A clinically validated 3D deep learning approach for quantifying vascular invasion in pancreatic cancer

Yajiao Zhang; Haoran Zhang; Yanzhao Yang; Chao Wu; Lei Zhang; Wei Xia; Xue Wang; Xiaohuan Zhang; Lixiu Cao; Manju Liu; Jing Zhang; Fuhua Yan; Baiyong Shen; Ning Wen

doi:10.1038/s41746-025-02260-3

A clinically validated 3D deep learning approach for quantifying vascular invasion in pancreatic cancer

NPJ Digit Med. 2025 Dec 31;9(1):79. doi: 10.1038/s41746-025-02260-3.

Authors

Yajiao Zhang^#^{1

2}, Haoran Zhang^#^{1

2}, Yanzhao Yang^#¹, Chao Wu³, Lei Zhang⁴, Wei Xia¹, Xue Wang^{1

2}, Xiaohuan Zhang⁴, Lixiu Cao⁵, Manju Liu⁶, Jing Zhang¹, Fuhua Yan^{7

8}, Baiyong Shen⁹, Ning Wen^{10

11

12}

Affiliations

¹ Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
² The SJTU-Ruijin-UIH Institute for Medical Imaging Technology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
³ Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China.
⁴ Department of Radiology, Wuxi No.2 People's Hospital (Jiangnan University Medical Center), Wuxi, China.
⁵ Department of Nuclear Medicine, Tangshan People's Hospital, Tangshan, Hebei Province, China.
⁶ Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu Province, China.
⁷ Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. yfh11655@rjh.com.cn.
⁸ The SJTU-Ruijin-UIH Institute for Medical Imaging Technology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. yfh11655@rjh.com.cn.
⁹ Department of General Surgery, Pancreatic Disease Center, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. shenby@shsmu.edu.cn.
¹⁰ Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. wn12479@rjh.com.cn.
¹¹ The SJTU-Ruijin-UIH Institute for Medical Imaging Technology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. wn12479@rjh.com.cn.
¹² The Global Institute of Future Technology, Shanghai Jiao Tong University, Shanghai, China. wn12479@rjh.com.cn.

^# Contributed equally.

Abstract

Vascular invasion assessment is critical for surgical planning in pancreatic ductal adenocarcinoma (PDAC). Current CT-based assessments often rely on radiologists' subjective 2D interpretations, which may not capture the continuous, three-dimensional tumor-vessel interactions and multiple vessel involvement, both essential for accurate preoperative evaluation. PAN-VIQ (Pancreatic Vascular Invasion Quantifier) is an automated deep learning framework to quantify tumor-vessel interactions from contrast-enhanced CT scans. It enables segmentation of pancreatic tumors and five critical vessels: celiac artery (CA), common hepatic artery (CHA), superior mesenteric artery (SMA), superior mesenteric vein (SMV), and portal vein (PV), quantifying vascular involvement through 3D encasement angles. PAN-VIQ was trained and internally validated on 2130 cases, and subsequently prospectively tested in 202 patients. External validation showed accuracies exceeding 90%. In prospective evaluation, the model outperformed junior radiologists and matched senior radiologists in accuracy and recall. These results underscore potential of PAN-VIQ to standardize vascular invasion assessment and reduce interobserver variability.

Abstract

Grants and funding