Soujanya Hazra

Electrical Engineering, IIT Kharagpur, India.

prof_pic.jpeg

Kharagpur, WB, India.

Hello! This is Soujanya. I’m currently a second-year master’s student at Indian Institute of Technology, Kharagpur, enrolled in Signal Processing and Machine Learning, where I am advised by Dr. Sanjay Ghosh. Before joining IIT-KGP, I obtained my bachelor’s degree at Kalyani Government Engineering College, working with Dr. Pritam Kumar Gayen.

My research interest lies in:

  • Machine learning and explainability
  • EEG-based study of neurological disorders
  • Brain functional connectivity modelling
  • Medical image analysis for oral cancer

I’m open to research collaborations! If you’re interested in working together, feel free to reach out.

news

Feb 13, 2026 Our abstract, “Wavelet Coherence Connectivity for EEG-based Classification of Major Depressive Disorder”, has been accepted for a poster presentation at the OHBM 2026.
Feb 05, 2026 Got the MICCAI Society Membership Grant!
Jan 19, 2026 Presented my poster on “From EEG Signals to Explainable Brain Graphs for Depression Analysis” at Workshop on AI for Science and Technology.
Jan 12, 2026 Our paper “RAA-MIL: A Novel Framework for Classification of Oral Cytology” got accepted at ISBI 2026 :sparkles:.
Jul 07, 2025 Attended the Joint Telematics Group/IEEE Information Theory Society Summer School on Signal Processing, Communications and Networks

community involvement

Reviewer: ISBI, ICASSP, IEEE TAFFC

selected publications

  1. ISBI 2026
    RAA-MIL.png
    RAA-MIL: A Novel Framework for Classification of Oral Cytology
    Rupam Mukherjee, Rajkumar Daniel, Soujanya Hazra, and 2 more authors
    International Symposium on Biomedical Imaging (ISBI 2026) Proceedings, 2026
  2. arXiv
    expanet.png
    Bridging Accuracy and Explainability in EEG-based Graph Attention Network for Depression Detection
    Soujanya Hazra and Sanjay Ghosh
    arXiv preprint, 2025
  3. arXiv
    graphunwrapnet.png
    Graph Guided Modulo Recovery of EEG Signals
    Soujanya Hazra and Sanjay Ghosh
    arXiv preprint, 2025