{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,30]],"date-time":"2026-10-30T09:39:27Z","timestamp":1793353167099,"version":"4.5.0"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurosci."],"abstract":"<jats:p>Strict privacy regulations in healthcare limit the centralized sharing of medical imaging data, creating challenges for developing robust and generalizable diagnostic models. Federated learning (FL) offers a privacy-preserving paradigm in which participating institutions collaboratively train a global model without directly exchanging patient images. This study presents a federated CNN-LSTM framework for privacy-preserving brain stroke classification from MRI scans, designed to learn discriminative spatial and sequential representations while retaining training data at the local client level. The proposed framework employs a CNN for hierarchical spatial feature extraction and an LSTM to model sequential dependencies among extracted features, followed by federated averaging (FedAvg) to construct the global model from local updates. Experimental evaluation on a private MRI dataset comprising 3,125 scans distributed across three non-IID clients enabled assessment under heterogeneous data distributions. The proposed federated framework achieved an accuracy of 95.20%, compared to 95.95% for centralized training, demonstrating that privacy-preserving decentralized training can achieve competitive diagnostic performance with only a marginal reduction in accuracy. Furthermore, Grad-CAM was employed to provide visual explanations of the model\u2019s predictions by highlighting the image regions that contribute to stroke classification. In five-fold cross-validation, the Federated CNN-LSTM model achieved an average accuracy of 92.10% \u00b1 1.50%, indicating consistent performance across folds. The findings demonstrate the feasibility of combining federated learning, hybrid CNN-LSTM representation learning, and model interpretability for privacy-preserving MRI-based stroke classification, while highlighting the need for multi-center validation and larger federated deployments before clinical translation.<\/jats:p>","DOI":"10.3389\/fnins.2026.1847277","type":"journal-article","created":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T05:36:04Z","timestamp":1789968964000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Federated deep learning for stroke detection: a privacy-preserving, explainable CNN-LSTM framework"],"prefix":"10.3389","volume":"20","author":[{"given":"Reeta","family":"Samuel","sequence":"first","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thanapal","family":"Pandi","sequence":"additional","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"key":"B1","first-page":"308","article-title":"Deep learning with differential privacy","volume-title":"Proceedings of the 23rd ACM Conference on Computer and Communications Security (CCS)","author":"Abadi","year":"2016"},{"key":"B2","doi-asserted-by":"publisher","first-page":"100145","DOI":"10.1016\/j.neuri.2023.100145","article-title":"Automatic brain ischemic stroke segmentation with deep learning: A review.","volume":"3","author":"Abbasi","year":"2023","journal-title":"Neurosci. Inform."},{"key":"B3","doi-asserted-by":"publisher","first-page":"1660479","DOI":"10.3389\/fradi.2025.1660479","article-title":"U-FDL-PPE: A unified federated deep learning framework with privacy-preserving explainability for early and accurate viral disease prediction.","volume":"5","author":"Agrawal","year":"2025","journal-title":"Front. Radiol"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1333","DOI":"10.1109\/TIFS.2017.2787987","article-title":"Privacy-preserving deep learning via additively homomorphic encryption.","volume":"13","author":"Aono","year":"2017","journal-title":"IEEE Trans. Inf. Forensics Secur."},{"key":"B5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1140\/epjs\/s11734-025-01516-z","article-title":"Federated learning for secure medical MRI brain tumor image classification.","volume":"234","author":"Appasami","year":"2025","journal-title":"Eur. Phys. J. Spec. Top"},{"key":"B6","doi-asserted-by":"publisher","first-page":"136","DOI":"10.1016\/j.envpol.2012.05.011","article-title":"Comment on \u201cAbundance of birds in fukushima as judged from chernobyl&quot.","volume":"169","author":"Beresford","year":"2012","journal-title":"Environ. Pollut."},{"key":"B7","doi-asserted-by":"publisher","first-page":"4381","DOI":"10.1109\/TNSM.2024.3418786","article-title":"A privacy-preserving federated learning framework with lightweight and fair in iot.","volume":"21","author":"Chen","year":"2024","journal-title":"IEEE Trans. Netw. Serv. Manag."},{"key":"B8","doi-asserted-by":"publisher","first-page":"80","DOI":"10.1007\/s40808-024-02094-0","article-title":"Innovative approaches to surface water quality management: Advancing nitrate (NO3) forecasting with hybrid CNN-LSTM and CNN- GRU techniques.","volume":"11","author":"Davoudi","year":"2025","journal-title":"Model. Earth Syst. Environ."},{"key":"B9","doi-asserted-by":"publisher","first-page":"24559","DOI":"10.1007\/s11227-024-06309-0","article-title":"Convolutional neural network (CNN) and federated learning-based privacy preserving approach for skin disease classification.","volume":"80","author":"Divya","year":"2024","journal-title":"J. Supercomput"},{"key":"B10","first-page":"21","article-title":"Privacy and security threat models and mitigation strategies of older adults","volume-title":"Proceedings 15th Symposium on Usable Privacy and Security","author":"Frik","year":"2019"},{"key":"B11","doi-asserted-by":"publisher","first-page":"135040","DOI":"10.1109\/ACCESS.2024.3462171","article-title":"FedADC: Federated average knowledge distilled mutual conditional learning (FedADC) for waste classification.","volume":"12","author":"Ghosh","year":"2024","journal-title":"IEEE Access."},{"key":"B12","doi-asserted-by":"crossref","first-page":"770","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep residual learning for image recognition","volume-title":"Proceedings of the 29th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016)","author":"He","year":"2016"},{"key":"B13","doi-asserted-by":"publisher","first-page":"19194","DOI":"10.1007\/s11227-023-05497-5","article-title":"A CNN-BiLSTM model with attention mechanism for earthquake prediction.","volume":"79","author":"Kavianpour","year":"2023","journal-title":"J. Supercomput."},{"key":"B14","doi-asserted-by":"publisher","first-page":"2021","DOI":"10.1109\/TII.2021.3098010","article-title":"A federated learning based privacy-preserving smart healthcare system.","volume":"18","author":"Li","year":"2021","journal-title":"IEEE Trans. Ind. Inform."},{"key":"B15","first-page":"93","article-title":"Brain MR image segmentation for tumor detection using artificial neural networks.","volume":"6","author":"Londhe","year":"2017","journal-title":"Int. J. Comput. Sci. Netw."},{"key":"B16","doi-asserted-by":"publisher","first-page":"175","DOI":"10.1007\/s00234-009-0633-5","article-title":"Computed tomography in acute ischemic stroke.","volume":"52","author":"L\u00f6vblad","year":"2010","journal-title":"Neuroradiology"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1535478","DOI":"10.3389\/fonc.2025.1535478","article-title":"Explainable AI in medical imaging: An interpretable and collaborative federated learning model for brain tumor classification.","volume":"15","author":"Mastoi","year":"2025","journal-title":"Front. Oncol"},{"key":"B18","doi-asserted-by":"publisher","first-page":"27024","DOI":"10.1109\/ACCESS.2025.3538773","article-title":"Reconstruction and classification of brain strokes using deep learning-based microwave imaging.","volume":"13","author":"Mousavi","year":"2025","journal-title":"IEEE Access."},{"key":"B19","doi-asserted-by":"publisher","first-page":"3206","DOI":"10.1109\/JBHI.2024.3375894","article-title":"Explainable federated medical image analysis through causal learning and blockchain.","volume":"28","author":"Mu","year":"2024","journal-title":"IEEE J. Biomed. Health Inform"},{"key":"B20","first-page":"256","article-title":"Contrastive re-localization and history distillation in federated CMR segmentation","volume-title":"MICCAI","author":"Qi","year":"2022"},{"key":"B21","doi-asserted-by":"publisher","first-page":"73145","DOI":"10.1109\/ACCESS.2022.3190798","article-title":"Applying combined action observation and motor imagery to enhance classification performance in a brain\u2013computer interface system for stroke patients.","volume":"10","author":"Rungsirisilp","year":"2022","journal-title":"IEEE Access."},{"key":"B22","doi-asserted-by":"publisher","first-page":"31215","DOI":"10.1038\/s41598-025-14358-5","article-title":"Optimizing brain stroke detection with a weighted voting ensemble machine learning model.","volume":"15","author":"Samuel","year":"2025","journal-title":"Sci. Rep"},{"key":"B23","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/ISAECT53699.2021.9668454","article-title":"Evaluation of transfer learning based deep learning architectures for waste classification","volume-title":"2021 4th International Symposium on Advanced Electrical and Communication Technologies (ISAECT)","author":"Singh","year":"2021"},{"key":"B24","doi-asserted-by":"crossref","first-page":"1018","DOI":"10.1109\/TENSYMP50017.2020.9230618","article-title":"Brain tumor classification with inception network based deep learning model using transfer learning","volume-title":"2020 IEEE Region 10 Symposium (TENSYMP)","author":"Soumik","year":"2020"},{"key":"B25","doi-asserted-by":"publisher","first-page":"100530","DOI":"10.1016\/j.eij.2024.100530","article-title":"Enhancing medical image classification via federated learning and pre-trained model.","volume":"27","author":"Srinivasu","year":"2024","journal-title":"Egypt. Inform. J."},{"key":"B26","doi-asserted-by":"publisher","first-page":"3763","DOI":"10.1109\/JIOT.2022.3151982","article-title":"Privacy-preserving ensemble infused enhanced deep neural network framework for edge cloud convergence.","volume":"10","author":"Stephanie","year":"2022","journal-title":"IEEE Internet Things J."},{"key":"B27","doi-asserted-by":"publisher","first-page":"186","DOI":"10.1016\/j.eswa.2015.08.016","article-title":"A non-invasive methodology for the grade identification of astrocytoma using image processing and artificial intelligence techniques.","volume":"43","author":"Subashini","year":"2016","journal-title":"Expert Syst. Appl."},{"key":"B28","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1007\/s13735-021-00218-1","article-title":"A review on deep learning in medical image analysis.","volume":"11","author":"Suganyadevi","year":"2022","journal-title":"Int. J. Multimed. Inf. Retr"},{"key":"B29","doi-asserted-by":"publisher","first-page":"701","DOI":"10.1016\/j.nimb.2009.10.012","article-title":"AMS measurement technique after 30 years: Possibilities and limitations of low energy systems.","volume":"268","author":"Synal","year":"2010","journal-title":"Nucl. Instrum. Methods Phys. Res. Sect. B"},{"key":"B30","doi-asserted-by":"publisher","first-page":"16301","DOI":"10.1109\/ACCESS.2024.3490693","article-title":"Neuro-VGNB: Transfer learning based approach for detecting brain stroke.","volume":"12","author":"Tanveer","year":"2024","journal-title":"IEEE Access."},{"key":"B31","doi-asserted-by":"publisher","first-page":"103768","DOI":"10.1016\/j.jnca.2023.103768","article-title":"RPIFL: Reliable and privacy preserving federated learning for the internet of things.","volume":"221","author":"Wang","year":"2024","journal-title":"J. Netw. Comput. Appl."},{"key":"B32","first-page":"559","article-title":"Differentially-private deep learning from an optimization perspective","volume-title":"Proceedings IEEE Conference on Computer Communications (INFOCOM)","author":"Xiang","year":"2016"},{"key":"B33","doi-asserted-by":"publisher","first-page":"55129","DOI":"10.1007\/s11356-021-14687-8","article-title":"A study on water quality prediction by a hybrid CNN-LSTM model with attention mechanism.","volume":"28","author":"Yang","year":"2021","journal-title":"Environ. Sci. Pollut. Res. Int"},{"key":"B34","doi-asserted-by":"publisher","first-page":"126791","DOI":"10.1016\/j.neucom.2023.126791","article-title":"FedBrain: A robust multi-site brain network analysis framework based on federated learning for brain disease diagnosis.","volume":"559","author":"Zhang","year":"2023","journal-title":"Neurocomputing"},{"key":"B35","doi-asserted-by":"publisher","first-page":"15884","DOI":"10.1109\/JIOT.2021.3056185","article-title":"Dynamic-fusion-based federated learning for COVID-19 detection.","volume":"8","author":"Zhang","year":"2021","journal-title":"IEEE Internet Things J"}],"container-title":["Frontiers in Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnins.2026.1847277\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T05:36:06Z","timestamp":1789968966000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnins.2026.1847277\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":35,"alternative-id":["10.3389\/fnins.2026.1847277"],"URL":"https:\/\/doi.org\/10.3389\/fnins.2026.1847277","relation":{},"ISSN":["1662-453X"],"issn-type":[{"value":"1662-453X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]},"article-number":"1847277"}}