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.
Pancreatic ductal adenocarcinoma (PDAC) is the deadliest solid malignancy worldwide, and causes approximately 466,000 deaths per year1. Despite the poor prognosis of PDAC, its early or incidental detection has been shown to substantially improve patient survival2,3,4,5,6,7. Recent studies indicate that high-risk individuals with screen-detected PDAC have a median overall survival of 9.8 years, substantially longer than the 1.5 years for those diagnosed outside of surveillance (for example, via standard clinical diagnostic techniques)6. As such, screening of PDAC holds the greatest promise to reduce PDAC-related mortality8. However, due to the relatively low prevalence of PDAC, effective screening in the general population requires high sensitivity and exceptionally high specificity to mitigate the risk of over-diagnosis. Current screening techniques are limited in this regard, and thus cannot be implemented in the general population as urgently needed9,10.