUs) showed greater variability.
Conclusions:
- AI demonstrates significant promise for improving diagnostic and predictive accuracy in clinical settings.
- Current evidence largely stems from retrospective studies, necessitating prospective validation.
- Prospective multicenter trials are crucial before routine AI implementation in healthcare.
