2026
doi: 10.12720/jait.17.4.800-815
Federated Multi-Omics Breast Cancer Prognosis Using Optimized Graph-Enhanced Capsule Tab Transformer with Explainability Support
Umme Najma1
,
Kalai Vani Yenamandram Sathyanarayana2 ,
Bhavya Ganga Reddy3,* ,
Savitha Suguna Kumar4 ,
Tirumalasetti Lakshmi Narayana5 ,
Sangita Chakraborty Bagchi6 ,
Sushil Lekhi7 ,
Belathur Ramanna Vatsala8 ,
Manoranjan Dash9 ,
Prasenjit Kumar Das10 ,
Manjunath Thimmasandra Narayanappa11 ,
Jyothi Nelahonne Mohan12
- 1Department of Biotechnology, Government Science College (Autonomous), Hassan, India
- 2Information Science and Engineering, BMS Institute of Technology and Management, Bengaluru, India
- 3Information Science and Engineering, BMS Institute of Technology and Management, Visvesvaraya Technological University, Belagavi, India
- 4Computer Science and Engineering, BMS Institute of Technology and Management, Bengaluru, India
- 5Department of Electrical and Electronics Engineering, Aditya University, Surampalem, India
- 6Computer Science and Engineering, ITM University, Gwalior, India
- 7Artificial intelligence and Machine Learning, Lovely Professional University, Jalandhar, India
- 8Computer Science and Engineering, The National Institute of Engineering, Mysuru, India
- 9Faculty of Management Sciences, Siksha O Anusandhan (Deemed to be University), Bhubaneswar, India
- 10Faculty of Computer Technology, Assam Down Town University, Guwahati, India
- 11Computer Science and Engineering, Sai Vidya Institute of Technology, Bengaluru, India
- 12Computer Science Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, India
- Manuscript receivedAugust 29, 2025
- revisedNovember 12, 2025
- acceptedNovember 18, 2025
- publishedApril 24, 2026
Abstract
Breast cancer prognosis benefits from multi-omics integration but faces challenges of class imbalance, calibration, and limited interpretability. To develop a federated multi-omics framework that enhances predictive accuracy, stability, and interpretability while preserving data privacy, the framework uses a Tab Transformer (TT) backbone with Graph Capsule (GC) for structural feature learning, Golden Eagle Optimization (GEO) for stable convergence, and explainability modules Shapley Additive Explanations (SHAP), Gradient-weighted Class Activation Mapping (Grad-CAM)) for transparent outputs. Federated learning was applied across the Cancer Genome Atlas-Breast Invasive Carcinoma (TCGA-BRCA), the Cancer Genome Atlas Pan-Cancer (TCGA-PANCAN) BRCA subset, and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) datasets. Evaluation used accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), Cohen’s kappa, Top-2 accuracy, and Jensen-Shannon divergence. The global model achieved 98.7% accuracy, an F1-score of 0.983, an MCC of 0.951, and the lowest Jensen-Shannon (JS) divergence of 0.039. GC improved feature interactions, GEO enhanced optimization stability, and the explainability modules supported biologically relevant insights. The framework improved accuracy, fairness, and calibration while ensuring interpretability and privacy, showing strong potential for clinical adoption.
Keywords
- breast cancer prognosis
- multi-omics
- federated learning
- tab transformer
- graph capsule
- golden eagle optimization
- explainable artificial intelligence
How to Cite
Umme Najma, Kalai Vani Yenamandram Sathyanarayana, Bhavya Ganga Reddy, Savitha Suguna Kumar, Tirumalasetti Lakshmi Narayana, Sangita Chakraborty Bagchi, Sushil Lekhi, Belathur Ramanna Vatsala, Manoranjan Dash, Prasenjit Kumar Das, Manjunath Thimmasandra Narayanappa, and Jyothi Nelahonne Mohan, "Federated Multi-Omics Breast Cancer Prognosis Using Optimized Graph-Enhanced Capsule Tab Transformer with Explainability Support," Journal of Advances in Information Technology, vol. 17, no. 4, pp. 800-815, 2026. https://doi.org/10.12720/jait.17.4.800-815
Copyright & License
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
