Explainable Ai for Bone Marrow Cancer Diagnosis: from Bench to Bedside
Abstract
Diagnosing bone marrow cancer continues to be a difficult process because of the large morphological variation of malignantly transformed cells and
insufficient interpretability of traditional deep learning approaches. We suggest a new efficient, interpretable, and clinically applicable artificial
intelligence (AI) framework for cancer diagnostics based on the combination of imaging and genomic data. Our framework uses INT8 quantization,
20% filter pruning, and dual-modality explainability, where SHAP is used to attribute genome sequencing features, and Grad-CAM++ is applied for
lesions localization based on MRI images. Performance evaluation is done by comparing the model to existing baselines, which include HA-UNet and
SE-ResHybrid models. According to our experiments, the suggested framework shows increased computational efficiency, decreasing the inference
time from 64 ms/image to 38 ms/image (by 41%), with a simultaneous improvement of segmentation (3.7%) and classification (4.5%) accuracy. In
addition, our model demonstrates high agreement with clinical specialists, achieving a Cohen's kappa value of 0.79, which proves its high clinical
relevancy. Localization accuracy also increases by 12-15%.
References
[2]. Deshpande, N. M., Gite, S., and Pradhan, B. 2024. Unlocking the Potential: Machine Learning and Deep Learning in Leukemia Diagnosis with Explainable AI. IoT Sensors, ML, AI and XAI in Healthcare. Springer.
[3]. Gimeno, M., San José-Enériz, E., Villar, S., et al. 2022. Explainable Artificial Intelligence for Precision Medicine in Acute Myeloid Leukemia. Frontiers in Oncology. 12. Frontiers Media SA.
[4]. Hossain, M. A., Islam, A. K. M. M., Islam, S., and Shatabda, S. 2022. Symptom Based Explainable Artificial Intelligence Model for Leukemia Detection. IEEE Access. 10: 10344–10357.
[5]. Sabile, J. M. G., Zhang, P., Parwani, A. V., Chobrutsiky, B., et al. 2025. Toward Clinically Actionable Machine Learning and Artificial Intelligence Algorithms in Acute Leukemia: A Systematic Narrative Review. Acta Haematologica. Karger.
[6]. Zhen, G., Fu, Y., Zhang, C., Ford, N. C., Wu, X., Wu, Q., and Yan, D. 2022. Mechanisms of Bone Pain: Progress in Research from Bench to Bedside. Bone Research. 10(1): 45. Nature Publishing Group.
[7]. Heymann, D. 2021. Bone Cancer: Bone Sarcomas and Bone Metastases—From Bench to Bedside. Elsevier.
[8]. Sun, L., Zhou, Y., Yin, M., Wang, F., and Yang, L. 2025. Approaches to Modeling Cancer Metastasis: From Bench to Bedside. Frontiers in Oncology. Frontiers Media SA.
[9]. Jones, R. J., and Matsui, W. 2007. Cancer Stem Cells: From Bench to Bedside. Biology of Blood and Marrow Transplantation. 13(1): 46–47. Elsevier.
[10]. Tripathi, R., Sharma, A., and Gupta, V. 2022. A Hybrid Deep Learning Model for Improved Bone Marrow Cancer Diagnosis Using MRI and Genomic Data. Artificial Intelligence in Medicine. 132: 102278. Elsevier.DOI: https://doi.org/10.1016/j.artmed.2022.102278
[11]. Meng, F., Al-Khafaji, Z., Muzammil, K., Sarbazjoda, E., et al. 2026. Biomarkers and Advances in AML-MRC: From Bench to Bedside. Annals of Hematology. Springer.
[12]. Huang, S., Huang, S., Huang, X., Feng, X., Wang, R., Liu, D., et al. 2026. Bench-to-Bedside Insights into the Challenges of Immunosuppression in Sepsis. Pathogens. MDPI.
[13]. Talukdar, J., Baishya, D., Karmakar, S., Ghosh, A., et al. 2026. From Bench to Bedside: Innovations and Challenges in CAR-T Cell Therapy Manufacturing and Regulation. Advancement in CAR-T Therapy. Springer.
[14]. Biswas, M., John, S., Murad, M. A., et al. 2026. Pharmacogenomics of Thiopurine Drugs: A Bench-to-Bedside Success Story in Thailand. Clinical and Translational Science. Wiley Online Library.
[15]. Zhang, G. K., Zhou, F. 2026. Exosomes-Loaded with Traditional Chinese Medicine Constituents: A Novel Therapeutic Avenue for Aplastic Anemia. Chinese Medicine. Springer.
[16]. Iqbal, H., Iqbal, M., Aman, J. 2026. Exosomes Therapeutics: Current and Future Perspectives. Immunology Research and Perspectives.
[17]. Convento, M. B., Borges, F. T. 2026. A Comprehensive Review of Engineered Bone Marrow Mesenchymal Stem Cell-Derived Exosomes as Nanotheranostic Platforms for Acute and Chronic Diseases. Journal of Nanotheranostics. MDPI.
[18]. Mohan, C. G., Menon, K. N., Pavithran, K., et al. 2026. Artificial Intelligence in Disease Modeling. Springer Handbook of Artificial Intelligence. Springer.
[19]. Bequette, B. W. 2026. Process Systems Engineering in Precision Medicine: Opportunities in Autologous CAR-T Therapy. Engineering in Life Sciences. Wiley Online Library.
[20]. Foti, R., Storti, G., Palmesano, M., Calicchia, A., et al. 2026. AI-Driven Innovations for Quality Control and Standardization: Future Strategies in Adipose-Derived Stem Cell Manufacturing. International Journal of Molecular Sciences. MDPI.
