Aditya Kumar

I am a PHD researcher with a strong focus on deep learning, computer vision, and medical image analysis. My research involves designing and implementing advanced hybrid and ensemble deep learning models aimed at improving the performance, robustness, and interpretability of automated image analysis systems. I actively explore the integration of modern architectures such as EfficientNet, ResNet, Vision Transformers (ViT), and U-Net-based frameworks with optimization algorithms and hybrid loss functions to enhance model efficiency and generalization.

My academic work is grounded in methodological innovation, emphasizing cross-dataset evaluation, explainability, and real-world deployment readiness. I am particularly interested in improving segmentation and classification tasks through encoder-decoder architectures, attention mechanisms, and post-processing techniques like Conditional Random Fields (CRF). My goal is to contribute to the development of intelligent, scalable, and interpretable AI solutions that meet high research and publication standards.

I am also actively involved in writing and contributing to high-impact SCI journal articles and conference papers, collaborating with interdisciplinary teams, and staying engaged with the latest trends in deep learning and image processing research.