Application of machine learning and deep learning in the diagnosis and treatment of inguinal hernia: a narrative review

With the rapid development of artificial intelligence (AI) technology, its application in the diagnosis and treatment of inguinal hernia (IH) has gradually become a research hotspot. As core components of AI, machine learning (ML) and deep learning (DL) demonstrate tremendous potential in medical imaging, disease prediction, and personalized treatment planning. Currently, models developed using ML can effectively predict the risks of postoperative surgical site infection, surgical site occurrence, intestinal resection in incarcerated IH, and postoperative lower extremity venous thromboembolism. DL, as a subset of ML, excels in processing unstructured data such as images and videos. It utilizes deep neural networks to automatically extract data features, thereby enhancing medical image diagnosis and intraoperative navigation capabilities. Studies have shown that DL is highly effective in identifying anatomical landmarks during surgery, which facilitates real-time feedback and surgical training. Generative AI, built on ML theories, shows promise in medical consultations, but its accuracy and reliability require further validation. Overall, ML and DL are revolutionizing the management of IH by improving diagnostic accuracy, optimizing surgical protocols, and enhancing patient outcomes. Future prospects include data integration, real-time feedback, and interdisciplinary collaboration. This article provides a review of the applications of ML and DL in the diagnosis and treatment of IH, offering references for clinical practice and technological innovation.