ABSTRACT
Background/Aims@#The accuracy of endosonographers in diagnosing gastric subepithelial lesions (SELs) using endoscopic ultrasonography (EUS) is influenced by experience and subjectivity. Artificial intelligence (AI) has achieved remarkable development in this field. This study aimed to develop an AI-based EUS diagnostic model for the diagnosis of SELs, and evaluated its efficacy with external validation. @*Methods@#We developed the EUS-AI model with ResNeSt50 using EUS images from two hospitals to predict the histopathology of the gastric SELs originating from muscularis propria. The diagnostic performance of the model was also validated using EUS images obtained from four other hospitals. @*Results@#A total of 2,057 images from 367 patients (375 SELs) were chosen to build the models, and 914 images from 106 patients (108 SELs) were chosen for external validation. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the model for differentiating gastrointestinal stromal tumors (GISTs) and non-GISTs in the external validation sets by images were 82.01%, 68.22%, 86.77%, 59.86%, and 78.12%, respectively. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in the external validation set by tumors were 83.75%, 71.43%, 89.33%, 60.61%, and 80.56%, respectively. The EUS-AI model showed better performance (especially specificity) than some endosonographers.The model helped improve the sensitivity, specificity, and accuracy of certain endosonographers. @*Conclusions@#We developed an EUS-AI model to classify gastric SELs originating from muscularis propria into GISTs and non-GISTs with good accuracy. The model may help improve the diagnostic performance of endosonographers. Further work is required to develop a multi-modal EUS-AI system.
ABSTRACT
Inguinal hernia is a defect disease in the abdominal wall. Surgeons have tried various ways to repair the defect for more than 100 years. The traditional herniorrhaphy destroys the normal anatomical structure, and the recurrence rate is quite high. After that, surgeons began to repair the defects with prostheses, from the initial use of rough metal materials such as silver, tantalum, stainless steel, to nylon, fiberglass, silicone rubber and other non-metallic materials, and also from artificial synthetic polymer non-absorbable materials such as polypropylene, polyester, ePTFE, to synthetic absorbable materials such as polyglycolic acid and the acellular extracellular matrix derived from biological meshes. However, these prostheses still can not meet the diverse needs of patients. Thus, multifunctional composite prostheses consisting of two or more materials were born, and various types of composite prostheses, stem cell coating meshes, 3D meshes, microstructure meshes were developed. The repair method evolved from traditional hernia repair to tension-free hernia repair and laparoscopic hernia repair. Surgeons are dedicated to finding idealized meshes for the perfect repair of defects, while considering postoperative complications, patient's quality of life, long-term efficacy and other issues. In the face of a wide variety of repair materials, the choice of surgeons is blind, and there is no standard to determine which prostheses are suitable for patients. Therefore, we have combed the development of different types of prostheses, summarized the development process of hernia repair materials for the past 100 years, and put forward the prospects for future development of prostheses, in order to provide reference for the selection of prostheses.