Advanced Imaging Modalities in Gastrointestinal Endoscopy: A Systematic Review of Diagnostic Accuracy and Clinical...

Conventional white light endoscopy (WLE) has limited sensitivity for detecting subtle gastrointestinal lesions, leading to missed diagnoses and interval cancers. Advanced imaging modalities, including narrow band imaging (NBI), linked color imaging (LCI), blue laser imaging (BLI), autofluorescence imaging (AFI), and artificial intelligence (AI)-based systems, have been developed to enhance lesion detection, characterization, and diagnostic confidence. This systematic review, conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, evaluated the diagnostic accuracy and clinical impact of these technologies compared with WLE and other standard approaches. Comprehensive searches of PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), Scopus, and Web of Science (from inception to September 2025) identified 14,288 records, of which 10,590 unique studies were screened, and 21 met the inclusion criteria. NBI demonstrated moderate-to-high diagnostic accuracy but offered limited improvement over high-definition WLE in adenoma detection. LCI improved sensitivity and lesion visibility across gastric, esophageal, and inflammatory disorders, showing strong performance in early gastric cancer and reflux esophagitis. BLI and BLI-bright enhanced real-time gastric cancer detection and supported colorectal optical diagnosis comparable to NBI. AFI showed inferior performance to dye-based chromoendoscopy in ulcerative colitis surveillance but effectively distinguished reflux phenotypes with high accuracy. AI-based systems consistently increased adenoma detection rate (ADR) and adenomas per colonoscopy (APC) through computer-aided detection (CADe) and achieved clinically actionable accuracy for optical diagnosis in selected settings using computer-aided diagnosis (CADx). Overall, advanced imaging modalities improve lesion detection and diagnostic precision compared with WLE, each offering distinct advantages. LCI and BLI are most effective for early gastric cancer and inflammatory conditions, NBI remains a validated tool for targeted diagnosis, and AI-based systems provide the greatest gains in adenoma yield while showing promise for reliable real-time optical diagnosis. Standardization, multicenter validation, and integration into clinical practice guidelines are essential to optimize their impact on patient outcomes.