A multimodal deep learning framework for smart health monitoring and early-stage classification of chronic kidney disease
Chronic kidney disease (CKD) is a disease that gets worse over time and raises the risk of heart disease, lowers the quality of life, and puts a lot of strain on the healthcare system. Conventional diagnostic methods predominantly depend on static clinical biomarkers, including serum creatinine, estimated glomerular filtration rate, and albuminuria, which may not effectively detect subtle physiological decline in the initial stages. To fill this gap, this paper suggests a strong, easy-to-understand, and scalable multimodal deep learning framework that combines structured clinical laboratory features with simulated real-time physiological signals, such as changes in blood pressure and heart rate. A dual-branch neural architecture is constructed, comprising a multilayer perceptron that encodes tabular clinical features and a one-dimensional convolutional neural network that extracts temporal representations from vital-sign sequences. For CKD stage-wise classification, the learned embeddings are combined at the feature level. The proposed model achieves a test accuracy of 90% and a macro-F1 score of 0.86, outperforming clinical-only and vitals-only baselines, as shown by experimental evaluation. SHapley Additive exPlanations-based feature attribution and temporal attention visualizations make it possible to understand models better by showing clinically relevant biomarkers and hemodynamic patterns. Robustness testing with noisy vital signs and low-latency inference profiling shows that the framework is good for real-time renal risk monitoring and tele-nephrology uses.
Alsekait, D. M., Saleh, H., Gabralla, L. A., Alnowaiser, K., El-Sappagh, S., Sahal, R., & El-Rashidy, N. (2023). Toward comprehensive chronic kidney disease prediction based on ensemble deep learning models. Applied Sciences, 13(6), 3937. https://doi.org/10.3390/app13063937
Balikci Cicek, I., & Kucukakcali, Z. (2023). Explanaible artificial intelligence method SHAP's prediction of risk factors associated with chronic kidney disease combined with black box methods. Journal of Community Medicine and Public Health Reports, 4(10). https://doi.org/10.38207/jcmphr/2023/nov041004138
Demrozi, F., Farmanbar, M., & Engan, K. (2025). Multimodal AI (MMAI) for next-generation healthcare: Data domains, algorithms, challenges, and future perspectives. Current Opinion in Biomedical Engineering, 100632. https://doi.org/10.1016/j.cobme.2025.100632
Ghosh, S. K., & Khandoker, A. H. (2024). Investigation on explainable machine learning models to predict chronic kidney diseases. Scientific Reports, 14(1), 3687. https://doi.org/10.1038/s41598-024-54375-4
Halder, R. K., Uddin, M. N., Uddin, M. A., Aryal, S., Saha, S., Hossen, R., Ahmed, S., Rony, M. A. T., & Akter, M. F. (2024). ML-CKDP: Machine learning-based chronic kidney disease prediction with smart web application. Journal of Pathology Informatics, 15, 100371. https://doi.org/10.1016/j.jpi.2024.100371
Hegde, G., Shenoy, P. D., & Canchi, A. (2025). A deep learning framework for chronic kidney disease stage classification. Healthcare Analytics, 7, 100398. https://doi.org/10.1016/j.health.2025.100398
Iliyas, I. I., Boukari, S., & Gital, A. Y. U. (2025). Recent trends in prediction of chronic kidney disease using different learning approaches: A systematic literature review. Journal of Medical Artificial Intelligence, 8, 62. https://doi.org/10.21037/jmai-24-256
Islam, M. A., Majumder, M. Z. H., & Hussein, M. A. (2023). Chronic kidney disease prediction based on machine learning algorithms. Journal of Pathology Informatics, 14, 100189. https://doi.org/10.1016/j.jpi.2023.100189
Jager, K. J., Kovesdy, C., Langham, R., Rosenberg, M., Jha, V., & Zoccali, C. (2019). A single number for advocacy and communication—Worldwide more than 850 million individuals have kidney diseases. Kidney International, 96(5), 1048–1050. https://doi.org/10.1016/j.kint.2019.07.012
Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. (2024). KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney International, 105(4S), S117–S314. https://doi.org/10.1016/j.kint.2023.10.018
Khan, N., Raza, M. A., Mirjat, N. H., Balouch, N., Abbas, G., Yousef, A., & Touti, E. (2024). Unveiling the predictive power: A comprehensive study of machine learning model for anticipating chronic kidney disease. Frontiers in Artificial Intelligence, 6, 1339988. https://doi.org/10.3389/frai.2023.1339988
Kotula, C. A., Martin, J., Carey, K. A., Edelson, D. P., Dligach, D., Mayampurath, A., Afshar, M., & Churpek, M. M. (2025). Comparison of multimodal deep learning approaches for predicting clinical deterioration in ward patients: Observational cohort study. Journal of Medical Internet Research, 27, e75340. https://doi.org/10.2196/75340
Li, J., Du, X., Zhang, R., Li, X., Xu, J., Song, X., Zhao, Y., Liu, L., Xu, G., & Fan, Y. (2025). Machine learning models for predicting short-term progression in patients with stage 4 chronic kidney disease: A multi-center validation study. Scientific Reports, 15(1), 39285. https://doi.org/10.1038/s41598-025-23037-4
Lu, D., Wu, S., & Huang, X. (2025). Multimodal health risk prediction system for chronic diseases via vision-language fusion and large language models. In Human-Computer Interaction and Natural Language Processing (ICAHN) (pp. 76–79). IEEE. https://doi.org/10.1109/icahn67688.2025.00022
Chandra Sekhar, Pasam Naga Kavitha, B. N. V. Uma Shankar, Amrutavalli Aakula, & Divya Boya. (2025). MedFusionAI: A deep learning framework for multi-modal health data fusion to predict chronic disease risks. International Journal of Computational and Experimental Science and Engineering, 11(2). https://doi.org/10.22399/ijcesen.2159
Rostamian, S., Kaura, A., Ariti, C., Gupta, A., Whitely, W., Poulter, N. R., & Sever, P. (2025). The impact of blood pressure variability on cardiovascular and renal outcomes in patients with and without diabetes: Insights from the Anglo-Scandinavian Cardiac Outcomes Trial. Journal of Hypertension, 43(12), 2033–2041. https://doi.org/10.1097/HJH.0000000000004153
Ruan, Y., Chen, Y., Huang, N., Wang, D., Xu, Y., Fan, J., Chen, W., & Wang, X. (2025). Long-term BPV is an independent risk factor for renal prognosis in hypertensive patients — A post-hoc analysis of the SPRINT study. International Journal of Medical Sciences, 22(10), 2298–2307. https://doi.org/10.7150/ijms.111843
Saif, D., Sarhan, A. M., & Elshennawy, N. M. (2024). Deep-kidney: An effective deep learning framework for chronic kidney disease prediction. Health Information Science and Systems, 12(1), 3. https://doi.org/10.1007/s13755-023-00261-8
Shu, P., Wang, X., Wen, Z., Chen, J., & Xu, F. (2025). SHAP combined with machine learning to predict mortality risk in maintenance hemodialysis patients: A retrospective study. Frontiers in Medicine, 12, 1615950. https://doi.org/10.3389/fmed.2025.1615950
Tisha, S. A., Bhattacharjee, B., Ahmed, M. P., Mukaddim, A. A., Hider, M. A., & Khan, M. T. (2025). Deep learning-driven early detection and risk prediction of chronic diseases using multimodal healthcare data. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.5784022
Tsai, M.-C., Lojanapiwat, B., Chang, C.-C., Noppakun, K., Khumrin, P., Li, S.-H., Lee, C.-Y., Lee, H.-C., & Khwanngern, K. (2023). Risk prediction model for chronic kidney disease in Thailand using artificial intelligence and SHAP. Diagnostics, 13(23), 3548. https://doi.org/10.3390/diagnostics13233548
Wang, G., Ma, K., Ma, Z., Guo, X., Wang, Y., Ma, L., Qi, C., Li, Y., & Zhou, X. (2022). Short-term blood pressure variability and outcomes in non-dialysis chronic kidney disease. Frontiers in Medicine, 9, 911205. https://doi.org/10.3389/fmed.2022.911205
Xiang, J. Z., Wang, Q. Y., Fang, Z. B., Esquivel, J. A., & Su, Z. X. (2025). A multi-modal deep learning approach for stress detection using physiological signals: Integrating time and frequency domain features. Frontiers in Physiology, 16, 1584299. https://doi.org/10.3389/fphys.2025.1584299
Zhu, Y., Bi, D., Saunders, M., & Ji, Y. (2023). Prediction of chronic kidney disease progression using recurrent neural network and electronic health records. Scientific Reports, 13(1), 22091. https://doi.org/10.1038/s41598-023-49271-2