Abstract
Purpose
Current diagnostic and treatment modalities for pancreatic cysts (PCs) are invasive and are associated with patient morbidity. The purpose of this study is to develop and evaluate machine learning algorithms to delineate mucinous from non-mucinous PCs using non-invasive CT-based radiomics.
Methods
A retrospective, single-institution analysis of patients with non-pseudocystic PCs, contrast-enhanced computed tomography scans within 1 year of resection, and available surgical pathology were included. A quantitative imaging software platform was used to extract radiomics. An extreme gradient boosting (XGBoost) machine learning algorithm was used to create mucinous classifiers using texture features only, or radiomic/radiologic and clinical combined models. Classifiers were compared using performance scoring metrics. Shapely additive explanation (SHAP) analyses were conducted to identify variables most important in model construction.
Results
Overall, 99 patients and 103 PCs were included in the analyses. Eighty (78%) patients had mucinous PCs on surgical pathology. Using multiple fivefold cross validations, the texture features only and combined XGBoost mucinous classifiers demonstrated an area under the curve of 0.72 ± 0.14 and 0.73 ± 0.14, respectively. By SHAP analysis, root mean square, mean attenuation, and kurtosis were the most predictive features in the texture features only model. Root mean square, cyst location, and mean attenuation were the most predictive features in the combined model.
Conclusion
Machine learning principles can be applied to PC texture features to create a mucinous phenotype classifier. Model performance did not improve with the combined model. However, specific radiomic, radiologic, and clinical features most predictive in our models can be identified using SHAP analysis.
Graphic abstract
Similar content being viewed by others
Data availability
Data was collected with HealthMyne (v5.0, Madison, WI).
Code availability
Git Hub Repository, https://github.com/uw-cmg/MeghanPanccystmachine.
References
Stark A, Donahue TR, Reber HA, Hines OJ (2016) Pancreatic Cyst Disease: A Review. JAMA 315(17):1882–1893. https://doi.org/10.1001/jama.2016.4690
Tanaka M, Fernández-Del Castillo C, Kamisawa T, Jang JY, Levy P, Ohtsuka T, Salvia R, Shimizu Y, Tada M, Wolfgang CL (2017) Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology 17(5):738–753. https://doi.org/10.1016/j.pan.2017.07.007
Vege SS, Ziring B, Jain R, Moayyedi P (2015) Clinical Guidelines Committee; American Gastroenterology Association. American gastroenterological association institute guideline on the diagnosis and management of asymptomatic neoplastic pancreatic cysts. Gastroenterology 148(4):819–822; quize12–3. https://doi.org/10.1053/j.gastro.2015.01.015
Wu J, Wang Y, Li Z, Miao H (2019) Accuracy of Fukuoka and American Gastroenterological Association Guidelines for predicting advanced neoplasia in pancreatic cyst neoplasm: A meta-analysis. Ann Surg Oncol 26(13):4522-4536. https://doi.org/10.1245/s10434-019-07921-8
Cho CS, Russ AJ, Loeffler AG, Rettammel RJ, Oudheusden G, Winslow ER, Weber SM (2013) Preoperative classification of pancreatic cystic neoplasms: the clinical significance of diagnostic inaccuracy. Ann Surg Oncol 20(9):3112-3119. https://doi.org/10.1245/s10434-013-2986-6
Jones MJ, Buchanan AS, Neal CP, Dennison AR, Metcalfe MS, Garcea G (2013) Imaging of indeterminate pancreatic cystic lesions: a systematic review. Pancreatology 13(4):436-442. https://doi.org/10.1016/j.pan.2013.05.007
Rift CV, Scheie D, Toxvaerd A, Kovacevic B, Klausen P, Vilmann P, Hansen CP, Lund EL, Hasselby JP (2021) Diagnostic accuracy of EUS-guided through-the-needle-biopsies and simultaneously obtained fine needle aspiration for cytology from pancreatic cysts: A systematic review and meta-analysis. Pathology-Research and Practice 220(1):153368-153381. https://doi.org/10.1016/j.prp.2021.153368
Zhu H, Jiang F, Zhu J, Du Y, Jin Z, Li Z (2017). Assessment of morbidity and mortality associated with endoscopic ultrasound-guided fine-needle aspiration for pancreatic cystic lesions: A systematic review and meta-analysis. Dig Endosc 29(1):667-675. https://doi.org/10.1111/den.12851
Gillies RJ, Kinahan PE, Hricak H (2016). Radiomics: Images are more than pictures, they are data. Radiology 278(2):563-577. https://doi.org/10.1148/radiol.2015151169
Haralick RM, Shanmugam K, Dinstein I (1973). Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics. SMC-3:610–621.
Espinasse M, Pitre-Champagnat S, Charmettant B, Bidault F, Volk A, Balleyguier C, Lassau N, Caramella C (2020). CT texture analysis challenges: influence of acquisition and reconstruction parameters: A comprehensive review. Diagnostics (Basel, Switzerland) 10(5):258-266. https://doi.org/10.3390/diagnostics10050258
Hanania AN, Bantis LE, Feng Z, Wang H, Tamm EP, Katz MH, Maitra A, Koay EJ (2016). Quantitative imaging to evaluate malignant potential of IPMNs. Oncotarget 7(52):85766–85784. https://doi.org/10.18632/oncotarget.11769
Permuth JB, Choi J, Balarunathan Y, et al. (2016). Combining radiomic features with a miRNA classifier may improve prediction of malignant pathology for pancreatic intraductal papillary mucinous neoplasms. Oncotarget 7(52):85785–85797. https://doi.org/10.18632/oncotarget.11768
Attiyeh MA, Chakraborty J, Gazit L, et al. (2019). Preoperative risk prediction for intraductal papillary mucinous neoplasms by quantitative CT image analysis. HPB 21(2):212-218. https://doi.org/10.1016/j.hpb.2018.07.016
Chakraborty J, Midya A, Gazit L, Attiyeh M, Langdon-Embry L, Allen PJ, Do RKG, Simpson AL (2018). CT radiomics to predict high-risk intraductal papillary mucinous neoplasms of the pancreas. Med Phys 45(11):5019-5029. https://doi.org/10.1002/mp.13159
Wang XX, Ding Y, Wang SW, Dong D, Li HL, Chen J, Hu H, Lu C, Tian J, Shan XH (2020). Intratumoral and peritumoral radiomics analysis for preoperative Lauren classification in gastric cancer. Cancer Imaging 20(1):83-92. https://doi.org/10.1186/s40644-020-00358-3
Jordan MI, Mitchell TM (2015). Machine learning: Trends, perspectives, and prospects. Science 349(6245):255-260. https://doi.org/10.1126/science.aaa8415
Suarez-Ibarrola R, Hein S, Reis G, Gratzke C, Miernik A (2020). Current and future applications of machine and deep learning in urology: a review of the literature on urolithiasis, renal cell carcinoma, and bladder and prostate cancer. World J Urol 38(10):2329-2347. https://doi.org/10.1007/s00345-019-03000-5
Bektas CT, Kocak B, Yardimci AH, Turkcanoglu MH, Yucetas U, Koca SB, Erdim C, Kilickesmez O (2019). Clear cell renal cell carcinoma: Machine learning-based quantitative computed tomography texture analysis for prediction of Fuhrman nuclear grade. Eur Radiol 29(3):1153-1163. https://doi.org/10.1007/s00330-018-5698-2
Qiu W, Duan N, Chen X, Ren S, Zhang Y, Wang Z, Chen R (2019) Pancreatic ductal adenocarcinoma: Machine learning-based quantitative computed tomography texture analysis for prediction of histopathological grade. Cancer Manag Res 11(1):9253-9264. https://doi.org/10.2147/CMAR.S218414
Zhang C, Chen T (2001). Efficient feature extraction for 2D/3D objects in mesh representation. Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205) 1(3):935–938.
Chen T, Guestrin C (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 1(1):785–794.
Gurbani S, Morgan D, Jog V, Dreyfuss L, Shen M, Das A, Abel EJ, Lubner MG (2021). Evaluation of radiomics and machine learning in identification of aggressive tumor features in renal cell carcinoma (RCC) [published online ahead of print, 2021 Apr 15]. Abdom Radiol (NY) 2021; 1–11. https://doi.org/10.1007/s00261-021-03083-y
Lundberg SM, Lee S (2017). A unified approach to interpreting model predictions. 31st conference on Neural Information Processing Systems 1(1):1–10.
Pezzilli R, Buscarini E, Pollini T, et al. (2020) Epidemiology, clinical features and diagnostic work-up of cystic neoplasms of the pancreas: Interim analysis of the prospective PANCY survey. Dig Liver Dis 52(5):547-554. https://doi.org/10.1016/j.dld.2020.02.003
Xie H, Ma S, Guo X, Zhang X, Wang X (2020). Preoperative differentiation of pancreatic mucinous cystic neoplasm from macrocystic serous cystic adenoma using radiomics: Preliminary findings and comparison with radiological model. Eur J Radiol 122(1):108747-108753. https://doi.org/10.1016/j.ejrad.2019.108747
Springer S, Masica DL, Dal Molin M, et al. (2019). A multimodality test to guide the management of patients with a pancreatic cyst. Sci Transl Med 11(501):1-29. https://doi.org/10.1126/scitranslmed.aav4772
Masica DL, Dal Molin M, Wolfgang CL, et al. (2017). A novel approach for selecting combination clinical markers of pathology applied to a large retrospective cohort of surgically resected pancreatic cysts. J Am Med Inform Assoc 24(1):145-152. https://doi.org/10.1093/jamia/ocw069
Elta GH, Enestvedt BK, Sauer BG, Lennon AM (2018). ACG clinical guideline: Diagnosis and management of pancreatic cysts. Am J Gastroenterol 113(4):464-479. https://doi.org/10.1038/ajg.2018.14
Funding
ML4MI internal funding.
Author information
Authors and Affiliations
Contributions
Not applicable.
Corresponding author
Ethics declarations
Conflict of interest
M Lubner: Prior Grant funding from Ethicon, Philips. The remaining authors do not have any disclosures. This data was previously presented as a scientific abstract at the virtual American Roentgen Ray Society meeting 2021; and as a scientific poster at the virtual Society of Abdominal Radiology meeting 2021.
Ethics approval
This study was approved by our Institutional Review board and was found to be HIPAA complaint. Thus, requirement for informed consent was waived.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Awe, A.M., Vanden Heuvel, M.M., Yuan, T. et al. Machine learning principles applied to CT radiomics to predict mucinous pancreatic cysts. Abdom Radiol 47, 221–231 (2022). https://doi.org/10.1007/s00261-021-03289-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00261-021-03289-0