{
    "@context": "https://schema.org",
    "@type": "ScholarlyArticle",
    "identifier": {
        "jsc_id": 30,
        "DOI": "10.5281/zenodo.23257468",
        "DOI_url": "https://doi.org/10.5281/zenodo.23257468"
    },
    "title": "Federated Learning with Differential Privacy for Healthcare IoT: A Privacy-Preserving Framework for Collaborative Clinical Modeling",
    "abstract": "Wearable sensors and connected medical devices have made healthcare Internet of Things (IoT) systems a rich and highly sensitive source of clinical data. Centralizing these data for model training raises privacy, regulatory, and security concerns. This paper proposes a privacy-preserving framework that combines federated learning with client-level differential privacy, enabling hospitals and device silos to train a shared clinical model without exchanging raw patient records. Clients train locally; their updates are clipped and perturbed with calibrated Gaussian noise, and only the protected updates are aggregated by a central server using federated averaging. We formalize the privacy accounting, present the training procedure as pseudocode, and evaluate the framework on a simulated non-independent and identically distributed (non-IID) healthcare dataset spanning ten clients. The experiments map the trade-off between the privacy budget and predictive utility, and show that strong privacy guarantees can be achieved at a modest, controllable cost to accuracy. The results indicate that differentially private federated learning is a viable approach for edge-based healthcare IoT settings in which utility, security, and compliance must be balanced.",
    "authors": [
        {
            "sequence": "first",
            "given": "Esmaeil",
            "family": "Shayesteh",
            "name": "Esmaeil Shayesteh",
            "affiliation": [
                {
                    "name": "Department of Industrial Engineering, CT.C., Islamic Azad University, Tehran"
                }
            ],
            "email": "es.shayesteh@iau.ir"
        },
        {
            "sequence": "additional",
            "given": "Vasileios",
            "family": "Lymperopoulos",
            "name": "Vasileios Lymperopoulos",
            "affiliation": [
                {
                    "name": "CEO, LiberoGroup, Greece"
                }
            ],
            "email": "vasileios.lymperopoulos@liberoservices.org",
            "is_corresponding": true
        }
    ],
    "keywords": [
        "Federated learning",
        "Differential privacy",
        "Healthcare IoT",
        "Privacy-preserving Machine Learning",
        "Edge computing"
    ],
    "section": "Research Article",
    "language": "en",
    "journal": {
        "name": "Journal of Smart Computing",
        "abbreviation": "JSC",
        "issn": "2979-8752",
        "publisher": "Center of Intelligent Operations and Technology Research, United Kingdom",
        "url": "https://ciotr.com/jsc/"
    },
    "issue": {
        "volume": 2,
        "number": 1,
        "year": 2026,
        "url": "https://ciotr.com/jsc/issue/8"
    },
    "pages": {
        "first": "1",
        "last": "10",
        "range": "1-10"
    },
    "dates": {
        "submitted": "2026-07-12",
        "published": "2026-02-05"
    },
    "urls": {
        "landing_page": "https://ciotr.com/jsc/article/30",
        "pdf": "https://ciotr.com/jsc/pdf/article-30.pdf",
        "metadata": "https://ciotr.com/jsc/article/30.json"
    },
    "license": {
        "name": "Creative Commons Attribution 4.0 International",
        "code": "CC BY 4.0",
        "url": "https://creativecommons.org/licenses/by/4.0/"
    },
    "citation": {
        "apa": "Shayesteh, E., & Lymperopoulos, V. (2026). Federated Learning with Differential Privacy for Healthcare IoT: A Privacy-Preserving Framework for Collaborative Clinical Modeling. Journal of Smart Computing, 2(1), 1-10. https://doi.org/10.5281/zenodo.23257468"
    }
}