AJAyesha Jabbar

Ayesha Jabbar

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ayesha Jabbar has a total of 13 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Automated Denoising of Mammographic Microcalcifications Using Wavelet-Based Attention Deep Network for the Medical Internet of Things

    by Khalil Ur Rehman, Anaa Yasin, Li Jianqiang, Ayesha Jabbar, Yibin Tian

    Breast microcalcifications are key radiological indicators of malignancy, and their assessment in denser breast tissues is challenging. We propose a Wavelet-based Attention Deep Network (WADN) that isolates high-frequency components linked to microcalcification clusters using denoising wavelet transforms and integrates them into a deep attention architecture for interpretable, localization-assisted classification. The system is deployed in an edge-computing framework enabled by Jetson Orin Nano to support real-time analysis and remote radiologist access for the Medical Internet of Things (MIoT). For the private PINUM and public DDSM datasets, WADN outperforms standard convolutional neural networks (CNNs) and classical machine learning baselines, achieving accuracies of 0.95 and 0.93, respectively, with consistent gains in sensitivity, specificity, precision, F1-score, and area under the curve (AUC). On the IoT edge device, the method achieves accuracies of 0.91 for PINUM and 0.93 for DDSM, demonstrating the feasibility of ondevice screening. These results indicate that combining wavelet transforms with attention mechanisms and IoT edge deployment improves diagnostic accuracy while enabling practical real-time breast-cancer screening workflows.

  • Radiomic V-Net Deep Learning Architecture for Breast Cancer Prediction Deployed on Medical Consumer Electronics

    by Khalil Ur Rehman, Yibin Tian, Jianqiang Li, Anaa Yasin, Ayesha Jabbar, Jungpil Shin, Mohammad Yahya H. Al-Shamri, Inam Ullah

    This paper introduces a radiomic-integrated 2D V-Net architecture designed for deployment on medical consumer electronics (MCE) platforms. The system targets real-time detection of architectural distortion in mammograms, a critical indicator for early breast cancer screening. Unlike conventional clinical models, our approach is optimized for consumer-grade edge devices, including Jetson Nano, Raspberry Pi 4, and Snapdragon 865, enabling diagnostic capability in home, rural, and mobile health settings. To achieve this, we employ lightweight design strategies such as top-hat morphological preprocessing, radiomic feature integration, and skip-connected V-Net layers. Deployment efficiency is further enhanced with INT8 quantization, pruning, and operator fusion, reducing model size and latency while preserving accuracy. Experimental evaluation on the CBIS-DDSM and PINUM datasets demonstrates robust performance, with accuracies of 0.98 and 0.97 and specificities of 0.97 and 0.87, respectively. A dedicated visualization interface for smartphones and tablets provides interactive feedback and encrypted report sharing, bridging patients and healthcare providers. By combining radiomic insights, edge AI deployment, and consumer-grade accessibility, this work contributes a scalable, privacy-aware solution for smart healthcare ecosystems and next-generation MCE.


  • Open Access

    Multi-Sensor Wearable-Based Sleep Stage Classification Using Federated Learning for Enhanced Privacy

    by Asad Ali, Huang Jianjun, Ayesha Jabbar, Muhammad Kashif Jabbar

    Sleep is an essential aspect of human health, and Obstructive Sleep Apnea (OSA) is a serious condition that is frequently unidentified because traditional Polysomnography (PSG) is costly and available only to a small number of patients. To address these limitations, this study develops and implements a Federated learning ensemble technique for privacy-preserving, sleep stage analysis with wearable devices. The framework adapts Random Forest classifiers for local client-side training and Logistic Regression for global aggregation of models, with a high level of classification accuracy and user data privacy. A principal approach in this methodology is the fusion of Blood Volume Pulse (BVP), Electrodermal Activity (EDA), heart rate, and triaxial accelerometry data for the assessment of sleep dynamics. The proposed approach highlights the possibility of developing low-cost, privacy-preserving health monitoring systems for multiple applications that often face obstacles such as data heterogeneity and resource limitations. This work provides support in progressing wearable sleep technologies and helps in the creation of effective, reliable applications in the healthcare industry.


  • Open Access

    Federated learning for connected autonomous vehicles: enhancing mobility, safety, and sustainability

    by Asad Ali, Jianjun Huang, Ayesha Jabbar, Mubashir Ali, Sijjad Ali

    Connected Autonomous Vehicles (CAVs) as vehicles of the future hold the social value as elements of an intelligent transport system that can contribute new advancements in mobility, safety and sustainability. However, there are issues like communication overhead, data heterogeneity and model convergence which limit the optimization of these techniques. Such a model training can be effectively done with the help of Federated Learning (FL), which helps training the same while restricting data transfer. This paper reports a systematic review of the recent literature on FL in CAVS, highlighting the state-of-the-art progress and challenges. It also discusses how FL may be applied to CAVs to enhance safety and traffic flow while achieving positive environmental impacts, addressing data privacy, distributed learning, and the ability to make real-time decisions. It also describes directions for future work to mitigate these issues, such as communication overhead, data heterogeneity, and model convergence in FL for CAVs. This paper discusses how FL enhances safety, productivity, and environmentally-friendly development in CAVs’ networks and the future research directions and issues regarding intelligent, privacy-preservation transportation systems.


  • Open Access

    Mamba-Based VoxelMorph Framework for Cardiovascular Disease Imaging and Risk Assessment

    by Muhammad Kashif Jabbar, Huang Jianjun, Ayesha Jabbar, Zaka Ur Rehman

    Cardiovascular diseases (CVDs), particularly coronary artery disease (CAD), remain the leading cause of global mortality, necessitating advanced diagnostic solutions. Accurate deformable image registration plays a crucial role in enhancing segmentation precision and classification performance in cardiovascular imaging. However, existing registration methods, including VoxelMorph, face limitations in computational efficiency and memory usage, restricting their real-time applicability for high-resolution cardiac imaging. This study proposes the Mamba-Optimized VoxelMorph framework, which leverages GPU-based parallelization and memory optimization to address these challenges. The framework achieves superior registration accuracy, yielding a Dice Similarity Coefficient (DSC) of 0.95 and Normalized Cross-Correlation (NCC) of 0.90, while reducing computational time by 40% and memory usage to 800 MB. These advancements ensure efficient alignment of complex cardiac structures, thereby improving segmentation accuracy and classification reliability. By addressing these critical limitations, the Mamba-Optimized VoxelMorph framework significantly enhances cardiovascular imaging, enabling precise, scalable, and real-time deformable image registration for improved CAD diagnosis and treatment planning.


  • Open Access

    Recent Advances in Federated Learning for Connected Autonomous Vehicles: Addressing Privacy, Performance, and Scalability Challenges

    by Asad Ali, Huang Jianjun, Ayesha Jabbar

    This review paper systematically explores the use of Federated Learning (FL) in Connected Autonomous Vehicles (CAVs) to improve vehicle performance, safety and user experience. FL presents a decentralized infrastructure that allows collaborative learning, while also ensuring data privacy, as CAVs increasingly rely on machine learning to process large amounts of sensor data. FL has been demonstrated to contribute to cooperative perception, vehicle trajectory prediction, traffic flow optimization, predictive maintenance, personalized in vehicle experiences, and advanced driver assistance systems (ADAS) through recent advances. Secondly, FL speeds up anomaly and intrusion detection, critical safety and security matters in CAV ecosystems. However, despite all these advancements, there are still challenges of data heterogeneity, communication overhead, security vulnerabilities, model convergence as well as regulatory compliance that constitute major barriers to FL deployment. This paper synthesizes recent contributions on this topic, identifies some gaps in the current research, and proposes future directions to fill those gaps. Federated Learning is a promising technology path towards driving innovations and improving the power and effectiveness of Autonomous Transportation Systems.