Machine Learning, Deep Learning, Computer Vision, Natural Language Processing
- Master of Science, Electrical and Computer Engineering, University of Florida (2021-2024)
- Bachelor of Science, Electrical and Electronic Engineering Bangladesh University of Engineering and Technology (BUET) (2012-2017)
- Competition Expert in Kaggle (1 Silver Meal + 1 Bronze Medal)
- Discussion Expert in Kaggle
- Second Runner-up, Divisional Champion, 1st Runner-up, Bangladesh Math Olympiad 2006-08
- Machine Learning Engineer, AI Samurai (2019-2020)
- Machine Learning Researcher, Semion Limited (2017-2019)
-
EVHA: Explainable Vision System for Hardware Testing and Assurance - An Overview" MD Mahfuz Al Hasan, Tahsin Mostafiz, Thomas An Le, Jake Julia, Nidish Vashistha, Shayan Taheri, and Dr. Navid Asadizanjani
-
Pathology Extraction from Chest X-Ray Radiological Reports: A Performance Comparison" Tahsin Mostafiz, Dr. Khalid Ashraf
-
Retinal Blood Vessel Segmentation using Residual Block Incorporated U-Net Architecture and Fuzzy Inference System" Tahsin Mostafiz, Ismat Jarin, Dr. Shaikh A. Fattah and Dr. Celia Shahnaz; IEEE WIECON-ECE 2018.
-
Photoplay: An Android Application to Stimulate Children’s Cognitive Development" Avijit Mitra, Tahsin Mostafiz, Raihan Ur Rashid; Humanitarian Technology Conference (R10-HTC), 2017 IEEE Region 10, Dhaka.
-
Algorithm
-
Head and Neck Cancer staging via lymph node segmentation using Convolutional Neural Network (CNN) and volumetric CT images.
-
Computed Tomography (CT) image reconstruction for Printed Circuit Boards (PCB) with semi-supervised denoising and artifact suppression techniques.
-
Counterfeit IC detection from SEM images of IC backsides using image processing and CNN.
-
Co-development of a semi-supervised CNN model for abnormality detection in dairy product images.
-
Backend deep learning algorithm development of , a web application for the detection and localization of Intracranial Hemorrhage from brain CT images.
-
Sentiment level analysis using transformer based language models for product review task.
-
Identification of Risk Factors for Heart Disease from i2b2 dataset Using a Bidirectional LSTM network with 50 Dimensional Glove Word Embedding.
-
-
Application
-
Co-development of a flutter based android app for electric pole detection from images captured using car dashboard cameras.
-
Co-development of SemRad, an inference tool and a class activation mapping (CAM) Tool Using ResNet101 for Detection and Localization of Abnormalities in Chest X-ray Images for this software.
-
semDDX, an Android app was designed to help the users navigate the vast landscape of differential diagnoses (DD) and help medical students to learn DD easily.
-
Differential Diagnoses, an Amazon Alexa skill was designed to help the users find all differential diagnoses for a symptom.
-
Symptom Checker an Amazon Alexa skill was designed to help the users detect disease from symptoms.
-
-
Deep Learning Competitions
-
Student Supervising
-
Mentored an undergraduate student in his thesis work titled "COVID Infection Analysis via Lung Lobe Segmentation using Deep Learning".
-
Supervised two high-school students to get them familiar with research work in hardware security and machine learning under the Student Science Training Program (SSTP).
-