Pancreatic Cancer Detection on CT Scans with Deep Learning: A Nationwide Population-based Study

Published Online:

An automatic end-to-end deep learning–based detection tool could detect pancreatic cancer on CT scans in a nationwide real-world test data set with 91% accuracy, without requiring manual image labeling or preprocessing.


Approximately 40% of pancreatic tumors smaller than 2 cm are missed at abdominal CT.


To develop and to validate a deep learning (DL)–based tool able to detect pancreatic cancer at CT.

Materials and Methods

Retrospectively collected contrast-enhanced CT studies in patients diagnosed with pancreatic cancer between January 2006 and July 2018 were compared with CT studies of individuals with a normal pancreas (control group) obtained between January 2004 and December 2019. An end-to-end tool comprising a segmentation convolutional neural network (CNN) and a classifier ensembling five CNNs was developed and validated in the internal test set and a nationwide real-world validation set. The sensitivities of the computer-aided detection (CAD) tool and radiologist interpretation were compared using the McNemar test.


A total of 546 patients with pancreatic cancer (mean age, 65 years ± 12 [SD], 297 men) and 733 control subjects were randomly divided into training, validation, and test sets. In the internal test set, the DL tool achieved 89.9% (98 of 109; 95% CI: 82.7, 94.9) sensitivity and 95.9% (141 of 147; 95% CI: 91.3, 98.5) specificity (area under the receiver operating characteristic curve [AUC], 0.96; 95% CI: 0.94, 0.99), without a significant difference (P = .11) in sensitivity compared with the original radiologist report (96.1% [98 of 102]; 95% CI: 90.3, 98.9). In a test set of 1473 real-world CT studies (669 malignant, 804 control) from institutions throughout Taiwan, the DL tool distinguished between CT malignant and control studies with 89.7% (600 of 669; 95% CI: 87.1, 91.9) sensitivity and 92.8% specificity (746 of 804; 95% CI: 90.8, 94.5) (AUC, 0.95; 95% CI: 0.94, 0.96), with 74.7% (68 of 91; 95% CI: 64.5, 83.3) sensitivity for malignancies smaller than 2 cm.


The deep learning–based tool enabled accurate detection of pancreatic cancer on CT scans, with reasonable sensitivity for tumors smaller than 2 cm.

© RSNA, 2022

Online supplemental material is available for this article.

See also the editorial by Aisen and Rodrigues in this issue.


  • 1. Rahib L, Smith BD, Aizenberg R, Rosenzweig AB, Fleshman JM, Matrisian LM. Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the United States. Cancer Res 2014;74(11):2913–2921. Crossref, MedlineGoogle Scholar
  • 2. Chari ST. Detecting early pancreatic cancer: problems and prospects. Semin Oncol 2007;34(4):284–294. Crossref, MedlineGoogle Scholar
  • 3. Kang JD, Clarke SE, Costa AF. Factors associated with missed and misinterpreted cases of pancreatic ductal adenocarcinoma. Eur Radiol 2021;31(4):2422–2432. Crossref, MedlineGoogle Scholar
  • 4. Chen PT, Chang D, Wu T, Wu MS, Wang W, Liao WC. Applications of artificial intelligence in pancreatic and biliary diseases. J Gastroenterol Hepatol 2021;36(2):286–294. Crossref, MedlineGoogle Scholar
  • 5. Liu KL, Wu T, Chen PT, et al. Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation. Lancet Digit Health 2020;2(6):e303–e313. Crossref, MedlineGoogle Scholar
  • 6. Yao X, Song Y, Liu Z. Advances on pancreas segmentation: a review. Multimed Tools Appl 2020;79(9):6799–6821. CrossrefGoogle Scholar
  • 7. National Health Insurance Administration Ministry of Health and Welfare. Published 2016. Accessed May 21, 2021. Google Scholar
  • 8. Wang P, Shen C, Roth HR, et al. Automated Pancreas Segmentation Using Multi-institutional Collaborative Deep Learning. In: Domain Adaptation and Representation Transfer, and Distributed and Collaborative Learning. Cham, Switzerland: Springer International Publishing, 2020; 192–200. CrossrefGoogle Scholar
  • 9. Shen C, Wang P, Roth HR, et al. Multi-task federate learning for heterogeneous pancreas segmentation. In: Clinical Image-Based Procedures, Distributed and Collaborative Learning, Artificial Intelligence for Combating COVID-19 and Secure and Privacy-Preserving Machine Learning. Cham, Switzerland: Springer International Publishing, 2021; 101–110. Google Scholar
  • 10. Yu Q, Yang D, Roth H, et al. C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, June 13–19, 2020. Piscataway, NJ: IEEE, 2020; 4125–4134. Google Scholar
  • 11. Simpson AL, Antonelli M, Bakas S, et al. A large annotated medical image dataset for the development and evaluation of segmentation algorithms. arXiv [preprint] Posted February 25, 2019. Accessed February 27, 2022. Google Scholar
  • 12. Pancreas-CT - The Cancer Imaging Archive (TCIA) Public Access - Cancer Imaging Archive Wiki. Accessed June 14, 2019. Google Scholar
  • 13. Gibson E, Giganti F, Hu Y, et al. Multi-organ Abdominal CT Reference Standard Segmentations. Published February 22, 2018. Accessed February 27, 2022. Google Scholar
  • 14. Deeks JJ, Altman DG. Diagnostic tests 4: likelihood ratios. BMJ 2004;329(7458):168–169. Crossref, MedlineGoogle Scholar
  • 15. Zhou XH. Statistical methods in diagnostic medicine. Hoboken, NJ: Wiley, 2011. CrossrefGoogle Scholar
  • 16. Royston P. PTREND: Stata module for trend analysis for proportions. Boston College Department of Economics. Published 2014. Accessed May 21, 2022. Google Scholar
  • 17. Health Promotion Administration. Published 2016. Accessed May 21, 2022. Google Scholar
  • 18. Tamada T, Ito K, Kanomata N, et al. Pancreatic adenocarcinomas without secondary signs on multiphasic multidetector CT: association with clinical and histopathologic features. Eur Radiol 2016;26(3):646–655. Crossref, MedlineGoogle Scholar
  • 19. Bai W, Sinclair M, Tarroni G, et al. Automated cardiovascular magnetic resonance image analysis with fully convolutional networks. J Cardiovasc Magn Reson 2018;20(1):65. Crossref, MedlineGoogle Scholar
  • 20. Barish M, Bolourani S, Lau LF, Shah S, Zanos TP. External validation demonstrates limited clinical utility of the interpretable mortality prediction model for patients with COVID-19. Nat Mach Intell 2020;3(1):25–27. CrossrefGoogle Scholar
  • 21. Gulshan V, Peng L, Coram M, et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA 2016;316(22):2402–2410. Crossref, MedlineGoogle Scholar
  • 22. Beede E, Baylor E, Hersch F, et al. A Human-Centered Evaluation of a Deep Learning System Deployed in Clinics for the Detection of Diabetic Retinopathy. In: CHI ‘20: Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. New York, NY: Association for Computing Machinery, 2020; 1–12. Google Scholar
  • 23. Kim DW, Jang HY, Kim KW, Shin Y, Park SH. Design Characteristics of Studies Reporting the Performance of Artificial Intelligence Algorithms for Diagnostic Analysis of Medical Images: Results from Recently Published Papers. Korean J Radiol 2019;20(3):405–410. Crossref, MedlineGoogle Scholar
  • 24. Si K, Xue Y, Yu X, et al. Fully end-to-end deep-learning-based diagnosis of pancreatic tumors. Theranostics 2021;11(4):1982–1990. Crossref, MedlineGoogle Scholar
  • 25. Turečková A, Tureček T, Komínková Oplatková Z, Rodríguez-Sánchez A. Improving CT Image Tumor Segmentation Through Deep Supervision and Attentional Gates. Front Robot AI 2020;7:106. Crossref, MedlineGoogle Scholar
  • 26. Zhang Y, Wu J, Liu Y, et al. A deep learning framework for pancreas segmentation with multi-atlas registration and 3D level-set. Med Image Anal 2021;68:101884. Crossref, MedlineGoogle Scholar
  • 27. Antonelli M, Reinke A, Bakas S, et al. The Medical Segmentation Decathlon. arXiv [preprint] Posted June 10, 2021. Accessed February 27, 2022. Google Scholar
  • 28. Dewitt J, Devereaux BM, Lehman GA, Sherman S, Imperiale TF. Comparison of endoscopic ultrasound and computed tomography for the preoperative evaluation of pancreatic cancer: a systematic review. Clin Gastroenterol Hepatol 2006;4(6):717–725;quiz 664. Crossref, MedlineGoogle Scholar
  • 29. Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal 2017;42:60–88. Crossref, MedlineGoogle Scholar
  • 30. Esteva A, Robicquet A, Ramsundar B, et al. A guide to deep learning in healthcare. Nat Med 2019;25(1):24–29. Crossref, MedlineGoogle Scholar

Article History

Received: Mar 5 2022
Revision requested: Apr 22 2022
Revision received: July 15 2022
Accepted: Aug 1 2022
Published online: Sept 13 2022
Published in print: Jan 2023