Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence
- PMID: 32874057
- PMCID: PMC7438201
- DOI: 10.3748/wjg.v26.i30.4453
Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence
Abstract
Background: Despite advancements in operative technique and improvements in postoperative managements, postoperative pancreatic fistula (POPF) is a life-threatening complication following pancreatoduodenectomy (PD). There are some reports to predict POPF preoperatively or intraoperatively, but the accuracy of those is questionable. Artificial intelligence (AI) technology is being actively used in the medical field, but few studies have reported applying it to outcomes after PD.
Aim: To develop a risk prediction platform for POPF using an AI model.
Methods: Medical records were reviewed from 1769 patients at Samsung Medical Center who underwent PD from 2007 to 2016. A total of 38 variables were inserted into AI-driven algorithms. The algorithms tested to make the risk prediction platform were random forest (RF) and a neural network (NN) with or without recursive feature elimination (RFE). The median imputation method was used for missing values. The area under the curve (AUC) was calculated to examine the discriminative power of algorithm for POPF prediction.
Results: The number of POPFs was 221 (12.5%) according to the International Study Group of Pancreatic Fistula definition 2016. After median imputation, AUCs using 38 variables were 0.68 ± 0.02 with RF and 0.71 ± 0.02 with NN. The maximal AUC using NN with RFE was 0.74. Sixteen risk factors for POPF were identified by AI algorithm: Pancreatic duct diameter, body mass index, preoperative serum albumin, lipase level, amount of intraoperative fluid infusion, age, platelet count, extrapancreatic location of tumor, combined venous resection, co-existing pancreatitis, neoadjuvant radiotherapy, American Society of Anesthesiologists' score, sex, soft texture of the pancreas, underlying heart disease, and preoperative endoscopic biliary decompression. We developed a web-based POPF prediction platform, and this application is freely available at http://popfrisk.smchbp.org.
Conclusion: This study is the first to predict POPF with multiple risk factors using AI. This platform is reliable (AUC 0.74), so it could be used to select patients who need especially intense therapy and to preoperatively establish an effective treatment strategy.
Keywords: Neural networks; Pancreatoduodenectomy; Postoperative pancreatic fistula; Recursive feature elimination.
©The Author(s) 2020. Published by Baishideng Publishing Group Inc. All rights reserved.
Conflict of interest statement
Conflict-of-interest statement: There are no financial or any potential personal conflicts of interest to declare for any of the authors.
Figures


Similar articles
-
Preoperative adiposity at bioimpedance vector analysis improves the ability of Fistula Risk Score (FRS) in predicting pancreatic fistula after pancreatoduodenectomy.Pancreatology. 2020 Apr;20(3):545-550. doi: 10.1016/j.pan.2020.01.008. Epub 2020 Jan 16. Pancreatology. 2020. PMID: 31980350
-
Preoperative anthropomorphic radiographic measurements can predict postoperative pancreatic fistula formation following pancreatoduodenectomy.Am J Surg. 2021 Jul;222(1):133-138. doi: 10.1016/j.amjsurg.2020.10.023. Epub 2020 Oct 20. Am J Surg. 2021. PMID: 33390246
-
Postoperative Pancreatic Fistula Following Pancreaticoduodenectomy-Stratification of Patient Risk.J Gastrointest Surg. 2019 Sep;23(9):1817-1824. doi: 10.1007/s11605-018-4045-x. Epub 2018 Nov 26. J Gastrointest Surg. 2019. PMID: 30478529
-
Risk stratification of postoperative pancreatic fistula and other complications following pancreatoduodenectomy. How far are we? A scoping review.Langenbecks Arch Surg. 2025 Feb 7;410(1):62. doi: 10.1007/s00423-024-03581-9. Langenbecks Arch Surg. 2025. PMID: 39915344 Free PMC article.
-
Prediction of postoperative pancreatic fistula and pancreatitis after pancreatoduodenectomy or distal pancreatectomy: A review.Scand J Surg. 2023 Jun;112(2):126-134. doi: 10.1177/14574969231167781. Epub 2023 Apr 21. Scand J Surg. 2023. PMID: 37083016 Review.
Cited by
-
Differential Performance of Machine Learning Models in Prediction of Procedure-Specific Outcomes.J Gastrointest Surg. 2022 Aug;26(8):1732-1742. doi: 10.1007/s11605-022-05332-x. Epub 2022 May 4. J Gastrointest Surg. 2022. PMID: 35508684 Free PMC article.
-
Exploring the application and future outlook of Artificial intelligence in pancreatic cancer.Front Oncol. 2024 Feb 21;14:1345810. doi: 10.3389/fonc.2024.1345810. eCollection 2024. Front Oncol. 2024. PMID: 38450187 Free PMC article. Review.
-
Preoperative evaluation to determine the difficulty of No. 6 lymphadenectomy in laparoscopic gastrectomy.BMC Surg. 2024 Feb 22;24(1):69. doi: 10.1186/s12893-024-02349-8. BMC Surg. 2024. PMID: 38389108 Free PMC article.
-
Deep learning-based prediction of post-pancreaticoduodenectomy pancreatic fistula.Sci Rep. 2024 Mar 1;14(1):5089. doi: 10.1038/s41598-024-51777-2. Sci Rep. 2024. PMID: 38429308 Free PMC article.
-
Research trends of artificial intelligence in pancreatic cancer: a bibliometric analysis.Front Oncol. 2022 Aug 2;12:973999. doi: 10.3389/fonc.2022.973999. eCollection 2022. Front Oncol. 2022. PMID: 35982967 Free PMC article.
References
-
- Chen BP, Chen M, Bennett S, Lemon K, Bertens KA, Balaa FK, Martel G. Systematic Review and Meta-analysis of Restrictive Perioperative Fluid Management in Pancreaticoduodenectomy. World J Surg. 2018;42:2938–2950. - PubMed
-
- Kambakamba P, Mannil M, Herrera PE, Müller PC, Kuemmerli C, Linecker M, von Spiczak J, Hüllner MW, Raptis DA, Petrowsky H, Clavien PA, Alkadhi H. The potential of machine learning to predict postoperative pancreatic fistula based on preoperative, non-contrast-enhanced CT: A proof-of-principle study. Surgery. 2020;167:448–454. - PubMed
MeSH terms
LinkOut - more resources
Full Text Sources