University of the Punjab

Muhammad Shaban

Head of Frontier AI Models · Oncology R&D · AstraZeneca

AI research scientist specializing in computational pathology and foundation models for oncology applications, with a focus on biomarker discovery and clinical trial decision support. Experienced in developing and applying frontier AI systems for multimodal translational research and precision medicine.


Experience

Head of Frontier AI Models

Oncology R&D department in AstraZeneca
  • Driving Strategic Adoption of Frontier AI Models to Advance Oncology R&D.
October 2025 - Present

Machine Learning Scientist

AI for Pathology Image Analysis Lab, Department of Pathology, Mass General Brigham, Harvard Medical School
  • Worked on self-supervised representation learning from multiplex spatial proteomics images.
November 2023 - September 2025

Postdoctoral Research Fellow

AI for Pathology Image Analysis Lab, Department of Pathology, Mass General Brigham, Harvard Medical School
  • Worked on multiplex image analysis projects, including cell phenotyping, and protein marker imputation and digital biomarker identification for classical Hodgkin lymphoma progression.
  • Worked on multimodal deep learning for fusing medical images and genomics for understanding cancer metastases and cancers of unknown primary.
  • Collaborated with labmates on different projects, including context-aware cancer prognosis using graph convolutional networks.
February 2021 - October 2023

Research Assistant

Kindi Lab, Department of Computer Science, Qatar University
  • Participated in the CAMELYON16 consortium that demonstrated the first breakthrough using deep learning for computational pathology in lymph node metastasis classification, published in JAMA.
  • Developed computer vision approaches for person pose detection via deep learning, published in Information Fusion.
October 2015 - September 2016

Software Engineer

Techlogix Inc., Multinational IT Services, Consulting, and Business Solutions Company
  • Worked on Business Process Modeling (BPM) web application tools for Fortune 100 companies to streamline their business processes and enhance their productivity and efficiency manifold.
  • Led Business Intelligence (BI) projects using Oracle BI and reporting tools, which provided hands-on experience with large amounts of data and how to make it understandable, presentable, and usable for a novice user.
July 2011 - August 2013

Selected Publications

Preprints

  1. A Foundation Model for Spatial Proteomics. M Shaban, ... F Mahmood. arXiv preprint arXiv:2506.03373, 2025.


Journals

  1. A multimodal whole-slide foundation model for pathology. Tong Ding, Sophia J. Wagner, Andrew H. Song, ... Muhammad Shaban, ... Faisal Mahmood. Nature Medicine, 2025.

  2. MAPS: Pathologist-level cell type annotation from tissue images through machine learning. M Shaban, ... F Mahmood. Nature Communications, 2024.

  3. Towards a general-purpose foundation model for computational pathology. Richard J. Chen, Tong Ding, Ming Y. Lu, ... Muhammad Shaban, ... Faisal Mahmood. Nature Medicine, 2024.

  4. Demographic bias in misdiagnosis by computational pathology models. Anurag Vaidya, Richard J. Chen, Drew F. K. Williamson, ... Muhammad Shaban, ... Faisal Mahmood. Nature Medicine, 2024.

  5. Pan-cancer integrative histology-genomic analysis via multimodal deep learning. Richard J. Chen, Ming Y. Lu, Drew F. K. Williamson, ... Muhammad Shaban, ... Faisal Mahmood. Cancer Cell, 2022.

  6. Artificial intelligence for multimodal data integration in oncology. Jana Lipkova, Richard J. Chen, Bowen Chen, ... Muhammad Shaban, ... Faisal Mahmood. Cancer Cell, 2022.

  7. A digital score of tumour-associated stroma infiltrating lymphocytes predicts survival in head and neck squamous cell carcinoma. M Shaban, ... NM Rajpoot. The Journal of Pathology, 2021.

  8. Context-aware convolutional neural network for grading of colorectal cancer histology images. M Shaban, ... NM Rajpoot. IEEE Transactions on Medical Imaging (TMI), 2020.

  9. A Novel Digital Score for Abundance of Tumour Infiltrating Lymphocytes Predicts Disease Free Survival in Oral Squamous Cell Carcinoma. M Shaban, ... NM Rajpoot. Scientific Reports, 2019.

  10. An information fusion framework for person localization via body pose in spectator crowds. M Shaban, ... NM Rajpoot. Information Fusion, 2019.

  11. Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer. Babak Ehteshami Bejnordi, Mitko Veta, Paul Johannes van Diest, ... Muhammad Shaban, ... Rui Venâncio. JAMA, 2017.

Note: Check the Google Scholar profile for an updated and complete list of publications.


Education

University of Warwick, Coventry, UK

PhD in Computer Science, Department of Computer Science
  • Proposed novel method for context-aware deep learning to leverage domain specific information in gigapixel histopathology slides for colorectal cancer grading.
  • Developed two automated and objective methods for the quantification of histopathologic biomarkers in tumor microenvironment for better patient prognosis.
  • Led teams of four or more researchers, including pathologists, in three different projects and successfully concluded those projects with journal publications.
  • Actively participated in computational pathology challenges and presented work at several medical conferences.
  • Published more than 15 publications in a span of four years.
  • Worked under the supervision of Prof. Nasir Rajpoot in the Tissue Image Analytics (TIA) Center.
November 2016 - December 2020

Lahore University of Management Sciences, Lahore, PK

MS in Computer Science, School of Science and Engineering
August 2013 - June 2015

Punjab University, Lahore, PK

BS in Computer Science, College of Information Technology
October 2007 - July 2011

Certificates

Specializations

Certificate for Agentic AI and AI Agents for Leaders

Agentic AI and AI Agents for Leaders

View certificate

Certificate for Deep Learning Specialization

Deep Learning

View certificate


Courses

Certificate for Foundations of Project Management

Foundations of Project Management

View certificate

Certificate for Generative AI with Large Language Models

Generative AI with Large Language Models

View certificate


Skills

Programming Languages & Tools

  • Python
  • PyTorch
  • Scikit-learn
  • TensorFlow
  • MATLAB

Image Processing and Analysis

  • Management of Large-Scale Pathology Datasets
  • Image Augmentation and Preprocessing
  • Supervised and Unsupervised Learning
  • Self-supervised/Representation Learning
  • Vision Foundation Model Training
  • Model Optimization
  • Multimodal Learning
  • Model Interpretability Analysis
  • Result Analysis

Domain Knowledge

  • Pathology Fundamentals
  • Histopathology Image Analysis
  • Multiplex Spatial Proteomics Image Analysis

Research and Communication Skills

  • Research Problem Identification
  • Research Methodology
  • Publication and Presentation
  • Interdisciplinary Collaboration

Research Talks

  • Healthcare AI at NVIDIA - October 10th, 2024
  • Stanford MedAI Group Exchange Session - April 8th, 2024
  • TIA Seminar Series - University of Warwick - April 8th, 2024