Extremal Contours: Gradient-driven contours for compact visual attribution
Training-free method that uses gradient-driven smooth contours for compact visual attribution.
I am a Ph.D. candidate in Computer Science at the University of Copenhagen with hands-on experience in computer vision and deep learning across both academic and applied settings. My work focuses on building reliable, end-to-end machine learning systems, with expertise in 3D medical image analysis, object detection, self-supervised learning, and explainable AI.
I have designed and implemented models ranging from YOLO-based detection pipelines to large-scale 3D segmentation and contrastive learning frameworks, taking projects from data preparation and preprocessing through model development, evaluation, and reproducible implementation. In parallel, I have industry experience developing production-oriented vision systems and leading small technical teams in interdisciplinary and international environments, and I am seeking a role where strong ML engineering and problem-solving skills can be applied to real-world products with an emphasis on robustness, interpretability, and impact.
Outside of research, I enjoy solving mechanical puzzles, exploring new technologies, and studying psychology and human behavior.
PhD Computer Science
Copenhagen University
MS in Biomedical Engineering
Sharif University of Technology
BS in Biomedical Engineering
Amirkabir University of Technology
I am a Ph.D. candidate in Computer Science at the University of Copenhagen, working on computer vision and deep learning with a focus on medical imaging/data, self-supervised learning, and explainable AI. My work emphasizes building robust, reproducible machine learning systems for real-world healthcare applications.
Please reach out to collaborate 😃
Training-free method that uses gradient-driven smooth contours for compact visual attribution.
Contrastive learning framework with overlap-aware modules for improved 3D medical image segmentation.
Extremal Contours; Gradient-driven contours for compact visual attribution
PiMPiC; An Overlap-Aware Contrastive Learning Framework for 3D Patch-Based Medical Image Segmentation