One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yu, Jia, Zhu, Yan, Fu, Peiyao, Chen, Tianyi, Wang, Zhihua, Wu, Fei, Li, Quanlin, Zhou, Pinghong, Wang, Shuo, Yang, Xian
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911347081478144
author Yu, Jia
Zhu, Yan
Fu, Peiyao
Chen, Tianyi
Wang, Zhihua
Wu, Fei
Li, Quanlin
Zhou, Pinghong
Wang, Shuo
Yang, Xian
author_facet Yu, Jia
Zhu, Yan
Fu, Peiyao
Chen, Tianyi
Wang, Zhihua
Wu, Fei
Li, Quanlin
Zhou, Pinghong
Wang, Shuo
Yang, Xian
contents Rare gastrointestinal lesions are infrequently encountered in routine endoscopy, restricting the data available for developing reliable artificial intelligence (AI) models and training novice clinicians. Here we present EndoRare, a one-shot, retraining-free generative framework that synthesizes diverse, high-fidelity lesion exemplars from a single reference image. By leveraging language-guided concept disentanglement, EndoRare separates pathognomonic lesion features from non-diagnostic attributes, encoding the former into a learnable prototype embedding while varying the latter to ensure diversity. We validated the framework across four rare pathologies (calcifying fibrous tumor, juvenile polyposis syndrome, familial adenomatous polyposis, and Peutz-Jeghers syndrome). Synthetic images were judged clinically plausible by experts and, when used for data augmentation, significantly enhanced downstream AI classifiers, improving the true positive rate at low false-positive rates. Crucially, a blinded reader study demonstrated that novice endoscopists exposed to EndoRare-generated cases achieved a 0.400 increase in recall and a 0.267 increase in precision. These results establish a practical, data-efficient pathway to bridge the rare-disease gap in both computer-aided diagnostics and clinical education.
format Preprint
id arxiv_https___arxiv_org_abs_2512_24278
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training
Yu, Jia
Zhu, Yan
Fu, Peiyao
Chen, Tianyi
Wang, Zhihua
Wu, Fei
Li, Quanlin
Zhou, Pinghong
Wang, Shuo
Yang, Xian
Computer Vision and Pattern Recognition
Artificial Intelligence
Rare gastrointestinal lesions are infrequently encountered in routine endoscopy, restricting the data available for developing reliable artificial intelligence (AI) models and training novice clinicians. Here we present EndoRare, a one-shot, retraining-free generative framework that synthesizes diverse, high-fidelity lesion exemplars from a single reference image. By leveraging language-guided concept disentanglement, EndoRare separates pathognomonic lesion features from non-diagnostic attributes, encoding the former into a learnable prototype embedding while varying the latter to ensure diversity. We validated the framework across four rare pathologies (calcifying fibrous tumor, juvenile polyposis syndrome, familial adenomatous polyposis, and Peutz-Jeghers syndrome). Synthetic images were judged clinically plausible by experts and, when used for data augmentation, significantly enhanced downstream AI classifiers, improving the true positive rate at low false-positive rates. Crucially, a blinded reader study demonstrated that novice endoscopists exposed to EndoRare-generated cases achieved a 0.400 increase in recall and a 0.267 increase in precision. These results establish a practical, data-efficient pathway to bridge the rare-disease gap in both computer-aided diagnostics and clinical education.
title One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2512.24278