Kalapala, Rakesh

ABSTRACT In response to the World Health Organization's (WHO) Global Initiative on Artificial Intelligence (AI) for Health, the World Endoscopy Organization (WEO) highlights the unique challenges and opportunities AI presents for gastrointestinal endoscopy, particularly in resource‐limited settings. While AI technologies have shown promise in improving diagnostic accuracy and efficiency in high‐resource environments, their implementation in low‐ and middle‐income countries is hindered by infrastructural, economic, regulatory, and training barriers. This commentary explores how these challenges may exacerbate existing healthcare disparities, emphasizing the need for localized datasets, affordable AI models, simplified regulatory frameworks, and workforce capacity building. The WEO supports WHO's call for equitable AI deployment and advocates for region‐specific solutions, including mobile and offline AI tools, public‐private partnerships, locally developed algorithms aligned with prevalent disease patterns, and a flexibly adapted regulatory framework. By leveraging WEO's training networks and fostering collaboration among governments, clinicians, and industry, the integration of AI into endoscopy can become more accessible and relevant to underserved populations. The commentary underscores that AI should not be seen as a luxury but as a tool to bridge global disparities in care quality. Ensuring responsible and inclusive AI integration requires both global coordination and context‐specific adaptations to truly benefit all healthcare systems.

Abstract Background and Aims Lack of clinical validation and inter-observer variability are two limitations of endoscopic assessment and scoring of disease severity in patients with ulcerative colitis [UC]. We developed a deep learning [DL] model to improve, accelerate and automate UC detection, and predict the Mayo Endoscopic Subscore [MES] and the Ulcerative Colitis Endoscopic Index of Severity [UCEIS]. Methods A total of 134 prospective videos [1550 030 frames] were collected and those with poor quality were excluded. The frames were labelled by experts based on MES and UCEIS scores. The scored frames were used to create a preprocessing pipeline and train multiple convolutional neural networks [CNNs] with proprietary algorithms in order to filter, detect and assess all frames. These frames served as the input for the DL model, with the output being continuous scores for MES and UCEIS [and its components]. A graphical user interface was developed to support both labelling video sections and displaying the predicted disease severity assessment by the artificial intelligence from endoscopic recordings. Results Mean absolute error [MAE] and mean bias were used to evaluate the distance of the continuous model’s predictions from ground truth, and its possible tendency to over/under-predict were excellent for MES and UCEIS. The quadratic weighted kappa used to compare the inter-rater agreement between experts’ labels and the model’s predictions showed strong agreement [0.87, 0.88 at frame-level, 0.88, 0.90 at section-level and 0.90, 0.78 at video-level, for MES and UCEIS, respectively]. Conclusions We present the first fully automated tool that improves the accuracy of the MES and UCEIS, reduces the time between video collection and review, and improves subsequent quality assurance and scoring.