Israt Jahan Khan

M.Sc. student at BRAC University. Researching machine learning, computer vision, and interdisciplinary AI. Published in IEEE Access and Elsevier.

Dhaka, Bangladesh · Open to PhD opportunities worldwide

4

Publications

30+

Citations

13+

Awards

2

Conferences

Israt Jahan Khan - Prospective PhD Candidate in Computer Science

Selected Publications

Peer-reviewed work in machine learning, computer vision, and applied AI. Published in IEEE and Elsevier venues.

View all publications
[01]
2025Journal12 cited

Enhanced ROI guided deep learning model for Alzheimer's detection using 3D MRI images

Israt Jahan Khan, Md Fahim Bin Amin, Md Delwar Shahadat Deepu, Hazera Khatun Hira, Asif Mahmud, Anas Mashad Chowdhury, Salekul Islam, Md Saddam Hossain Mukta, Swakkhar Shatabda, Alzheimer's Disease Neuroimaging Initiative

IMU · Elsevier

[02]
2024Journal6 cited

SafeTrax: Smart Collision Prediction and Alert System Using IoT for Sustainable Traffic Safety

Md Fahim Bin Amin, Israt Jahan Khan, Faria Akter, Adib Ahmed, Md Motaharul Islam

IEEE Access

What I Study

Explore research

Computer Vision

Vision-based systems for image understanding, object detection, and visual reasoning.

Machine Learning

Supervised, unsupervised, and ensemble approaches for classification and regression.

Deep Learning

Neural architectures (CNNs, attention mechanisms, 3D ResNets) for complex AI tasks.

Interdisciplinary AI

AI applied to healthcare, traffic safety, data security, and other real-world domains.

Selected Work

View all projects
Upgraded-SoupX
ResearchBioinformatics

Upgraded-SoupX

Full Python port and extension of the original SoupX R package (Young & Behjati, 2020). Drops the R dependency entirely, adds probabilistic per-cell decontamination (DecontX), doublet-aware estimation, gene-heterogeneity correction, a complete downstream analysis pipeline, and eight quantitative benchmark metrics, all on top of the same core SoupChannel abstraction.

PythonscRNA-seqDecontXNumPy
VAE for Audio Representation & Clustering
ResearchDeep Learning

VAE for Audio Representation & Clustering

Comprehensive study on Variational Autoencoders (VAEs) for unsupervised representation learning and clustering of audio data. Investigates how different VAE architectures and latent space enhancement techniques affect clustering performance across datasets of varying scale and complexity.

PythonPyTorchVAELatent Space
Open to Opportunities

Seeking PhD Positions

Actively pursuing PhD opportunities in Computer Science, Machine Learning, and Computer Vision at research-intensive universities worldwide.