Privacy-Preserving Federated Learning with Secure Aggregation and Homomorphic Encryption for Brain Tumor Segmentation

Wireless sensor networks are now in widespread use to monitor regions, detect events and acquire information. To reduce the amount of sending data, an aggregation approach can be applied along the path from sensors to the sink. However, usually the carried information contains confidential data. Therefore, an end-to-end secure aggregation approach is required to ensure a healthy data reception. End-to-end encryption schemes that support operations over cypher-text have been proved important for private party sensor network implementations. Unfortunately, nowadays these methods are very complex and not suitable for sensor nodes having limited resources. In this paper, we propose a secure end-to-end encrypted-data aggregation scheme. It is based on elliptic curve cryptography that exploits a smaller key size. Additionally, it allows the use of higher number of operations on cypher-texts and prevents the distinction between two identical texts from their cryptograms. These properties permit to our approach to achieve higher security levels than existing cryptosystems in sensor networks. Our experiments show that our proposed secure aggregation method significantly reduces computation and communication overhead and can be practically implemented in on-the-shelf sensor platforms. By using homomorphic encryption on elliptic curves, we thus have realized an efficient and secure data aggregation in sensor networks.

ABSTRACTVehicular Ad Hoc Networks (VANETs) play a pivotal role in enabling intelligent transportation systems, yet their decentralized and dynamic nature exposes them to a wide range of cyber threats, including Sybil attacks, black hole attacks, replay, and message spoofing. To address these vulnerabilities, we propose HyDra‐VANET, a novel hybrid security framework that integrates deep learning, federated learning, and homomorphic encryption for robust and privacy‐preserving intrusion detection. At the vehicle level, a convolutional–recurrent neural network (CRNN) is employed to extract both spatial and temporal patterns from real‐time vehicular communication and telemetry data, ensuring accurate anomaly detection. Federated learning coordinates decentralized model training across vehicles, enabling collaborative intelligence while eliminating the need to share raw data. To further enhance privacy, a lightweight lattice‐based homomorphic encryption scheme allows encrypted inference and secure aggregation, preventing sensitive information leakage at intermediate nodes such as roadside units. Experimental evaluation using multiple datasets and adversarial scenarios demonstrates that HyDra‐VANET significantly outperforms baseline intrusion detection systems in detection accuracy, resilience to adversarial manipulation, scalability, and communication efficiency.

This article presents a comprehensive framework for implementing privacy-preserving cross-cloud monitoring using federated learning techniques. As organizations increasingly adopt multi-cloud strategies, maintaining unified observability without violating data sovereignty or regulatory requirements becomes challenging. The innovative system employs federated learning architecture to develop detection models across decentralized, encrypted transaction records, exchanging only model parameter updates between segregated cloud environments while preserving data locality and privacy. The architecture incorporates federated graph neural networks to discover hidden dependencies across cloud boundaries, secure aggregation through homomorphic encryption and secure multi-party computation, and differential privacy safeguards. Through case studies spanning defense, financial services, and healthcare sectors, Article demonstrates significant improvements in incident detection capability, reduction in false positives, and accelerated mean time to resolution while maintaining strict compliance with data protection regulations. The results establish federated learning as a viable solution for achieving cross-cloud observability without compromising sensitive operational data.

The digital transformation of healthcare in India has resulted in the rapid growth of Electronic Health Records (EHRs) across hospitals [1][3], paving the way for advanced data-driven clinical decision support systems [2][11]. However, centralized machine learning models require aggregating patient data into a single repository, which raises major concerns about privacy, data ownership, regulatory compliance, and cyber-security risks [4][9][13][14]. To address these challenges, this research proposes a federated learning-based privacy-preserving health data analysis framework for Indian hospitals [3][8]. In the proposed system, deep learning models are trained locally within each hospital without transferring raw patient data to external servers [1][12]. Instead, only encrypted model parameters or gradients are communicated to a central coordinator for secure aggregation [7][20], ensuring that sensitive medical data never leaves hospital premises [4][24]. The federated learning approach enables collaborative model development across multiple hospitals while maintaining data sovereignty, eliminating the risks of data leakage or unauthorized access [3][15]. The framework is enhanced with secure aggregation, differential privacy, and homomorphic encryption [6][10], making it resilient to inference attacks and modelreconstruction threats [16][24]. Experiments conducted over heterogeneous datasets representing different hospital types—government, private, and multi-specialty centers—demonstrate that the proposed federated model achieves predictive performance comparable to centralized learning while offering significantly higher privacy guarantees [5][12][21]. The system proves effective for key clinical tasks including disease prediction, patient risk stratification, and length-of-stay forecasting [15][17][25]. This research provides a scalable, ethically compliant, and regulatory-aligned paradigm for privacy-preserving AI adoption in Indian healthcare, supporting collaborative analytics without compromising patient confidentiality [14][23].

This systematic review examines the role of federated learning (FL) as a privacy-preserving paradigm for enterprise decision systems, synthesizing evidence from 187 peer-reviewed studies. Guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, the review integrates algorithmic, systems, security, sectoral, and governance perspectives to provide a comprehensive account of current knowledge. Findings highlight that foundational algorithms such as FedAvg, FedProx, and SCAFFOLD dominate the methodological landscape, with significant adaptations emerging to address non-IID and unbalanced datasets across distributed organizational silos. Privacy-preserving mechanisms—including differential privacy, secure aggregation, homomorphic encryption, and multiparty computation—were consistently applied as layered defenses, balancing mathematical guarantees with empirical resilience. The synthesis further revealed critical vulnerabilities to model poisoning, backdoor attacks, and gradient leakage, alongside defensive strategies such as robust aggregation, anomaly detection, and differential privacy clipping. Sector-specific implementations demonstrate FL’s practical utility in healthcare, finance, retail, logistics, telecommunications, and public services, where it enables collaborative modeling without violating data residency or confidentiality requirements. Governance and ethical frameworks, particularly GDPR, CCPA, and the NIST Privacy Framework, were found to shape deployment practices, while documentation artifacts such as datasheets, model cards, and privacy budget ledgers ensure accountability and transparency. Comparative surveys position FL as an integrative socio-technical architecture that unites distributed optimization, privacy engineering, adversarial robustness, and AI governance into a coherent enterprise-ready model. The review concludes that federated learning provides enterprises with a scalable, secure, and ethically aligned approach to leveraging distributed data while preserving trust and compliance.

In light of the emergence of privacy breaches in federated learning, secure aggregation protocols, which mainly adopt either homomorphic encryption or threshold secret sharing techniques, have been extensively developed to preserve the privacy of each client's local gradient. Nevertheless, many existing schemes suffer from either poor capability of privacy protection or expensive computational and communication overheads. Accordingly, in this paper, we propose an efficient and multi-private key secure aggregation scheme for federated learning. Specifically, we skillfully design a multi-private key secure aggregation algorithm that achieves homomorphic addition operation, with two important benefits: 1) both the server and each client can freely select public and private keys without introducing a trusted third party, and 2) the plaintext space is relatively large, making it more suitable for deep models. Besides, for dealing with the high dimensional deep model parameter, we introduce a super-increasing sequence to compress multi-dimensional data into one dimension, which greatly reduces encryption and decryption times as well as communication for ciphertext transmission. Detailed security analyses show that our proposed scheme can achieve semantic security of both individual local gradients and the aggregated result while achieving optimal robustness in tolerating client collusion. Extensive simulations demonstrate that the accuracy of our scheme is almost the same as the non-private approach, while the efficiency of our scheme is much better than the state-of-the-art baselines. More importantly, the efficiency advantages of our scheme will become increasingly prominent as the number of model parameters increases.

Federated Learning (FL) has transformed AI applications in healthcare by enabling collaborative model training across multiple institutions while preserving patient data privacy. Despite its advantages, FL remains susceptible to security vulnerabilities, including model inversion attacks, adversarial data poisoning, and communication inefficiencies, necessitating enhanced privacy-preserving mechanisms. In response, this study introduces Privacy-Preserving Federated Learning (PPFL), an advanced FL framework integrating Secure Multi-Party Computation (SMPC), Differential Privacy (DP), and Homomorphic Encryption (HE) to ensure data confidentiality while maintaining computational efficiency. I rigorously evaluate PPFL using Federated Averaging (FedAvg), Secure Aggregation (SecAgg), and Differentially Private Stochastic Gradient Descent (DP-SGD) across real-world healthcare datasets. The results indicate that this approach achieves up to an 85% reduction in model inversion attack success rates, enhances privacy efficiency by 30%, and maintains accuracy retention between 95.2% and 98.3%, significantly improving security without compromising model performance. Furthermore, comparative visual analyses illustrate trade-offs between privacy and accuracy, scalability trends, and computational overhead. This study also explores scalability challenges, computational trade-offs, and real-world deployment considerations in multi-institutional hospital networks, paving the way for secure, scalable, and privacy-preserving AI adoption in healthcare environments.

A Validation Framework for Brain Tumor Segmentation

Segmentation and classification of brain tumor are time-consuming and challenging chore in clinical image processing. Magnetic Resonance Imaging (MRI) offers more information related to human soft tissues that assists in diagnosing brain tumor. Precise segmentation of the MRI images is vital to diagnose brain tumor by means of computer-aided medical tools. Afterwards suitable segmentation of MRI brain tumor images, tumor classification is performed that is a hard chore owing to complications. Therefore, Gannet Aquila Optimization Algorithm_deep maxout network (GAOA_DMN) and GAOA_K-Net+speech enhancement generative adversarial network (GAOA_K-Net+Segan) is presented for classification and segmentation of brain tumor utilizing MRI images. Here, pre-processing phase performs noise removal from input image utilizing the Laplacian filter and also the region of interest (ROI) extraction is also carried out. Then, segmentation of brain tumor is conducted by K-Net+Segan, which is combined by Motyka similarity. However, K-Net+Segan for segmentation is trained by GAOA that is an amalgamation of Gannet Optimization Algorithm (GOA) and Aquila Optimizer (AO). From segmented image, features are extracted for performing classification phase. At last, brain tumor classification is conducted by DMN, which is tuned by GAOA and thus, output is obtained. Furthermore, GAOA_K-Net+Segan obtained better outcomes in terms of segmentation accuracy whereas devised GAOA_DMN achieved maximum accuracy, true negative rate (TNR) and true positive rate (TPR) of 92.7%, 94.5% and 91.5%.

A Review on Convolutional Neural Networks for Brain Tumor Segmentation: Methods, Datasets, Libraries, and Future Directions

In this work, we propose a fully automatic brain tumor and edema segmentation technique in brain magnetic resonance (MR) images. Different brain tissues are characterized using the novel texture features such as piece-wise triangular prism surface area (PTPSA), multi-fractional Brownian motion (mBm) and Gabor-like textons, along with regular intensity and intensity difference features. Classical Random Forest (RF) classifier is used to formulate the segmentation task as classification of these features in multi-modal MRIs. The segmentation performance is compared with other state-of-art works using a publicly available dataset known as Brain Tumor Segmentation (BRATS) 2012 [1]. Quantitative evaluation is done using the online evaluation tool from Kitware/MIDAS website [2]. The results show that our segmentation performance is more consistent and, on the average, outperforms other state-of-the art works in both training and challenge cases in the BRATS competition.

Federated Learning (FL) is a transformative paradigm that enables collaborative machine learning across multiple decentralized data sources without requiring direct data sharing. This approach has emerged as a critical solution for privacy-preserving machine learning, particularly in sensitive domains such as healthcare, finance, and IoT systems. FL frameworks employ advanced techniques such as secure aggregation, differential privacy, and homomorphic encryption to protect user data while facilitating model convergence. These frameworks address key challenges, including communication efficiency, data heterogeneity, and robustness against malicious actors. This paper reviews the state-of-the-art FL frameworks, exploring their architecture, privacy-preserving techniques, and application scenarios. Furthermore, it highlights recent advancements in secure federated learning protocols, multi-party computation strategies, and scalability improvements. The discussion extends to open challenges, such as ensuring fairness, optimizing resource consumption, and maintaining security guarantees in adversarial settings. By advancing the development of privacy-preserving FL frameworks, researchers and practitioners can unlock the potential of collaborative machine learning while upholding stringent data privacy standards.

The rapid expansion of the Internet of Things (IoT) ecosystem has transformed industries but also exposed significant cybersecurity vulnerabilities. Traditional centralized methods for securing IoT networks struggle to balance privacy preservation with real-time threat detection. This study presents a Federated Learning-Driven Cybersecurity Framework designed for IoT environments, enabling decentralized data processing through local model training on edge devices to ensure data privacy. Secure aggregation using homomorphic encryption supports collaborative learning without exposing sensitive information. The framework employs GRU-based recurrent neural networks (RNNs) for anomaly detection, optimized for resource-constrained IoT networks. Experimental results demonstrate over 98% accuracy in detecting threats such as distributed denial-of-service (DDoS) attacks, with a 20% reduction in energy consumption and a 30% reduction in communication overhead, showcasing the framework’s efficiency over traditional centralized approaches. This work addresses critical gaps in IoT cybersecurity by integrating federated learning with advanced threat detection techniques. It offers a scalable, privacy-preserving solution for diverse IoT applications, with future directions including blockchain integration for model aggregation traceability and quantum-resistant cryptography to enhance security.

Brain tumors are one of the most common neurological complications that affect millions each year. Current studies have focused only on either the segmentation of brain tumors or healthy anatomical regions. Unfortunately, there is a lack of research on whole brain segmentation of patients with existing tumors. The segmentation of these regions is cruical for understanding patient pathologies. Due to the distortions in the brain caused by tumors, whole-brain segmentation models trained on healthy brains will not perform accurately on diseased ones. To address this issue, this paper proposes a novel unified 3D nnU-Net model for the segmentation of both brain tumor and whole-brain structures for the first time. Training a model on datasets that include both healthy and cancerous brain regions could enhance the generalizability of machine learning (ML) models for performing both whole-brain segmentation and tumor segmentation tasks. Currently, there are no datasets that are publicly available for this purpose. This paper proposes to address this challenge by augmenting data from OASIS-1 and BraTS2023 to train the baseline model. The model successfully achieves a Dice Similarity Coefficient (DSC) of 0.814 for whole brain structures and 0.874 for tumor segmentation, demonstrating its effectiveness across diverse regions.

In brain tumor diagnosis, clinicians integrate their medical knowledge and brain magnetic resonance imaging (MRI) scans to obtain the nature and pathological characteristics of brain tumors and to decide on treatment options. However, manually detecting and segmenting brain tumors in today's brain MRI, where a large number of MRI scans taken for each patient, is tedious and subjected to inter and intra observer detection and segmentation variability. As result a number of methods have been proposed in recent years to fill this gap, but still there is no commonly accepted automated technique by clinicians to be used in clinical floor due to accuracy and robustness issues. In our approach, an automatic brain tumor detection and segmentation framework that consists of techniques from skull stripping to detection and segmentation of brain tumors is proposed with fuzzy Hopfield neural network as its final tumor segmentation technique. Through preprocessing, image fusion and initial tumorous slice classification, the final hybrid intelligent fuzzy Hopfield neural network algorithm based tumor segmentation, and tumor region detection and extraction is achieved. The performance of the proposed framework is evaluated on various MR images including simulated and real, normal and tumorous. Quantitatively the method is validated against available ground truth using commonly used validation metrics. The final segmentation mean and standard deviation result in Jaccard similarity index, Dice similarity score, sensitivity and specificity are 0.8569+/-0.0896, 0.9186+/-0.0638, 0.9480+/-0.0402 and 0.9917+/-0.0387 respectively. Quantitative and qualitative segmentation result indicates the potential of the proposed framework.