The threat model the researchers target is specific and technically subtle. In a software-defined network (SDN), a central controller manages traffic using the OpenFlow protocol, deciding how packets flow between IoT devices and hospital servers. A zero-day DDoS attack exploits undisclosed vulnerabilities in this arrangement: compromised medical IoT devices flood the controller with oversized Packet-in messages, often disguised as urgent traffic, until its CPU and memory are exhausted and legitimate users are locked out. Because these attacks are novel by definition, conventional signature-based detection systems simply have nothing in their databases to match against. Hospitals, the authors note, often lack the cyber threat intelligence capacity to identify such new attack patterns at all, making the problem acute in healthcare settings where downtime can cost lives.

The core of the new framework is a tightly coupled combination of two deep learning architectures with complementary strengths. The first is a variational autoencoder (VAE), an unsupervised generative model that learns a probabilistic representation of normal network traffic. Unlike a traditional autoencoder, which compresses each input into a single point in a latent space, a VAE maps inputs to a distribution characterized by mean and variance, using a KL-divergence regularization term to keep that latent space smooth and continuous. During detection, the model encodes incoming traffic, samples latent vectors via the reparameterization trick, and attempts to reconstruct the original input. If the reconstruction error is high, or the latent representation strays far from the learned distribution of benign traffic, the input is flagged as anomalous. This generative approach is precisely what makes zero-day detection possible: the model does not need to have seen an attack before, only to recognize that the traffic deviates statistically from normal behavior.

On its own, however, a VAE has a blind spot. DDoS attacks evolve over time, shifting intensity and behavior across minutes and hours, and a purely reconstruction-based model can miss these temporal dynamics. The researchers therefore feed sequences of latent vectors, built over a sliding time window, into a long short-term memory (LSTM) network. The LSTM’s gated memory cells, governed by input, forget, and output gates, learn long-range temporal dependencies in traffic patterns and can spot irregularities that unfold across sequences rather than within single snapshots. When the two models disagree, for instance when the VAE flags traffic as malicious but the LSTM scores it as normal, the framework normalizes both losses and combines them into a single hybrid anomaly score weighted by a tunable parameter, balancing unsupervised anomaly detection against supervised classification of known attacks.

Training such a model normally requires pooling enormous volumes of traffic data, and in a hospital that data is laced with sensitive medical information. The team’s answer is federated learning: each medical IoT device trains the hybrid VAE-LSTM model locally on its own data, and only the resulting model parameters, never the raw records, are sent to a central aggregator. Crucially, the authors chose the FedProx aggregation method over the standard Federated Averaging (FedAvg). Medical networks are notoriously heterogeneous, with non-independent and identically distributed (non-IID) data, since a patient’s heartbeat fluctuates with activity and stress, and devices vary widely in computing power. FedProx adds a proximal term to the local loss function that penalizes large deviations from the global model and allows slower devices, or stragglers, to contribute partial work without stalling aggregation, stabilizing convergence across this messy, real-world landscape.

Even a highly accurate detector is useless if clinicians and security analysts cannot trust it, and deep learning models are notorious black boxes. To address this, the framework incorporates the SHapley Additive exPlanations (SHAP) algorithm, which quantifies how much each input feature contributes to a given prediction. The resulting visualizations, including decision plots, waterfall plots, and summary plots, reveal which traffic characteristics push a prediction toward an attack alert. On the IoT healthcare dataset, features such as MQTT header flags, TCP push flags, and packet rate drove predictions toward the attack zone, while flow duration and HTTP request rate pushed toward benign classifications. On the CICDDoS2019 dataset, flow bytes per second, packet size consistency, and inter-arrival jitter emerged as the dominant indicators, exactly the volumetric and temporal signatures expected of DDoS behavior. This feature-level transparency lets analysts attribute attacks to specific protocols rather than generic anomalies, supporting auditing, debugging, and regulatory compliance.

The experimental evaluation was conducted in a simulated SDN-enabled IoT healthcare environment built with Mininet 2.3 and an ONOS controller, with the detection models implemented in TensorFlow 2.12. The team trained and validated the framework on two datasets: CICDDoS2019, which includes eight known DDoS attack classes and six zero-day attacks withheld from training and introduced only during testing, and an IoT healthcare security dataset. Data was split 60-30-10 for training, testing, and validation, with ten-fold cross-validation and careful preprocessing to prevent data leakage. Against baseline methods including standard autoencoders, standalone LSTMs, CNNs, and MLPs, the hybrid model achieved 99.75 percent accuracy and 99.63 percent precision on known attacks, and, more impressively, 98.61 percent accuracy and 98.90 percent precision on the zero-day attacks it had never encountered. On the healthcare dataset, it reached 99.21 percent accuracy on known attacks and 98.1 percent accuracy on zero-day variants.

The statistical rigor of the evaluation strengthens these claims. Paired t-tests across the cross-validation folds showed that the hybrid model’s superiority over every baseline was significant at p < 0.0001, with exact p-values ranging from 1.77 × 10⁻¹¹ to 7.86 × 10⁻⁷ and Cohen’s d effect sizes exceeding 4.0 in all comparisons, indicating substantial practical significance. Against the BFLIDS baseline, the accuracy advantage was 7.33 percent and the precision advantage 8 percent, with 95 percent confidence intervals confirming the gains. The authors translate these numbers into clinical terms: the accuracy improvement corresponds to correctly identifying roughly 160 additional attacks per 1,000 attempts, potentially protecting ventilators and electronic health records from compromise, while the precision gains reduce false alarms and the alert fatigue that plagues hospital security teams.

Scalability results add a practical dimension that many federated learning studies omit. Simulated deployments of 250 and 500 medical IoT devices, with a realistic 30 percent participation rate per round, showed that FedProx-based aggregation cut communication costs by about 17 percent compared with baseline aggregation methods, reaching 106.7 gigabytes, while converging in 118 rounds and consuming roughly 40 percent less energy per round and 20 percent less aggregation time than alternatives. The method earned the highest feasibility score of 0.89 in a weighted composite metric, and performance degraded only moderately as device counts grew, suggesting robustness to the heterogeneity of real hospital networks. The authors caution that the work remains simulation-based and outline future directions: validation on larger multi-institution datasets such as TON_IoT and Edge-IIoT, deployment on real edge hardware to measure CPU load, memory footprint, and latency, and extension of the framework to other threats including spoofing and man-in-the-middle attacks. For now, the study offers a template for how generative anomaly detection, privacy-preserving federated training, and explainable AI can be combined into a single defensive stack, one that may prove essential as healthcare’s Industry 5.0 ambitions collide with an increasingly hostile internet.

Subject of Research: A privacy-preserving, explainable hybrid deep learning framework for detecting zero-day DDoS attacks in SDN-enabled IoT healthcare networks

Article Title: A privacy preserving and trustworthy hybrid deep learning model for zero-day DDoS detection in healthcare industry 5.0

Article References: Ali, A., Khan, M. F., Azim, N., & Aadil, F. (2026). A privacy preserving and trustworthy hybrid deep learning model for zero-day DDoS detection in healthcare industry 5.0. Cluster Computing, 29(13), Article 750. https://doi.org/10.1007/s10586-026-06463-5

Image Credits: AI Generated

DOI: 10.1007/s10586-026-06463-5

Keywords: zero-day DDoS, federated learning, variational autoencoder, LSTM, explainable AI, SHAP, software-defined networking, healthcare IoT, Industry 5.0, intrusion detection, FedProx, privacy preservation

Federated learning, first popularized by researchers at Google in 2017, flips the conventional training pipeline on its head. Instead of moving data to the model, the model moves to the data. Each device—or client—trains a local copy of the algorithm on its own measurements and then transmits only the resulting parameter updates, typically small numerical gradients, to a coordinating server. The server averages these updates into a shared global model and sends it back down for another round of local training. Because raw sensor readings never leave the device, the approach reduces direct exposure of sensitive information while slashing the volume of traffic that must cross the network. For IoT deployments, where a single greenhouse or smart building may host dozens of constrained devices, that distinction is not a luxury but a prerequisite.

What makes the Moroccan team’s contribution distinctive is where the federation logic lives. The OneM2M standard is an internationally recognized service layer designed to sit between heterogeneous IoT devices and applications, providing common primitives for registration, data storage, subscription, and notification regardless of whether devices speak HTTP, MQTT, or CoAP. Rather than bolting federated learning onto an ad hoc middleware stack, FL-OneM2M embeds it into this operational backbone, turning OneM2M into an orchestration infrastructure for decentralized intelligence. The authors argue that this integration addresses a genuine gap: most federated learning research assumes idealized communication conditions, while most IoT standards work ignores machine learning entirely. By unifying the two, the framework promises both quality-of-service awareness and scalable learning in a single architecture.

The experimental setup drew on data generated from IoT agricultural deployments, collected through Azure IoT devices and processed via OneM2M-based data injectors. The raw sensor streams included humidity, temperature, environmental plant indicators, and a suite of agricultural growth metrics, partitioned across multiple simulated greenhouse clients. On top of this dataset, the researchers deployed logistic-regression clients and compared two canonical federated optimization algorithms. Federated Averaging, known as FedAvg, is the simplest and most widely used scheme: clients train locally and the server weights each update by client data volume. Federated Proximal, or FedProx, adds a proximal term to the local objective that penalizes clients whose local updates drift too far from the global model, a modification intended to stabilize learning when devices are heterogeneous or perform varying amounts of local work.

The results, reported for an independently and identically distributed—or IID—data setting, tell a nuanced story. FedAvg reached a final accuracy of 80 percent, a gain of 50 percentage points over its starting point, and achieved a higher peak accuracy than FedProx, which topped out at 75 percent. FedProx, however, exhibited smoother loss convergence, suggesting that its regularization term does damp oscillations even when it costs some final performance on this particular problem. The authors are careful to frame these numbers as trends observed under their specific conditions rather than universal verdicts. With a single model class, a single dataset, and at most ten clients, the study deliberately scopes its claims, and the paper discusses the resulting limitations—large-scale behavior, statistical heterogeneity across clients, and security—explicitly rather than glossing over them.

Perhaps the most striking findings concern communication efficiency, the resource that often determines whether federated learning is practical at the edge at all. Under the study’s test conditions, switching the transport protocol from HTTP to MQTT reduced the average round-trip time by 98.4 percent, and the system reached peak accuracy roughly 40 training rounds earlier. The explanation lies in protocol design. HTTP is a request-response protocol that carries substantial header overhead and requires a new exchange for every message, whereas MQTT is a lightweight publish-subscribe protocol built for constrained networks, using minimal headers and persistent connections. In a federated loop, where the server and clients exchange model updates hundreds of times, shaving milliseconds off every round compounds into a dramatic acceleration of the entire training process. For anyone deploying learning across battery-powered sensors, that difference can separate a viable system from an intractable one.

Scalability tests at small scale—three, five, and ten clients—added another practical insight: five clients produced the most stable convergence in the studied configuration. This counterintuitive result reflects a real tension in federated systems. With very few clients, each local update carries outsized weight, so a single noisy or poorly conditioned client can destabilize the global model. As the client count grows, averaging smooths out individual noise, but coordination overhead rises and, in real deployments, stragglers and dropouts multiply. The team also probed resilience by simulating clients dropping out mid-training and found that the system retained 80 percent accuracy under a 10 percent client-dropout rate, an encouraging signal that the OneM2M-based orchestration can absorb a degree of real-world unreliability without collapsing.

The choice of agriculture as the application domain is more than incidental. Smart farming is emerging as a natural proving ground for federated learning because farms are simultaneously data-rich and privacy-sensitive: yield forecasts, disease detections, and growth metrics carry commercial value, and farmers may be reluctant to surrender raw operational data to third parties. Recent work in the field has explored federated crop disease detection from images, model-pruned yield forecasting, and blockchain-assisted trust mechanisms, all pointing toward the same conclusion that collaborative intelligence is only acceptable to many stakeholders if the data stays local. FL-OneM2M fits squarely into this movement, demonstrating that a standardized service layer can serve as the connective tissue between field sensors and a learning pipeline without a central data lake.

The authors are refreshingly candid about what their study does not establish. Ten clients is a toy scale compared with the thousands or millions of devices a production IoT network might involve, and behavior at large scale—where communication bottlenecks, stragglers, and server load become dominant—remains untested. The IID assumption is similarly generous: real sensor networks are notoriously non-IID, with each greenhouse, factory floor, or city block exhibiting its own statistical fingerprint, a condition known to degrade federated averaging and motivate algorithms like FedProx, SCAFFOLD, and adaptive federated optimization in the broader literature. Security is another open front, since model updates themselves can leak information or be poisoned by malicious clients, and the paper positions these challenges as explicit future work rather than solved problems. The researchers also note that the underlying dataset was generated specifically for the study and contains sensitive operational information, so it is not publicly available, though processed or anonymized subsets can be requested from the corresponding author.

Even within those limits, the study offers a genuinely useful template. It shows that federated learning need not live in bespoke middleware but can ride on an open, interoperable standard that the IoT industry already deploys, and it quantifies how much protocol choice—MQTT over HTTP—matters to the speed of decentralized training. As edge hardware grows more capable and privacy regulation tightens worldwide, the pressure to move intelligence out of the cloud and into the network’s periphery will only intensify. Frameworks like FL-OneM2M suggest a path in which the same infrastructure that routes sensor data today becomes the stage on which devices learn together tomorrow, one carefully averaged model update at a time.

Subject of Research: Federated learning integrated with the OneM2M service layer for quality-of-service-aware, scalable machine learning in Internet of Things networks

Article Title: A novel federated learning approach for enhancing QoS and scalability in IoT networks

Article References: Et-Tousy, J., & Zyane, A. (2026). A novel federated learning approach for enhancing QoS and scalability in IoT networks. Cluster Computing, 29(14), Article 825. https://doi.org/10.1007/s10586-026-06633-5

Image Credits: AI Generated

DOI: 10.1007/s10586-026-06633-5

Keywords: federated learning, Internet of Things, OneM2M, edge intelligence, MQTT, quality of service, FedAvg, FedProx, smart agriculture, privacy-preserving machine learning, scalability, Cluster Computing

The research, led by Suresh Arumugam of Dayananda Sagar University in Bengaluru together with colleagues from three other Indian institutions, introduces FedXAI-Health, a federated learning framework designed for multi-label thoracic disease classification. In federated learning, the model travels to the data rather than the other way around: each participating hospital trains a shared neural network on its own machines, and only the resulting model updates, never the patient images themselves, are sent to a central server for aggregation. The framework was tested in a simulated network of five hospital clients, each holding a differently distributed slice of the publicly available NIH Chest X-ray 14 dataset.

That heterogeneity is the crux of the problem the researchers set out to solve. In real-world federated networks, no two hospitals see the same mix of diseases, scanner types, or patient demographics, a condition known in machine learning as non-IID data. When local datasets diverge this sharply, the averaged model produced by standard federated algorithms can drift away from what any individual site needs. To simulate this faithfully, the team partitioned the data using Dirichlet distributions with concentration parameters ranging from 0.1, which produces extremely skewed client datasets, to 5.0, which approaches a more balanced split, and validated robustness across the entire spectrum.

The headline result is genuinely counterintuitive. Under identical experimental conditions, using an EfficientNetB0 backbone, a batch size of 512, and the Adam optimiser with a weight decay of 10 to the power of minus 4, the classic federated averaging algorithm known as FedAvg achieved a Macro-AUC of 0.8060, plus or minus 0.0048. That figure not only held up under severe non-IID conditions but actually exceeded the centralised baseline, in which all data would have been pooled on one server, by 2.0 percent. The centralised EfficientNetB0 model managed only 0.7859, plus or minus 0.0142. In other words, the privacy-preserving approach did not merely match the gold standard it is usually measured against; it beat it.

The authors attribute part of this advantage to a phenomenon familiar from the regularisation literature: training across heterogeneous sites forces the model to encounter a wider effective variety of disease presentations during optimisation, which can act as an implicit form of data augmentation and reduce overfitting to any single distribution. The team also benchmarked FedProx, a popular federated variant that adds a proximal term to keep local models from straying too far from the global one. FedProx reached a Macro-AUC of 0.7879, plus or minus 0.0020, respectable but below FedAvg in this setting.

That gap prompted one of the study’s most practically useful findings, an ablation study across five values of FedProx’s regularisation coefficient, mu, spanning 0, 0.001, 0.01, 0.1, and 1.0. Performance degraded monotonically as the coefficient increased, meaning the stronger the algorithm tried to tether local models to the global average, the worse the multi-label classification became. The lesson for practitioners is that minimal regularisation is optimal for multi-label medical imaging under these conditions, a direct challenge to the intuition that heterogeneous data always demands stronger constraints. For clinical teams designing federated deployments, this kind of tuning guidance can save months of trial and error.

Accuracy alone, however, is not enough to earn the trust of radiologists, and this is where the explainability component of FedXAI-Health becomes central. The framework pairs its federated training pipeline with two complementary attribution techniques: SHAP, which assigns each input pixel a contribution score for a given prediction, and Grad-CAM, which produces class-specific heatmap localisations over the image. The researchers compared the maps generated by both methods and confirmed that they produced clinically coherent, disease-specific localisations, meaning the models appeared to base their diagnoses on the actual anatomical regions associated with each condition rather than on spurious correlations such as hospital-specific imaging artefacts.

This matters because multi-label thoracic classification is a genuinely hard task. The NIH Chest X-ray 14 dataset covers fourteen distinct conditions, including cardiomegaly, pleural effusion, pneumonia, and pneumothorax, and a single radiograph can carry several of them simultaneously. A model that flags a disease without showing where it saw the evidence is of limited clinical value, and a federated model whose explanations differ wildly across sites would be even harder to trust. By demonstrating that attribution maps remain coherent across a decentralised, heterogeneous network, the study addresses the trustworthiness question at the same time as the privacy one, rather than treating explainability as an afterthought bolted onto a finished model.

The experimental hygiene underlying these claims deserves attention as well. All results are reported as means with standard deviations across three independent runs, and the dataset was split at the patient level between training and test sets, ensuring zero data leakage, a persistent pitfall in medical imaging research where multiple images of the same patient can silently straddle both sets. The team also ran a supplementary experiment on CIFAR-10 to contextualise the findings beyond the medical domain. Crucially for reproducibility, the complete source code, trained model checkpoints, experiment logs, and result files have been released publicly on GitHub, and the NIH dataset itself is openly available.

The broader significance of the work lies in what it suggests about the future of clinical AI. Federated learning has already been deployed in real hospital networks, most famously in a Nature Medicine study predicting clinical outcomes in COVID-19 patients across twenty institutions, and surveys of the field consistently identify non-IID heterogeneity as its central unsolved challenge. By showing that a well-tuned, lightweight federated setup can outperform centralised training on a benchmark thoracic dataset while producing verifiable explanations, FedXAI-Health strengthens the case that hospitals need not choose between collaboration and confidentiality. The code is open, the data is public, and the recipe is documented, which means other research groups can stress-test the results on their own distributions. If the findings generalise beyond simulated clients to genuine multi-site deployments, the era in which privacy regulations condemned each hospital to train diagnostic AI in isolation may finally be drawing to a close, replaced by networks of institutions that learn together while their patients’ images never leave the building.

Subject of Research: Explainable federated learning for privacy-preserving multi-label thoracic disease classification from chest X-rays under non-IID data distributions

Article Title: Explainable Federated Learning for Trustworthy Thoracic Disease Detection Under Non-IID Data Distributions

Article References: Arumugam, S., Sindhu, A., Saroja, M. N., Kannan, S., & Esakkiammal, A. (2026). Explainable Federated Learning for Trustworthy Thoracic Disease Detection Under Non-IID Data Distributions. Machine Learning, 115(10), Article 228. https://doi.org/10.1007/s10994-026-07174-z

Image Credits: AI Generated

DOI: 10.1007/s10994-026-07174-z

Keywords: federated learning, non-IID data, chest X-ray, explainable AI, SHAP, Grad-CAM, FedAvg, FedProx, EfficientNetB0, privacy-preserving machine learning, thoracic disease detection, NIH Chest X-ray 14