Benefit of Linked‐Color Imaging in Artificial Intelligence‐Assisted Diagnosis of Early Gastric Cancer: A Pilot Study With Propensity Score Adjustment
DEN Open · 2027-04 · 7 (1)
DOI: 10.1002/deo2.70381 ↗Kurz gefasst
Fachartikel in der Zeitschrift DEN Open, online erschienen am 28.07.2026. Verfasst von Ahmed Nashaat Mohamed und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medicine and Bioregulatory Science, Graduate School of Medical Sciences Kyushu University Fukuoka Japan. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background and Aims Artificial intelligence (AI)‐assisted endoscopy represents a promising approach for lesion detection, yet frequent false‐positive detections impair clinical utility by disrupting examinations and diminishing physician confidence. Linked‐color imaging (LCI), an image‐enhanced endoscopy technique that amplifies mucosal and vascular contrast, may address this limitation. This investigation evaluated whether LCI reduces false‐positive AI detections compared with white‐light imaging (WLI).
Methods
This retrospective study analyzed consecutive AI‐assisted upper endoscopies performed between March 2024 and June 2025. WLI and LCI were performed sequentially within the same endoscopic session in each patient. False‐positive AI detections were compared between modalities using two computer‐aided detection (CAD) versions. Propensity score adjustment was used as a sensitivity analysis for baseline differences between CAD Versions I and II.
Results
Of 66 initially screened cases, 63 remained after excluding patients with prior gastric surgery. LCI reduced false‐positive AI detections compared with WLI (median 2 vs. 5; p < 0.001). In CAD version–stratified sensitivity analyses, LCI reduced false‐positive AI detections in both Version I (5 to 2; p = 0.01) and Version II (2 to 0; p = 0.03). This reduction remained consistent across atrophic grades. Both imaging modalities identified all gastric lesions, achieving 100% detection sensitivity.
Conclusions
LCI assessment performed after WLI observation yielded fewer false‐positive CAD‐EYE detections while maintaining lesion detection sensitivity. However, because the observation sequence was fixed, these findings should be interpreted cautiously and require confirmation in prospective or counterbalanced studies. Trial Registration : N/A (retrospective study).
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Medicine and Bioregulatory Science, Graduate School of Medical Sciences Kyushu University Fukuoka Japan.
- International Medical Department Kyushu University Hospital Fukuoka Japan.
- The National Hepatology and Tropical Medicine Research Institute Cairo Egypt.
- Department of Endoscopic Diagnostics and Therapeutics Kyushu University Hospital Fukuoka Japan.
Zitieren
APA
Mohamed, A. N., Minoda, Y., Maruoka, R., Suzuki, Y., Fukuya, H., Tsuru, H., Wada, M., Tanaka, Y., Chinen, T., Moriyama, T., Ogino, H., Ihara, E., & Ogawa, Y. (2027). Benefit of Linked‐Color Imaging in Artificial Intelligence‐Assisted Diagnosis of Early Gastric Cancer: A Pilot Study With Propensity Score Adjustment. DEN Open, 7(1). https://doi.org/10.1002/deo2.70381
Vancouver
Mohamed AN, Minoda Y, Maruoka R, Suzuki Y, Fukuya H, Tsuru H, et al. Benefit of Linked‐Color Imaging in Artificial Intelligence‐Assisted Diagnosis of Early Gastric Cancer: A Pilot Study With Propensity Score Adjustment. DEN Open. 2027;7(1). doi:10.1002/deo2.70381
BibTeX
@article{Mohamed2027,
author = {Mohamed, Ahmed Nashaat and Minoda, Yosuke and Maruoka, Ryohei and Suzuki, Yusuke and Fukuya, Hiroki and Tsuru, Hirotaka and Wada, Masafumi and Tanaka, Yoshimasa and Chinen, Takatoshi and Moriyama, Tomohiko and Ogino, Haruei and Ihara, Eikichi and Ogawa, Yoshihiro},
title = {Benefit of Linked‐Color Imaging in Artificial Intelligence‐Assisted Diagnosis of Early Gastric Cancer: A Pilot Study With Propensity Score Adjustment},
journal = {DEN Open},
year = {2027},
volume = {7},
number = {1},
doi = {10.1002/deo2.70381}
}
EUVIMED