Journal DOI 10.68956/journal.human-performance-futures

Creative Commons source reading · Machine learning for performance

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

Liu Y; Xie TH; Fu Y; Ha SN; Jin XS; Ren XX

Originally published in Front Med (Lausanne) · 2026-04-01

From the article

Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.

Abstract

AbstractWith 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…

Conclusion

Significant progress has been made in the application of ML and DL to the diagnosis and treatment of IHs, providing robust support for early disease detection, risk assessment, prediction and management of postoperative complications, as well as optimization of the surgical process. In the future, with the continuous development of data integration and standardization, multi - modal data fusion, real - time feedback, rigorous external validation and dynamic adjustment, as well as interdisciplinary collaboration and training, the application prospects of ML and DL in the diagnosis and treatment of IHs will be even more promising.

Editorial reading guide

  • What data created and tested the model, and was evaluation independent?
  • What is the cost of a false positive, false negative or out-of-distribution decision?