| Autor: |
Ligato I; Department of Medical-Surgical Sciences and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, 00185 Roma, Italy., De Magistris G; Department of Computer, Control, and Management Engineering, Sapienza University of Rome, Via Ariosto 25, 00185 Rome, Italy., Dilaghi E; Department of Medical-Surgical Sciences and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, 00185 Roma, Italy., Cozza G; Department of Medical-Surgical Sciences and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, 00185 Roma, Italy., Ciardiello A; Department of Physics, Sapienza University of Rome, P.le A. Moro 5, 00185 Rome, Italy., Panzuto F; Department of Medical-Surgical Sciences and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, 00185 Roma, Italy., Giagu S; Department of Physics, Sapienza University of Rome, P.le A. Moro 5, 00185 Rome, Italy., Annibale B; Department of Medical-Surgical Sciences and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, 00185 Roma, Italy., Napoli C; Department of Computer, Control, and Management Engineering, Sapienza University of Rome, Via Ariosto 25, 00185 Rome, Italy., Esposito G; Department of Medical-Surgical Sciences and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, 00185 Roma, Italy. |
| Abstrakt: |
Gastric cancer (GC) is a significant healthcare concern, and the identification of high-risk patients is crucial. Indeed, gastric precancerous conditions present significant diagnostic challenges, particularly early intestinal metaplasia (IM) detection. This study developed a deep learning system to assist in IM detection using image patches from gastric corpus examined using virtual chromoendoscopy in a Western country. Utilizing a retrospective dataset of endoscopic images from Sant'Andrea University Hospital of Rome, collected between January 2020 and December 2023, the system extracted 200 × 200 pixel patches, classifying them with a voting scheme. The specificity and sensitivity on the patch test set were 76% and 72%, respectively. The optimization of a learnable voting scheme on a validation set achieved a specificity of 70% and sensitivity of 100% for entire images. Despite data limitations and the absence of pre-trained models, the system shows promising results for preliminary screening in gastric precancerous condition diagnostics, providing an explainable and robust Artificial Intelligence approach. |
