Cross-Modal Attention Fusion of Radiomic Lung Nodule Features and Transcriptomic Phase-Separation Signatures for Explainable Cancer Risk Stratification
Keywords:
cross-modal attention, radiomics, transcriptomics, phase separation, lung nodule, risk stratification, explainable AI, multimodal fusion, governance, clinical deploymentAbstract
Precision oncology demands risk stratification frameworks that synthesize heterogeneous biological evidence into actionable and interpretable predictions. This paper presents a system-level architecture for cross-modal attention fusion that integrates radiomic features extracted from lung nodule imaging with transcriptomic signatures shaped by biomolecular phase separation, aiming to deliver explainable cancer risk stratification. The work does not propose a single novel model; rather, it offers a sustained examination of the structural, governance, infrastructure, and policy dimensions associated with deploying such a cross-modal system in real clinical contexts. We argue that contemporary multimodal fusion methods, while powerful, often obscure the reasoning behind predictive outcomes, creating barriers to clinical trust, regulatory approval, and equitable deployment. By centering attention mechanisms that explicitly decouple modality-specific feature extraction from cross-modal interaction, the architecture promotes transparency through interpretable attention pathways. We discuss how radiomic features, long established for noninvasive tumor phenotyping, can be enriched by dual-attention segmentation networks, and how transcriptomic phase-separation signatures, which reveal condensate-driven gene regulation, provide a dynamic molecular complement. The paper critically examines trade-offs between predictive performance and explainability, explores infrastructure requirements for ingesting and aligning radiology and sequencing data at scale, and addresses fairness, robustness, and governance challenges that arise when such systems are embedded in clinical decision support. Throughout, the discussion emphasizes systemic resilience, policy alignment with evolving medical AI regulation, and sustainability of cross-institutional data collaborations. By treating the fusion pipeline as a socio-technical infrastructure, we provide a forward-looking perspective that connects attention-driven representation learning with the broader imperatives of responsible translation in oncology.
References
1. Gillies, R. J., Kinahan, P. E., & Hricak, H. (2016). Radiomics: Images are more than pictures, they are data. Radiology, 278(2), 563–577. https://doi.org/10.1148/radiol.2015151169
2. Lambin, P., Rios-Velazquez, E., Leijenaar, R., Carvalho, S., van Stiphout, R. G. P. M., Granton, P. V., … & Aerts, H. J. W. L. (2012). Radiomics: Extracting more information from medical images using advanced feature analysis. European Journal of Cancer, 48(4), 441–446. https://doi.org/10.1016/j.ejca.2011.11.036
3. Aerts, H. J. W. L., Velazquez, E. R., Leijenaar, R. T. H., Parmar, C., Grossmann, P., Carvalho, S., … & Lambin, P. (2014). Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nature Communications, 5, 4006. https://doi.org/10.1038/ncomms5006
4. Chang, C., Fu, M., Chen, X., et al. (2025, November). Research on PDU-Net Lung Nodule Segmentation Algorithm Based on Path Aggregation and Dual Attention. In 2025 4th International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) (pp. 1897-1900). IEEE.
5. Setio, A. A. A., Traverso, A., de Bel, T., Berens, M. S. N., van den Bogaard, C., Cerello, P., … & van Ginneken, B. (2017). Pulmonary nodule detection in CT images: false positive reduction using multi-view convolutional networks. IEEE Transactions on Medical Imaging, 36(5), 1160–1169. https://doi.org/10.1109/TMI.2017.2650162
6. Ardila, D., Kiraly, A. P., Bharadwaj, S., Choi, B., Reicher, J. J., Peng, L., … & Shetty, S. (2019). End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature Medicine, 25(6), 954–961. https://doi.org/10.1038/s41591-019-0447-x
7. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., … & Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems 30 (pp. 5998–6008).
8. Baltrusaitis, T., Ahuja, C., & Morency, L.-P. (2019). Multimodal machine learning: A survey and taxonomy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(2), 423–443. https://doi.org/10.1109/TPAMI.2018.2798607
9. Shin, Y., & Brangwynne, C. P. (2017). Liquid phase condensation in cell physiology and disease. Science, 357(6357), eaaf4382. https://doi.org/10.1126/science.aaf4382
10. Boeynaems, S., Alberti, S., Fawzi, N. L., Mittag, T., Polymenidou, M., Rousseau, F., … & Fuxreiter, M. (2018). Protein phase separation: A new phase in cell biology. Trends in Cell Biology, 28(6), 420–435. https://doi.org/10.1016/j.tcb.2018.02.004
11. Chung, C.-I., et al. (2024). Phase separation of YAP-MAML2 differentially regulates the transcriptome. Proceedings of the National Academy of Sciences, 121, e2310430121.
12. Cancer Genome Atlas Research Network. (2014). Comprehensive molecular profiling of lung adenocarcinoma. Nature, 511(7511), 543–550. https://doi.org/10.1038/nature13385
13. Gentles, A. J., Bratman, S. V., Lee, L. J., et al. (2015). Integrating tumor and stromal gene expression signatures with clinical indices for survival stratification in lung cancer. Journal of the National Cancer Institute, 107(10), djv168. https://doi.org/10.1093/jnci/djv168
14. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems 30 (pp. 4765–4774).
15. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). “Why should I trust you?”: Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135–1144). https://doi.org/10.1145/2939672.2939778
16. Ding, Y., Wang, X., Yuan, H., Qu, M., & Jian, X. (2025). Decoupling feature-driven and multimodal fusion attention for clothing-changing person re-identification. Artificial Intelligence Review, 58(8), 241.
17. Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019). Dissecting racial bias in an algorithm used to manage the health of populations. Science, 366(6464), 447–453. https://doi.org/10.1126/science.aax2342
18. Mittelstadt, B. D., Allo, P., Taddeo, M., Wachter, S., & Floridi, L. (2016). The ethics of algorithms: Mapping the debate. Big Data & Society, 3(2), 2053951716679679. https://doi.org/10.1177/2053951716679679
19. Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115–118. https://doi.org/10.1038/nature21056
20. Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision (pp. 618–626). https://doi.org/10.1109/ICCV.2017.74
21. Finlayson, S. G., Bowers, J. D., Ito, J., Zittrain, J. L., Beam, A. L., & Kohane, I. S. (2019). Adversarial attacks on medical machine learning. Science, 363(6433), 1287–1289. https://doi.org/10.1126/science.aaw4399
22. Chung, C.-I. et al. Phase separation of YAP-MAML2 differentially regulates the transcriptome. Proc. Natl Acad. Sci. Usa. 121, e2310430121 (2024).
23. Hnisz, D., Shrinivas, K., Young, R. A., Chakraborty, A. K., & Sharp, P. A. (2017). A phase separation model for transcriptional control. Cell, 169(1), 13–23. https://doi.org/10.1016/j.cell.2017.02.007
24. Price, W. N., & Cohen, I. G. (2019). Privacy in the age of medical big data. Nature Medicine, 25(1), 37–43. https://doi.org/10.1038/s41591-018-0272-7
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Clinical and Translational Medicine
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
