Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost

Tuğçe Taşar Yildirim; Nevzat Gözel; Selahattin Barış Çelebi; Beyda Taşar; Orhan Yaman; Sedef Onay Demir; Aykut Bulu; Gülşah Karaduman

doi:10.70058/cjm.1831642

Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost

Abstract

Objective: This study aims to construct a high-precision decision support model by integrating explainable artificial intelligence (XAI) techniques utilizing urinary and plasma biomarkers. Methods: We used an open-access dataset published by Debernardi et al. (2020) containing the biomarkers Lymphatic Vessel Endothelial Hyaluronan Receptor 1 (LYVE1), Regenerating Family Member 1 Beta (REG1B), Trefoil Factor 1 (TFF1), and plasma CA19-9. The preprocessing phase involved missing value imputation, Z-score normalization, and feature engineering. An autoencoder (AE)-based unsupervised learning framework was employed for dimensionality reduction. Classification was performed using an XGBoost algorithm optimized via the Optuna framework. Class imbalance was addressed through the Synthetic Minority Over-sampling Technique (SMOTE). Model interpretability was ensured using SHapley Additive Explanations (SHAP). Results: The proposed Autoencoder–XGBoost model optimized with Optuna outperformed conventional methods, achieving an accuracy of 95.8%, 95% precision, 93% recall, 93% F1-score, and an AUC of 0.984. SHAP analysis identified plasma CA19-9, LYVE1, creatinine, and age as the most influential predictors contributing to model decisions. Conclusion: The developed XAI framework offers high diagnostic accuracy and transparent decision logic for the early detection of PDAC. By leveraging the clinical potential of urinary biomarkers, the model demonstrates strong applicability for integration into screening and risk stratification modules of clinical decision support systems.

Keywords

References

1. Debernardi S, O’Brien H, Algahmdi AS, Malats N, Stewart GD, Plješa-Ercegovac M, et al. A combination of urinary biomarker panel and PancRISK score for earlier detection of pancreatic cancer: a case-control study. PLoS Med. 2020;17(12):e1003489. doi:10.1371/journal.pmed.1003489.
2. Hinestrosa JP, Dhani H, Schroeder GK, Lewis JM, Balcer HI, Kurzrock R, et al. Abstract 6515: Pancreatic ductal adenocarcinoma (PDAC) early detection. Cancer Res. 2023;83(7 Suppl):Abstract nr 6515. doi:10.1158/1538-7445. AM2023-6515.
3. Nakamura K, Zhu Z, Roy S, Jun E, Han H, Munoz RM, et al. An exosome-based transcriptomic signature for noninvasive, early detection of patients with pancreatic ductal adenocarcinoma: a multicenter cohort study. Gastroenterology. 2022;163(5):1252-66.e2. doi:10.1053/j.gastro.2022.06.090.
4. Bengtsson A, Andersson R, Ansari D. The actual 5-year survivors of pancreatic ductal adenocarcinoma based on realworld data. Sci Rep. 2020;10:16425. doi:10.1038/s41598-020-73525-y.
5. Yaman O, Dogan S, Tuncer T, Subasi A. Skin cancer classification model based on hybrid deep feature generation and iterative mRMR. In: Bajaj V, Ansari IA, editors. Computational intelligence based solutions for vision systems. IOP Publishing; 2022. p. 1-24. doi:10.1088/978-0-7503-4821-8ch4.
6. Murthy AV, Nagendran A, Rao D. Pancreatic cancer: early detection and novel therapies. Journal of Translational Gastroenterology. 3(2):74-92. doi:10.14218/JTG.2024.00037.
7. Yaman O, Tuncer T. Exemplar pyramid deep feature extraction based cervical cancer image classification model using pap-smear images. Biomed Signal Process Control. 2022;73:103428. doi:10.1016/j.bspc.2021.103428.
8. Daher H, Punchayil SA, Ismail AAE, Fernandes RR, Jacob J, Algazzar MH, et al. Advancements in pancreatic cancer detection: integrating biomarkers, imaging technologies, and machine learning for early diagnosis. Cureus. 2024;16(3):e56583. doi:10.7759/cureus.56583.

9. Yang J, Xu R, Wang C, Qiu J, Ren B, You L. Early screening and diagnosis strategies of pancreatic cancer: a comprehensive review. Cancer Commun (Lond). 2021;41(12):1257-74. doi:10.1002/cac2.12204.
10. O’Brien DP, Sandanayake NS, Jenkinson C, Gentry-Maharaj A, Apostolidou S, Fourkala EO, et al. Serum CA19-9 is significantly upregulated up to 2 years before diagnosis with pancreatic cancer: implications for early disease detection. Clin Cancer Res. 2015;21(3):622-31. doi:10.1158/1078-0432.CCR-14-0365.
11. Zakari S, Niels NK, Olagunju GV, Nnaji PC, Ogunniyi O, Tebamifor M, et al. Emerging biomarkers for non-invasive diagnosis and treatment of cancer: a systematic review. Front Oncol. 2024;14:1405267. doi:10.3389/fonc.2024.1405267.
12. Sequeira-Antunes B, Ferreira HA. Urinary biomarkers and point-of-care urinalysis devices for early diagnosis and management of disease: a review. Biomedicines. 2023;11(4):1051. doi:10.3390/biomedicines11041051.
13. Lepowsky E, Ghaderinezhad F, Knowlton S, Tasoglu S. Paper-based assays for urine analysis. Biomicrofluidics. 2017;11(5):051501. doi:10.1063/1.4996768.
14. Jordaens S, Zwaenepoel K, Tjalma W, Deben C, Beyers K, Vankerckhoven V, et al. Urine biomarkers in cancer detection: a systematic review of preanalytical parameters and applied methods. Int J Cancer. 2023;152(10):2186-205. doi:10.1002/ijc.34434.
15. Li L. The investigation of the correlation between 4 urine biomarkers and intelligent diagnosis of PDAC. In: 2022 IEEE 2nd international conference on electronic technology, communication and information (ICETCI); 2022 May 27-29; Changchun, China. Piscataway (NJ): IEEE; 2022. p. 593-6. doi:10.1109/ICETCI55101.2022.9832090.
16. Hoffmann W. Trefoil factor family (TFF) peptides and their links to inflammation: a re-evaluation and new medical perspectives. Int J Mol Sci. 2021;22(9):4909. doi:10.3390/ijms22094909.
17. Reddy P, Chandrasekar M. PAD: a pancreatic cancer detection based on extracted medical data through ensemble methods in machine learning. International Journal of Advanced Computer Science and Applications. 2022;13(2):149-56. doi:10.14569/IJACSA.2022.0130218.
18. Upadhyay A, Lalwaney A, Sharma A. Pancreatic cancer detection using radial basis neural network. SN Comput Sci. 2023;4(3):261. doi:10.1007/s42979-022-01643-7.
19. Ashfaq N, Suhail Z, Khalid A, Sarwar N, Irshad A, Yaman O, et al. Skinsight: advancing deep learning for skin cancer diagnosis and classification. Discover Computing. 2025;28:63. doi:10.1007/s10791-025-09541-1.
20. Kaczor-Urbanowicz KE, Cheng J, King JC, Sedarat A, Pandol SJ, Farrell JJ, et al. Reviews on current liquid biopsy for detection and management of pancreatic cancers. Pancreas. 2020;49(9):1141-52. doi:10.1097/MPA.0000000000001662.
21. Ali H. Artificial intelligence in multi-omics data integration: advancing precision medicine, biomarker discovery and genomic-driven disease interventions. International Journal of Science and Research Archive. 2023;8(1):1012-30. doi:10.30574/ijsra.2023.8.1.0189.
22. Backes C, Meese E, Keller A. Specific miRNA disease biomarkers in blood, serum and plasma: challenges and prospects. Mol Diagn Ther. 2016;20(6):509-18. doi:10.1007/s40291-016-0221-4.
23. Pu X, Zhang C, Ding G, Gu H, Lv Y, Shen T, et al. Diagnostic plasma small extracellular vesicles miRNA signatures for pancreatic cancer using machine learning methods. Transl Oncol. 2024;40:101847. doi:10.1016/j.tranon.2023.101847.
24. Dang C, Bian Q, Wang F, Wang H, Liang Z. Machine learning identifies SLC6A14 as a novel biomarker promoting the proliferation and metastasis of pancreatic cancer via Wnt/ β-catenin signaling. Sci Rep. 2024;14:2116. doi:10.1038/s41598-024-52646-8.
25. Mahawan T, Luckett T, Mielgo Iza A, Pornputtapong N, Caamaño Gutiérrez E. Robust and consistent biomarker candidates identification by a machine learning approach applied to pancreatic ductal adenocarcinoma metastasis. BMC Med Inform Decis Mak. 2024;24(Suppl 4):175. doi:10.1186/s12911-024-02578-0.
26. Severeyn E, La Cruz AL, Velásquez J, Huerta M. Early diagnosis of pancreatic cancer using urinary biomarkers and machine learning. In: 2024 IEEE 8th Ecuador Technical Chapters Meeting (ETCM); 2024 Oct; Cuenca, Ecuador. Piscataway (NJ): IEEE; 2024. p. 267-71. doi:10.1109/ETCM63562.2024.10746008.
27. Çelebi SB, Emiroğlu BG. Leveraging deep learning for enhanced detection of Alzheimer’s disease through morphometric analysis of brain images. Traitement du Signal. 2023;40(4):1355-65. doi:10.18280/ts.400405.
28. Patel H, Zanos T, Hewitt DB. Deep learning applications in pancreatic cancer. Cancers (Basel). 2024;16(2):436. doi:10.3390/cancers16020436.
29. Çelebi SB, Emiroğlu BG. A novel deep dense block-based model for detecting Alzheimer’s disease. Applied Sciences. 2023;13(15):8686. doi:10.3390/app13158686.
30. Sijithra P, Santhi N, Ramasamy N. A review study on early detection of pancreatic ductal adenocarcinoma using artificial intelligence assisted diagnostic methods. Eur J Radiol. 2023;166:110972. doi:10.1016/j.ejrad.2023.110972.
31. Acer İ, Bulucu FO, İçer S, Latifoğlu F. Early diagnosis of pancreatic cancer by machine learning methods using urine biomarker combinations. Turkish Journal of Electrical Engineering & Computer Sciences. 2023;31(1):112-25. doi:10.55730/1300-0632.3974.
32. Chen PT, Wu T, Wang P, Chang D, Liu KL, Wu MS, et al. Pancreatic cancer detection on CT scans with deep learning: a nationwide population-based study. Radiology. 2023;306(1):172-82. doi:10.1148/radiol.220152.
33. Almisned FA, Usanase N, Ozsahin DU, Ozsahin I. Incorporation of explainable artificial intelligence in ensemble machine learning-driven pancreatic cancer diagnosis. Sci Rep. 2025;15(1):14038. doi:10.1038/s41598-025-98298-0.
34. National Cancer Institute. The cancer genome atlas program [Internet]. [cited 2025 Nov 25]. Available from: https://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tcga
35. National Center for Biotechnology Information. GEO DataSets [Internet]. [cited 2025 Nov 25]. Available from: https://www.ncbi.nlm.nih.gov/gds
36. Zhang J, Baran J, Cros A, Guberman JM, Haider S, Hsu J, et al. International cancer genome consortium data portal—a one-stop shop for cancer genomics data. Database (Oxford). 2011;2011:bar026. doi:10.1093/database/bar026.
37. Edwards NJ, Oberti M, Thangudu RR, Cai S, McGarvey PB, Jacob S, et al. The CPTAC data portal: a resource for cancer proteomics research. J Proteome Res. 2015;14(6):2707-13. doi:10.1021/pr501254j.
38. Karar ME, El-Fishawy N, Radad M. Automated classification of urine biomarkers to diagnose pancreatic cancer using 1-D convolutional neural networks. J Biol Eng. 2023;17:28. doi:10.1186/s13036-023-00340-0.
39. Blyuss O, Zaikin A, Cherepanova V, Munblit D, Kiseleva EM, Prytomanova OM, et al. Development of PancRISK, a urine biomarker-based risk score for stratified screening of pancreatic cancer patients. Br J Cancer. 2020;122(5):692-6. doi:10.1038/s41416-019-0694-0.
40. Shelly M, Sivakumari S. Enhancing pancreatic cancer diagnostics: ensemble-based model for automated urine biomarker classification. Comput Biol Med. 2025;189:109997. doi:10.1016/j.compbiomed.2025.109997.
41. Iwano T, Okazaki M, Tsumura H, Morita S, Akimoto M, Sugiyama R, et al. High-performance collective biomarker from liquid biopsy for diagnosis of pancreatic cancer based on mass spectrometry and machine learning. J Cancer. 2021;12(24):7477-87. doi:10.7150/jca.63244.
42. Davis JJ IV. Urinary biomarkers for pancreatic cancer [Internet]. 2020 [cited 2025 Nov 25]. Available from: https://www.kaggle.com/datasets/johnjdavisiv/urinary-biomarkers-for-pancreatic-cancer
43. He Y, Deng K, Han J. Patent value prediction in biomedical textiles: a method based on a fusion of machine learning models. PLoS One. 2025;20(4):e0322182. doi:10.1371/journal.pone.0322182.
44. Lundberg SM, Lee SI. A unified approach to interpreting model predictions. arXiv [Preprint]. 2017 [cited 2025 Nov 25]. Available from: https://arxiv.org/abs/1705.07874 doi:10.48550/arXiv.1705.07874.
45. Hinton GE, Salakhutdinov RR. Reducing the dimensionality of data with neural networks. Science. 2006;313(5786):504-7. doi:10.1126/science.1127647.
46. Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP. SMOTE: synthetic minority over-sampling technique. J Artif Intell Res. 2002;16:321-57. doi:10.1613/jair.953.
47. Chen T, Guestrin C. XGBoost: a scalable tree boosting system. In: Proc 22nd ACM SIGKDD Int Conf Knowledge Discovery and Data Mining. ACM; 2016. p. 785-94. doi:10.1145/2939672.2939785.
48. Akiba T, Sano S, Yanase T, Ohta T, Koyama M. Optuna: a next-generation hyperparameter optimization framework. In: Proc 25th ACM SIGKDD Int Conf Knowledge Discovery & Data Mining. ACM; 2019. p. 2623-31. doi:10.1145/3292500.3330701.
49. Fawcett T. An introduction to ROC analysis. Pattern Recognit Lett. 2006;27(8):861-74. doi:10.1016/j.patrec.2005.10.010.
50. Pan S, Liu Z, Han Y, Zhang D, Zhao X, Li J, et al. Using the Pearson’s correlation coefficient as the sole metric to measure the accuracy of quantitative trait prediction: is it sufficient? Front Plant Sci. 2024;15:1480463. doi:10.3389/fpls.2024.1480463.

Details

Primary Language

English

Subjects

Clinical Oncology

Journal Section

Research Article

Authors

Tuğçe Taşar Yildirim ^*
0000-0003-0769-6357
Türkiye

Nevzat Gözel
0000-0001-7326-6860
Türkiye

Selahattin Barış Çelebi
0000-0002-6235-9348
United Kingdom

Beyda Taşar
0000-0002-4689-8579
Türkiye

Orhan Yaman
0000-0001-9623-2284
Türkiye

Sedef Onay Demir
0000-0002-6448-5153
Türkiye

Aykut Bulu
0000-0002-5792-3557
Türkiye

Gülşah Karaduman
0000-0001-8034-3019
Türkiye

Early Pub Date

January 26, 2026

Publication Date

January 26, 2026

Submission Date

November 27, 2025

Acceptance Date

January 2, 2026

Published in Issue

Year 2026 Volume: 3 Number: 1

DOI

https://doi.org/10.70058/cjm.1831642

IZ

https://izlik.org/JA24RN43SP

Cite

RIS / Bibtex

APA

Taşar Yildirim, T., Gözel, N., Çelebi, S. B., Taşar, B., Yaman, O., Onay Demir, S., Bulu, A., & Karaduman, G. (2026). Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost. Cerasus Journal of Medicine, 3(1), 37-56. https://doi.org/10.70058/cjm.1831642

AMA

1.Taşar Yildirim T, Gözel N, Çelebi SB, et al. Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost. Cerasus J Med. 2026;3(1):37-56. doi:10.70058/cjm.1831642

Chicago

Taşar Yildirim, Tuğçe, Nevzat Gözel, Selahattin Barış Çelebi, et al. 2026. “Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost”. Cerasus Journal of Medicine 3 (1): 37-56. https://doi.org/10.70058/cjm.1831642.

EndNote

Taşar Yildirim T, Gözel N, Çelebi SB, Taşar B, Yaman O, Onay Demir S, Bulu A, Karaduman G (January 1, 2026) Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost. Cerasus Journal of Medicine 3 1 37–56.

IEEE

[1]T. Taşar Yildirim et al., “Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost”, Cerasus J Med, vol. 3, no. 1, pp. 37–56, Jan. 2026, doi: 10.70058/cjm.1831642.

ISNAD

Taşar Yildirim, Tuğçe - Gözel, Nevzat - Çelebi, Selahattin Barış - Taşar, Beyda - Yaman, Orhan - Onay Demir, Sedef - Bulu, Aykut - Karaduman, Gülşah. “Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost”. Cerasus Journal of Medicine 3/1 (January 1, 2026): 37-56. https://doi.org/10.70058/cjm.1831642.

JAMA

1.Taşar Yildirim T, Gözel N, Çelebi SB, Taşar B, Yaman O, Onay Demir S, Bulu A, Karaduman G. Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost. Cerasus J Med. 2026;3:37–56.

MLA

Taşar Yildirim, Tuğçe, et al. “Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost”. Cerasus Journal of Medicine, vol. 3, no. 1, Jan. 2026, pp. 37-56, doi:10.70058/cjm.1831642.

Vancouver

1.Tuğçe Taşar Yildirim, Nevzat Gözel, Selahattin Barış Çelebi, Beyda Taşar, Orhan Yaman, Sedef Onay Demir, Aykut Bulu, Gülşah Karaduman. Explainable AI Framework for Urinary Biomarker-Based Early Detection of Pancreatic Cancer: Clinical Implications of Autoencoder-Optimized XGBoost. Cerasus J Med. 2026 Jan. 1;3(1):37-56. doi:10.70058/cjm.1831642