Browsing by Author "Alam, Jawad Bin"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    An efficient deep learning framework for diabetic retinopathy classification using generative data augmentation and knowledge distillation
    (BRAC University, 2025-10) Fahim, Tasnimul Mohammad; Ayon, Tanzim Hossain; Datta, Shuvo; Alam, Jawad Bin; Sabur, Sarika Bintey; Chakrabarty, Amitabha
    Diabetic retinopathy (DR) affects 93 million people globally and is a leading cause of preventable blindness. Automated screening systems face two critical barriers: severe class imbalance in medical datasets where healthy cases vastly outnumber vision-threatening stages, and high computational demands that limit deployment in resource-constrained settings. This research presents a three-stage framework integrating Progressive Growing GANs for synthetic fundus image generation, a dualbranch ensemble architecture combining EfficientNet-V2-S with Vision Transformer- B/16, and knowledge distillation to compress the model into MobileNetV2. The generative approach created a balanced training corpus of 10,000 images across five severity classes, addressing the fundamental imbalance problem while maintaining pathological authenticity validated through perceptual metrics. On APTOS 2019, the ensemble teacher achieves 95.8% accuracy with 0.975 quadratic weighted kappa. Cross-dataset validation on DDR and Messidor-2 confirms robust generalization across diverse clinical populations and imaging protocols. The distilled MobileNetV2 student retains 92.4% accuracy and 0.938 kappa while reducing model parameters by over ninety-six percent and computational operations by over ninetyseven percent. Inference latency decreases by over eighty percent on CPU, with model size compressed to enable mobile deployment. Grad-CAM visualizations confirm clinically relevant attention to microaneurysms, hemorrhages, and neovascularization. This framework demonstrates that generative augmentation combined with heterogeneous ensemble learning and knowledge distillation overcomes the accuracydeployability trade-off, enabling accessible DR screening in resource-limited settings.
  • Thumbnail Image
    Item
    IoT based intelligent wireless substation monitoring and controlling system
    (Brac University, 2022-11) Khan, Serajul Islam; Abdullah, S.M.Sifat; Alam, Jawad Bin; Das, Pramit; Azad, AKM Abdul Malek; Mohsin, Abu S.M.; Rahman, Md. Mosaddequr
    As the advancement of technologies, the automation of substations has become a key concept in the current world. In order to improve the power quality it is necessary to have an idea about the constraints that occurred. As a result, this project proposes an IoT based system for the monitoring, controlling and protection of the system and analysis of data in a simpler cost effective manner. The sensor data is conveniently accessed by any device wirelessly through IoT. It incorporates real time error free data of substation. The remote electrical parameters we focused on this project are voltage, current, temperature and frequency. In this prototype, we have a set of predefined values for the parameters. A relay ensures the protective part and sends alerts to the users during any faults and once the problem is resolved the system works again.