Mustafa Alper Akay, Ozan Can Tatar, Semih Metin, Elif Tatar, Onursal Varlıklı, Gülşen Ekingen Yıldız
{"title":"Development and clinical assessment of a novel AI-based diagnostic model for Hirschsprung's disease.","authors":"Mustafa Alper Akay, Ozan Can Tatar, Semih Metin, Elif Tatar, Onursal Varlıklı, Gülşen Ekingen Yıldız","doi":"10.1007/s13304-025-02284-0","DOIUrl":null,"url":null,"abstract":"<p><p>This study aimed to develop an AI-based diagnostic model for Hirschsprung's disease (HD) using deep learning on contrast enema (CE) images, with the goal of improving diagnostic accuracy while reducing invasiveness. The dataset included 725 CE images from histopathologically confirmed HD patients from 2013 to 2022. Employing Python and PyTorch, a deep learning model based on the YOLOv8 algorithm was trained and validated, emphasizing key metrics like mean average precision (mAP), precision, recall, and F1 score. This model exhibited high precision (0.87477) and recall (0.87317), with an mAP50 score of 0.91. External validation showed promising results, including a sensitivity of 86.96%, a specificity of 72.22%, and an overall accuracy of 80.49%. This AI model offers a less-invasive and accurate alternative to traditional HD diagnostics, especially beneficial for initial screening in pediatric gastroenterology, with the potential to enhance healthcare diagnostics through AI integration.</p>","PeriodicalId":23391,"journal":{"name":"Updates in Surgery","volume":" ","pages":""},"PeriodicalIF":2.2000,"publicationDate":"2025-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Updates in Surgery","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s13304-025-02284-0","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"SURGERY","Score":null,"Total":0}
引用次数: 0
Abstract
This study aimed to develop an AI-based diagnostic model for Hirschsprung's disease (HD) using deep learning on contrast enema (CE) images, with the goal of improving diagnostic accuracy while reducing invasiveness. The dataset included 725 CE images from histopathologically confirmed HD patients from 2013 to 2022. Employing Python and PyTorch, a deep learning model based on the YOLOv8 algorithm was trained and validated, emphasizing key metrics like mean average precision (mAP), precision, recall, and F1 score. This model exhibited high precision (0.87477) and recall (0.87317), with an mAP50 score of 0.91. External validation showed promising results, including a sensitivity of 86.96%, a specificity of 72.22%, and an overall accuracy of 80.49%. This AI model offers a less-invasive and accurate alternative to traditional HD diagnostics, especially beneficial for initial screening in pediatric gastroenterology, with the potential to enhance healthcare diagnostics through AI integration.
本研究旨在开发一种基于人工智能的巨结肠病（HD）诊断模型，利用对比灌肠（CE）图像的深度学习，以提高诊断准确性，同时减少侵入性。该数据集包括2013年至2022年组织病理学证实的HD患者的725张CE图像。采用Python和PyTorch，对基于YOLOv8算法的深度学习模型进行了训练和验证，强调了mean average precision （mAP）、precision、recall和F1 score等关键指标。该模型具有较高的准确率（0.87477）和召回率（0.87317），mAP50得分为0.91。外部验证结果令人满意，灵敏度为86.96%，特异性为72.22%，总体准确度为80.49%。这种人工智能模型为传统的HD诊断提供了一种侵入性更小、更准确的替代方案，尤其有利于儿科胃肠病学的初步筛查，并有可能通过人工智能集成来增强医疗保健诊断。
期刊介绍：
Updates in Surgery (UPIS) has been founded in 2010 as the official journal of the Italian Society of Surgery. It’s an international, English-language, peer-reviewed journal dedicated to the surgical sciences. Its main goal is to offer a valuable update on the most recent developments of those surgical techniques that are rapidly evolving, forcing the community of surgeons to a rigorous debate and a continuous refinement of standards of care. In this respect position papers on the mostly debated surgical approaches and accreditation criteria have been published and are welcome for the future.
Beside its focus on general surgery, the journal draws particular attention to cutting edge topics and emerging surgical fields that are publishing in monothematic issues guest edited by well-known experts.
Updates in Surgery has been considering various types of papers: editorials, comprehensive reviews, original studies and technical notes related to specific surgical procedures and techniques on liver, colorectal, gastric, pancreatic, robotic and bariatric surgery.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
