Know
About me

Experienced AI Researcher, interested in designing high-performance deep learning models that are understandable to humans. I believe the language of explanations should include higher-level, human-friendly concepts. My current focus is on medical imaging and conceptually controllable\interpretable methods. Currently working as a Graduate Researcher at QUANTIMB Lab. Before that, I worked as a Research Assistant at National Center of Artificial Intelligence Pakistan.

02

My
Experience

Nov 2022 - Present

Graduate Researcher

QUANTIMB Lab (Quantitaive Imaging Biomarker)

> Working on Brain Tumor/Metastasis
Jan 2022 - Sept 2022

Research Assistant

Medical Imaging and Diagnostics Lab, National Center of Artificial Intelligence

> Developed an AI-based DICOM viewer for the diagnosis of Tuberculosis, Breast Cancer and Brain tumour.
> Worked on fully automated AI-Based PACS Picture Archiving and Communication System).
>Presented demo at Brain Oncology Symposium organized by Aga Khan Medical University.
> Designed a Logo for MID Lab-NCAI.
> Presented AI models and DICOM Viewer to hospitals and Labs.
> Collected and preprocessed the data from hospitals.
> As a member of the organizing team, organized AI industrial events and training sessions.
> Assisted MS students with their thesis, project and publications.
Dec 2020 - Nov 2021

Research Scholar

Medical Imaging and Diagnostics Lab, National Center of Artificial Intelligence

> Developed models for visual attribution of medical images, particularly for domains where pixel-level labels are difficult to attain (such as Alzheimer). Evaluated and performed experiments on three datasets including synthetic, Alzheimer;s disease Neuro imaging Initiative(ADNI) and, BraTS dataset.
> Got hands-on experience in Generative Adversarial Networks, Zero-shot learning, EXplainable and interpretable deep learning methods, and Visual Feature Attribution techniques. Worked on several deep learning projects, including skin lesion detection, generated unseen objects via zero-shot learning, receptive field view, diabetic retinopathy, Mask R-CNN, YOLO, TbX11k, concept extraction and activation vectors , Alzheimer’s disease and Brain tumor abnormal to normal image conversion and vice versa.
Mar 2019 - Oct 2020

IT-Consultant

Buildingz, USA, Miami, FL

Designed and Deployed Real Estate Application focusing IoT based Solutions

03

My
Education

November 2022

PhD Computer Science (Artificial Intelligence)

York University, Toronto

Thesis Title
In Progress
September 2021

Master of Science in Computer Science (MSCS)

COMSATS University Islamabad, Islamabad

Thesis Title
CNET: A Concept‑Controlled Deep Learning Architecture for Interpretable Image Classification
September 2018

Bachelor of Science in Software Engineering (BSSE), Silver Medal

University of Sargodha, Sargodha, Pakistan

Thesis Title
READ: Requirements Engineering Analysis and Design

04

Selected Publications

SoFTNet: A concept-controlled deep learning architecture for interpretable image classification (IF 8.139)

View

Zia, T., Bashir, N., Ullah, M. A., & Murtaza, S. (2022). SoFTNet: A concept-controlled deep learning architecture for interpretable image classification. Knowledge-Based Systems, 108066.
VANT-GAN: Adversarial Learning for Discrepancy-Based Visual Attribution in Medical Imaging (IF 4.757)

View

Tehseen Zia, Shakeeb Murtaza, Nauman Bashir Bhatti, David Windridge, Zeeshan Nisar, VANT-GAN: Adversarial Learning for Discrepancy-Based Visual Attribution in Medical Imaging, Pattern Recognition Letters, 2022, ISSN 0167-8655
Modeling Class Diagram using NLP in Object-Oriented Designing

View

Bashir, N., Bilal, M., Liaqat, M., Marjani, M., Malik, N., & Ali, M. (2021, March). Modeling Class Diagram using NLP in Object-Oriented Designing. In 2021 National Computing Colleges Conference (NCCC) (pp. 1-6). IEEE.