AI-powered predictive analytics framework for precision prognosis in liver cancer patients
Suman Mannsumanmann2021@gmail.comDepartment of Computer Science and EngineeringSingapore Institute of TechnologyPungoll Digital District, 828608, Singapore0000-0002-1067-2586View full profile → , Sivanesan Bala Krishnansivaneasan@singaporetech.edu.sgDepartment of Electrical and Electronics EngineeringSingapore Institute of TechnologyPungoll Digital District, 828608, Singapore0000-0002-4271-2677View full profile → , Amit Kumar Goeldrgoelkamit@gmail.comDepartment of Computer Science & EngineeringSir Padampat Singhania UniversityUdaipur, Rajasthan, 313601, India0000-0003-4204-926XView full profile → , Prasun Chakrabartidrprasun.cse@gmail.comDepartment of Computer Science & EngineeringSir Padampat Singhania UniversityUdaipur, Rajasthan, 313601, lndia0000-0003-0907-2433View full profile → , *Akanksha MahajanCorresponding authorer.aakanksha@yahoo.co.inDepartment of Computer Science & EngineeringPanipat Institute of Engineering & TechnologySamalkha, Haryana, 132102, India0000-0003-2775-7751View full profile →
* Corresponding author · click or hover a name for details
- Received:
- 01 Jun 2026
- Published Online:
- 30 Sep 2026
- Article type:
- Research Article
- Language:
- EN
- Article no.:
- JIOS-2438
- Pages:
- 3415–3424
Abstract
Liver cancer is one of the most deadly malignancies in the world with little survival opportunities. The conventional mathematical models of statistics do not have the ability to indicate intricate non-linear relationships between various forms of patient data. In this paper, a machine learning (Gradient Boosting Decision Trees) and deep neural networks (DNNs) algorithm with explainable AI (XAI) elements proposed are integrated to create a predictive analytics framework that precision predicts the likelihood of liver cancer. The model digests multimodal data such as clinical, biochemical and imaging features with state of art preprocess, feature selection and explainability piping like SHAP and LIME. On TCGA-LIHC data, the experimental assessment recorded an accuracy of 92.5 percent and C-index of 0.89, which is higher than that of the conventional models. The framework suggested offers clear patient-centered predictions, which will consequently lead to the achievement of precision oncology and AI-based clinical decision support.
Keywords
Subject Classifications
References
[1] J. M. Llovet, R. K. Kelley, A. Villanueva, A. G. Singal, E. Pikarsky, C. Roayaie, L. Lencioni, K. Koike, J. Zucman-Rossi, and T. L. Greten, “Hepatocellular carcinoma,” Nat. Rev. Dis. Primers, vol. 7, no. 1, Art. no. 6 (2021), doi: 10.1038/s41572-021-00241-2.
[2] World Health Organization, Global Cancer Observatory: Liver Cancer Fact Sheet. Geneva, Switzerland: WHO (2025). [Online]. Available: https://gco.iarc.fr
[3] C. H. Lee, H. Kim, and J. Cho, “Predicting survival in liver cancer using clinical and radiomic data,” Front. Oncol., vol. 13, Art. no. 1190532 (2023), doi: 10.3389/fonc.2023.1190532.
[4] Y. Chen, L. Zhang, and Q. Sun, “Random forest based survival analysis in liver cancer,” Comput. Biol. Med., vol. 146, Art. no. 105622 (2022), doi: 10.1016/j.compbiomed.2022.105622.
[5] T. Wang, H. Zhao, and J. Liu, “Gradient boosting machine learning for liver cancer outcomes,” IEEE Access, vol. 11, pp. 55231–55242 (2023), doi: 10.1109/ACCESS.2023.3278349
[6] J. Li, H. Zhang, and P. Lin, “Deep convolutional networks for liver tumor prognosis from CT imaging,” Front. Oncol., vol. 12, Art. no. 1145689 (2023), doi: 10.3389/fonc.2023.1145689.
[7] X. Zhou, Y. Zhang, W. Wang, L. Liu, and H. Chen, “Multi-omics integration via deep autoencoders for cancer prognosis,” Brief. Bioinform., vol. 25, no. 1, Art. no. bbad488 (2024), doi: 10.1093/bib/bbad488.
[8] A. Kumar, M. Patel, and R. Srivastava, “Hybrid ensemble model for multimodal liver cancer analytics,” Expert Syst. Appl., vol. 238, Art. no. 121962 (2024), doi: 10.1016/j.eswa.2023.121962.
[9] Y. Li and Q. Zhang, “Interpretable deep survival analysis using SHAP,” Artif. Intell. Med. (2025), doi: 10.1016/j.artmed.2025.102562.
[10] D. Patel, N. Kaur, and G. Singh, “Explainable artificial intelligence for oncology prognosis,” IEEE Trans. Biomed. Eng. (2025), doi: 10.1109/TBME.2025.3390547.
[11] L. Chen, Y. Wang, and X. Zhao, “Predicting hepatocellular carcinoma recurrence with deep learning and EHR data,” BMC Med. Inform. Decis. Mak., vol. 23, Art. no. 97 (2023), doi: 10.1186/s12911-023-02105-2.
[12] M. Sahoo, R. Dash, and S. Panda, “Radiogenomic hybrid learning for HCC recurrence,” BMC Bioinformatics, vol. 25, no. 2, Art. no. 245 (2024), doi: 10.1186/s12859-024-05690-z.
[13] P. Srivastava and R. Tripathi, “Explainable AI models in cancer informatics,” ACM Comput. Surv., vol. 56, no. 5, pp. 1–38 (2024), doi: 10.1145/3620342.
[14] S. Johnson, K. Roberts, and C. Tan, “Integrating AI and clinical features for survival prediction,” Sci. Rep., vol. 14, Art. no. 12752 (2024), doi: 10.1038/s41598-024-12752-4.
[15] L. Yang and W. Zhao, “Transformers for longitudinal patient data in oncology,” IEEE J. Biomed. Health Inform. (2025), doi: 10.1109/JBHI.2025.3341057.
[16] P. Gupta, T. Sharma, and A. Banerjee, “Federated learning for healthcare data privacy,” Nat. Mach. Intell., vol. 6, pp. 872–883 (2024), doi: 10.1038/s42256-024-00729-5.
[17] S. A. Raza and H. Imran, “Multimodal data fusion for cancer prognosis,” Knowl.-Based Syst., vol. 277, Art. no. 110859 (2024), doi: 10.1016/j.knosys.2023.110859.
[18] J. Brown, D. O’Neil, and Y. Xu, “AI-driven clinical decision support in hepatology,” Lancet Digit. Health, vol. 5, no. 7, pp. e412–e425 (2023), doi: 10.1016/S2589-7500(23)00091-4.
[19] N. Tandon, R. Gupta, and A. Mahajan, “Digital twins for precision oncology,” npj Digit. Med. (2025), doi: 10.1038/s41746-025-00812-1.
[20] M. Arora and G. Sharma, “Bias mitigation in AI based medical predictions,” IEEE Access, vol. 12, pp. 116122–116135 (2024), doi: 10.1109/ACCESS.2024.3421543.
[21] J. Xu, F. Wang, and Z. He, “Deep learning-based liver cancer survival prediction using multi-omics integration,” Bioinform. Adv., vol. 3, no. 1, Art. no. vbad024 (2023), doi: 10.1093/bioadv/vbad024.
[22] S. Kim, D. Lee, and J. Park, “Feature selection strategies for interpretable cancer prognosis models,” Pattern Recognit. Lett., vol. 168, pp. 72–81 (2023), doi: 10.1016/j.patrec.2023.05.018.
[23] L. Zhao and P. Liu, “Evaluating explainability methods in medical AI: A comparative study of SHAP, LIME, and Grad-CAM,” Artif. Intell. Rev., vol. 57, no. 2, pp. 1507–1533 (2024), doi: 10.1007/s10462-023-10412-5.
[24] H. Nguyen, K. Tran, and N. Pham, “Hybrid gradient boosting and deep neural networks for cancer risk prediction,” Comput. Biol. Med., vol. 170, Art. no. 107842 (2025), doi: 10.1016/j.compbiomed.2025.107842.
[25] R. Alabi, T. Adeyemi, and O. Akintola, “Validation of ensemble survival models for hepatocellular carcinoma,” Front. Public Health, vol. 12, Art. no. 1331247 (2024), doi: 10.3389/fpubh.2024.1331247.
[26] V. M. Wadhwa, A. Mahajan, T. Jena, and J. Garg, “An explainable artificial intelligence-based approach for prediction and analysis of mental health disorder due to work pressure in the tech industry,” Journal of Information and Optimization Sciences, vol. 46, no. 6, pp. 1821–1829 (2025), doi: 10.47974/JIOS-2011.
[27] I. Lekshmi and M. Regees, “A fuzzy NoVaRM vaccine selection decision-making model for Covid-19 : A case study in India,” Journal of Information and Optimization Sciences, vol. 47, no. 4, pp. 1–13 (2026), doi: 10.47974/JIOS-1842.
