|
I am currently a MASc. student, working on computer vision and machine learning with applications in medical imaging advised by Prof. A. Ben Hamza at Concordia University. I also recieved the MITACS Accelerate Fellowship to work jointly with Concordia and Décathlon Canada on semi-supervised learning for visual recognition using large-scale sport image data. Previously, I was a deep learning software engineer at Think Bricks LLC where I worked on Smart Retina, a medical software to analyze fundus images. I also developed open-source robots at The Tech Academy! During my undergrad at North South University in Bangladesh, I was first exposed to research when I was advised by Dr. Nabeel Mohammed. I served as the founding chair of IEEE Robotics and Automation Society (RAS) Student Branch Chapter which is the first RAS student chapter in Bangladesh. I like to cook fancy meals (beauty is in the eye of the beholder ;)) and play competitive e-sports, Dota 2 and CSGO, during my free time. If you're into esports, add me on Steam! I also like funny dog videos! 🐕 Email / LinkedIn / GitHub / Google Scholar / Twitter / Medium |
|
|
My work involves building artificial intelligence (AI)-based computer vision systems from planning, development, to deployment. I have experience in coding up end-to-end deep learning workflows such as semi-supervised/ transfer/ adversarial learning for classification, detection, segmentation and generation of 2D images as well as 3D (videos, CT & MRI scans). Recently I have been working on developing data-efficient deep learning techniques which can reduce the negative effects of severe class imbalance, and also those which do not depend heavily on intensive manual labeling efforts. During my undergrad, I built 😬 a lot of other random stuff over time. |
|
We pretrain a model on videos for human activity recognition which leads to better representations for the downstream tuberculosis type classification task, especially for under-represented class samples. Our method achieved 2nd place in the ImageCLEF 2021 Tuberculosis Type Classification Challenge. |
|
Automating the tasks of detecting, segmenting, and classifying cell nuclei can free up the pathologists’ time for higher value tasks and reduce errors due to fatigue and subjectivity. This paper summarizes and publicly releases the challenge dataset, and compile key findings of the methods developed by various participants. |
|
Propose the first-of-its-kind open dataset of synthetic COVID-19 chest X-ray images using unsupervised domain adaptation by leveraging class conditioning and adversarial training. |
|
Transformed a high variation malaria localization dataset into a malaria classification dataset. Several classification algorithm benchmarks are provided for the task of classifying presence of malaria from microscopic images of isolated red blood cells. |
|
Showed that analyzing 3D medical images in a per slice basis is a sub-optimal approach, that can be improved by 3D context. Ranked 5-th in ImageCLEF 2019. |
|
Improved classification performance by synthesizing under-represented class samples from the over-represented ones. Synthetic samples are used as additional training data to reduce class imbalance. |
|
SincNet models based on joint angular margin loss not only consistently outperformed current prior models, but also generalizes well on unseen and diverse tasks such as Bengali speaker recognition. |
|
Benchmarked several classification algorithms for the task of detecting malaria from microscopic images of red blood cells. Transfer learning approach worked best in our study. |
|
A 3D CNN with a slice selection method employed in the task of chest CT image analysis for predicting tuberculosis (TB). Our method achieved 10-th place in the ImageCLEF 2019 Tuberculosis SVR - Severity scoring. |
|
An accuracy of 97.09% was achieved on the NumtaDB Bengali handwritten digit dataset, which was obtained by freezing intermediate layers. |
|
Implementation of an automated multifunctional attendance system which uses rfid, fingerprint, and real time facial recognition. |
|
A wireless sensor network with a real time web application for monitoring multiple ships to prevent catastrophic events due to overloading. |
|
A low-cost autonomous mobile security robot based on a multisensor system for the purpose of sending alarms remotely. |
|
|
|
This is a demonstration of a full stack deep learning project from training a model to deploying it, using a REST API endpoint as well a separate end-user prototype web application. The model is built for predicting the presence of melanoma from dermoscopic skin lesions using neural networks. |
|
Train a 3D convolutional neural network to predict presence of pneumonia. |
|
Trained a U-Net architecture with a pretrained ResNet backbone on knee MRIs at the slice level. The goal was to reconstruct high resolution images from the given undersampled image. |
|
Change monitor screen when your boss is coming towards you! |
|
A Python package built to send email notifications which can be integrated into any existing piece of python software. |
|
|
|
Teaching Assistant (Lab Demonstrator), COMP 478 Winter 2021 Teaching and assisting graduate students with implementation of image processing algorithms and course projects. |
|
Instructor, Image Classification with Python and Keras, 2019. Technical workshops where participants get to know about programming in Python, image processing fundamental using OpenCV and building image classification models using Keras. |