Linked color imaging for detection and characterization of gastrointestinal neuroendocrine tumors
Year: 2025
Type: article
Study design: Observational Study
Abstract: BACKGROUND AND AIMS: The endoscopic detection of gastrointestinal neuroendocrine tumors (GI-NETs) remains clinically challenging. Linked color imaging (LCI), which enhances color contrast, may improve the identification of tumorous lesions. This study aims to evaluate the endoscopic characteristics of GI-NETs using LCI in comparison with white light imaging (WLI). METHODS: We retrospectively reviewed two pictures both with WLI and LCI for consecutive patients with GI-NETs who underwent endoscopic resection at our center. The endoscopic features and clinic characteristics were analyzed. The edges of lesions were identified using the Canny operator, and color differences (∆E) between lesions and surrounding mucosa in gastric and rectal NETs were measured using the CIE Lab* color space(CIE: Commission Internationale d'Eclairage). Three expert endoscopists and three non-experts independently assessed LCI and WLI images using a validated visibility scores scale. RESULTS: Sixty patients were included in the analysis. The average ∆E values for LCI were significantly higher than those for WLI in both gastric and rectal NETs (P < 0.05). In rectal NETs, the increased contrast was predominantly driven by enhancement of the yellow-blue component (b*) (P < 0.001). LCI provided superior delineation of lesion edges and more detailed visualization compared to WLI. LCI significantly improves the visibility scores of GI-NETs compared to WLI among non-expert endoscopists (P < 0.001). CONCLUSION: LCI significantly improves the endoscopic visibility of GI-NETs by enhancing color contrast between lesions and surrounding mucosa and optimizing edges visualization. These benefits may allow easy recognition of GI-NETs even for non-experts, suggesting that LCI may be a valuable tool for enhancing the endoscopic detection and characterization of GI-NETs.
Source: BMC Gastroenterology
Authors: Xin Pu, Yang Xu, Xueting Zhang, Zijin Yin, Yan Liu, Min Min
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,190
Award: G3297427067
OpenAlex ID: W4415914210
