arXiv Open Access 2025

Subspecialty-Specific Foundation Model for Intelligent Gastrointestinal Pathology

Lianghui Zhu Xitong Ling Minxi Ouyang Xiaoping Liu Tian Guan +22 lainnya
Lihat Sumber

Abstrak

Gastrointestinal (GI) diseases represent a clinically significant burden, necessitating precise diagnostic approaches to optimize patient outcomes. Conventional histopathological diagnosis suffers from limited reproducibility and diagnostic variability. To overcome these limitations, we develop Digepath, a specialized foundation model for GI pathology. Our framework introduces a dual-phase iterative optimization strategy combining pretraining with fine-screening, specifically designed to address the detection of sparsely distributed lesion areas in whole-slide images. Digepath is pretrained on over 353 million multi-scale images from 210,043 H&E-stained slides of GI diseases. It attains state-of-the-art performance on 33 out of 34 tasks related to GI pathology, including pathological diagnosis, protein expression status prediction, gene mutation prediction, and prognosis evaluation. We further translate the intelligent screening module for early GI cancer and achieve near-perfect 99.70% sensitivity across nine independent medical institutions. This work not only advances AI-driven precision pathology for GI diseases but also bridge critical gaps in histopathological practice.

Penulis (27)

L

Lianghui Zhu

X

Xitong Ling

M

Minxi Ouyang

X

Xiaoping Liu

T

Tian Guan

M

Mingxi Fu

Z

Zhiqiang Cheng

F

Fanglei Fu

M

Maomao Zeng

L

Liming Liu

S

Song Duan

Q

Qiang Huang

Y

Ying Xiao

J

Jianming Li

S

Shanming Lu

Z

Zhenghua Piao

M

Mingxi Zhu

Y

Yibo Jin

S

Shan Xu

Q

Qiming He

Y

Yizhi Wang

J

Junru Cheng

X

Xuanyu Wang

L

Luxi Xie

H

Houqiang Li

S

Sufang Tian

Y

Yonghong He

Format Sitasi

Zhu, L., Ling, X., Ouyang, M., Liu, X., Guan, T., Fu, M. et al. (2025). Subspecialty-Specific Foundation Model for Intelligent Gastrointestinal Pathology. https://arxiv.org/abs/2505.21928

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2025
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓