KADER MOHIDEEN NAFEEZ AHAMED SHAFI

Adaptive Myoelectric Hand Control Using Reinforcement Learning (01/25 to 05/25)

• Developed a reinforcement learning-based adaptive control system for a myoelectric prosthetic arm integrating EMG signals.
• Designed and implemented the RL model training pipeline and performed algorithm tuning for optimized control.
• Conducted analysis of results to evaluate system adaptability and performance in dynamic conditions.
• Co-authored a research paper based on the project, highlighting its potential impact in biomedical technology.

Acute Lymphoblastic Leukemia Severity prediction using CNN (07/24 to 12/24)

• Engineered the core deep learning model (CNN) for ALL severity classification, achieving a validated 98.6% classification accuracy in differentiating four malignant/benign subtypes.
• Developed a robust image preprocessing pipeline using OpenCV (Gaussian filtering, segmentation) to consistently isolate and standardize lymphoblast cells for analysis.
• Optimized the CNN architecture via transfer learning and rigorous hyperparameter tuning (Adam, batch normalization) to ensure strong generalization and minimize overfitting.