Journal Article

Large-scale pancreatic cancer detection via non-contrast CT and deep learning

Yao, Jiawen, Guo, Wenchao, Zhou, Jian, Shao, Chengwei, Yuille, Alan L., Zhang, Qi, Lu, Le, Fang, Wei, Xia, Yingda, Li, Xuezhou, Hager, Gregory D., Chen, Yingli, Nogues, Isabella, Lu, Yimei, Zhang, Rong, Jin, Gang, Zhang, Ling, Hou, Yang, Liang, Tingbo, Wang, Yu, Lu, Jianping, Fang, Xu, Jiang, Hui, Qiu, Mingyan, Zhang, Tingting, Xie, Chuanmiao, Han, Xu, Liu, Zaiyi, Lu, Hong, Vocka, Michal, Chen, Xin, Tang, Wei, Lambert, Lukas, Shi, Yu, Cao, Kai, Kovarnik, Tomas

2023

Overview

Pancreatic ductal adenocarcinoma (PDAC), the most deadly solid malignancy, is typically detected late and at an inoperable stage. Early or incidental detection is associated with prolonged survival, but screening asymptomatic individuals for PDAC using a single test remains unfeasible due to the low prevalence and potential harms of false positives. Non-contrast computed tomography (CT), routinely performed for clinical indications, offers the potential for large-scale screening, however, identification of PDAC using non-contrast CT has long been considered impossible. Here, we develop a deep learning approach, pancreatic cancer detection with artificial intelligence (PANDA), that can detect and classify pancreatic lesions with high accuracy via non-contrast CT. PANDA is trained on a dataset of 3,208 patients from a single center. PANDA achieves an area under the receiver operating characteristic curve (AUC) of 0.986–0.996 for lesion detection in a multicenter validation involving 6,239 patients across 10 centers, outperforms the mean radiologist performance by 34.1% in sensitivity and 6.3% in specificity for PDAC identification, and achieves a sensitivity of 92.9% and specificity of 99.9% for lesion detection in a real-world multi-scenario validation consisting of 20,530 consecutive patients. Notably, PANDA utilized with non-contrast CT shows non-inferiority to radiology reports (using contrast-enhanced CT) in the differentiation of common pancreatic lesion subtypes. PANDA could potentially serve as a new tool for large-scale pancreatic cancer screening. A deep learning model provides high accuracy in detecting pancreatic lesions in multicenter data, outperforming radiology specialists.

• Publisher: Nature Publishing Group US,Nature Publishing Group
• Subject: 631/114/1305 / 692/699/67/1504/1713 / 692/699/67/2322 / 692/700/1421/1846/2771 / Adenocarcinoma / Artificial Intelligence / Biomedical and Life Sciences / Biomedicine / Cancer / Cancer Research / Cancer screening / Carcinoma, Pancreatic Ductal - diagnostic imaging / Carcinoma, Pancreatic Ductal - pathology / Computed tomography / Deep Learning / Humans / Infectious Diseases / Lesions / Malignancy / Medical screening / Metabolic Diseases / Model accuracy / Molecular Medicine / Neurosciences / Pancreas - diagnostic imaging / Pancreas - pathology / Pancreatic cancer / Pancreatic Neoplasms - diagnostic imaging / Pancreatic Neoplasms - pathology / Radiology / Retrospective Studies / Sensitivity / Tomography, X-Ray Computed