Deponker Sarker Depto `

Hello! I am a Research Assistant at NSU Optics Lab of North South University, where I work on computer vision and machine learning with a focus on medical imaging.

As a Reseach Assistant I work under supervision of Dr. M. Sohel Rahman and Dr. Mahdy Rahman Chowdhury


And, I love cats. They warm my heart and make me genuinely smile all the time. 🐈 💖



Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn

profile photo

Research

I'm interested in computer vision and machine learning with a focus on medical imaging (Microscopy, WSI, Histopathology, Radiology, Ultrasounds etc). In particular, my research interest lies in the following areas.

  • The application of Deep learning and Machine leraning algorithms for medical diagonosis and prognosis.
  • Usage of deep learning to design human level CAD (Computer-Aided-Diagonostic) systems.
  • Exploring the effectiveness and inefficiences of Deep learnng and Machine learning algorithms & techniques.
  • Pushing the SOTA (state-of-the-art) of algorithmic design of very deep neural networks.

I am also actively involved in Quantum Computing research projects in NSU Optics and Quantum Mechanics Lab.


News

  • 17, Dec, 2020: One Paper is Accepted at Tissue and Cell .
  • 27, Nov, 2022: One Paper is Accepted at Computers in Biology and Medicine .


       Publications
Automatic segmentation of blood cells from microscopic slides: A comparative analysis
Authors: Deponker Sarker Depto, Shazidur Rahman, Md. Mekayel Hosen, Mst Shapna Akter, Tamanna Rahman Reme, Aimon Rahman, Hasib Zunair, M. Sohel Rahman, M.R.C. Mahdy
Tissue and Cell, 2021
Paper / Dataset / Code /

With the advent of deep learing algorithms in medial domain, there is a need for quality and large datasets. In this work, we introduced the largest microscopic blood cell segmentation dataset and benchmark different state-of-the-art algorithms on it. Our findings and contributions are particularly helpful for researchers working in deep learning with applications in medial domain.

Quantifying Imbalanced Classification Methods for Leukemia Detection
Deponker Sarker Depto, Md. Mashfiq Rizvee, Aimon Rahman, Hasib Zunair, M. Sohel Rahman, M.R.C. Mahdy
Computers in Biology and Medicine, 2022
Paper / Code / Data

In a plethora of imbalanced classification techniques, to decide on a specific one for imbalance handeling is an ambiguous task. With our effort, in this paper, we have performed extensive analysis on State-Of-The-Art imbalanced calssification techniques. We have presented emperical evidance that loss-based tecniques perform better than input-based and GAN-based techniques in high imbalance scenario.


Teaching

  • Taught Quantum Computing coding (Qiskit) 19 senior year students in spring 2022 and 20 senior year students in summer 2022 respectively at North South University.
  • Covered topics are quantum gates, superposition, entanglement, heisenberg’s uncertainty principle, teleportation, super dense coding, quantum error correction along with building and debugging quantum circuits in simulator and running on real quantum devices on cloud.
  • Taught Python Programming and Deep Learning to University Students in Mahdy Research Academy.


Publication Recognition Certificates



Template credit goes to Jon Barron.
This document was last modified on: