Related Experiment Videos
Cross-Domain Diagnostic Performance of Analytical and Artificial Intelligence-Based Methods: A Systematic Review With
Snehal Himmatlal Shah1, Niraj Kumar Yadav2, Anjali Virani3
1Department of Ophthalmology, Dr. N.D. Desai Faculty of Medical Science and Research, Dharmsinh Desai University, Nadiad, IND.
Cureus
|May 25, 2026
Summary
Data-driven and artificial intelligence methods show high diagnostic accuracy, especially in ophthalmic imaging. Their clinical utility depends on aligning analytical approaches with specific healthcare goals and implementation contexts.
Area of Science:
- Biomedical Data Science
- Artificial Intelligence in Medicine
- Ophthalmic Imaging Analysis
Background:
- Data-driven and AI methods are increasingly used for medical diagnosis.
- Evaluating their performance across diverse biomedical fields is crucial for clinical adoption.
- Ophthalmic imaging is a key area for AI application.
Purpose of the Study:
- To systematically review the diagnostic performance and clinical applicability of data-driven and AI methods.
- To compare performance across various biomedical domains, focusing on ophthalmic imaging.
- To understand how methodological characteristics impact diagnostic accuracy and translational utility.
Main Methods:
- Systematic literature search adhering to PRISMA guidelines (studies from 2015 onwards).
- Inclusion of original research with primary data, clear methodology, and measurable outcomes.
- Exclusion of studies lacking detail or non-primary research.
Main Results:
- Eleven studies were included, covering ophthalmic imaging, biomarkers, cognitive assessment, and vascular disease.
- High diagnostic performance (sensitivity/specificity >90%) observed in imaging-based screening.
- Moderate, clinically meaningful performance reported for biomarker and cognitive domains.
Conclusions:
- Analytical approaches offer greatest clinical utility when aligned with healthcare objectives and context.
- Integration of performance metrics and translational relevance aids clinical adoption.
- A structured framework supports future development of AI diagnostic tools.