Usman Mohammad, AI Researcher
Open to collaborations

Usman Mohammad

AI Researcher

I develop AI methods at the intersection of Deep Learning, Computer Vision, and Computational Biology. My work spans single-cell analysis, medical imaging, and virtual cell modeling — building models that are not only powerful but also interpretable.

PhD Researcher at the Mathematics & Computer Sciences Department, University of Barcelona. Affiliated with Hospital Clínic Creatio and the Computational Biology group at BSC. Previously an ML Engineer in the medical domain. Honours Master's from COMSATS University Islamabad.

Research

Research Interests

Developing robust and interpretable AI for complex biological and medical data — from single-cell omics to virtual cell modeling.

Computer Vision

Object detection, segmentation, and scene understanding in medical and biological imagery.

Deep Learning

Novel architectures and training strategies for robust representation learning.

Medical Imaging

AI-driven analysis for diagnosis, treatment planning, and advanced therapies.

Single-Cell & Omics

AI methods for single-cell data analysis to understand disease at the cellular level.

Foundation Models

Large-scale pretraining and transfer learning for biological and medical applications.

Virtual Cell Modeling

Deep learning approaches to simulate and predict cellular behavior and disease mechanisms.

Explainable AI & Interpretability

Mechanistic interpretability and XAI methods to understand what deep models learn — essential for trustworthy medical AI.

Projects

Selected Projects

Applying AI to fundamental problems in neurodegeneration, alternative disease modeling, and computational biology.

Foundation Model
Active · Ongoing · BSC Computational Biology

Virtual Cell Modeling

Deep learning for virtual cell simulation and prediction. Building foundation models to simulate cellular behavior, predict disease mechanisms, and accelerate discovery in computational biology — working towards AI that can model the complexity of living cells.

Foundation Models Virtual Cell Computational Biology BSC ↗
Brain-on-Chip
Completed · 2022 – 2025 · Ministry of Science and Innovation

AlterNED

Alternative human in vitro methods to study neurodegenerative diseases. Developing brain-on-chip platforms as alternative models for neurodegeneration research, reducing reliance on animal models while enabling more precise study of disease mechanisms.

Neurodegeneration Brain-on-Chip In Vitro Models Creatio ↗
Single-Cell
Completed · 2021 – 2024 · "La Caixa" Foundation

HDREP

In vivo reprogramming to rescue alterations in Huntington's disease. Developed AI methods for single-cell data analysis to understand disease development and progression at the cellular level, enabling identification of key molecular signatures in Huntington's disease models.

Huntington's Disease Single-Cell Analysis In Vivo Reprogramming Creatio ↗
Vision Transformer
Completed · During Master's · COMSATS University Islamabad

Vision Transformers for Chest X-rays

Transfer learning with ViT architectures for chest X-ray analysis. Developed and adapted Vision Transformer models for medical imaging, exploring transfer learning strategies to bridge the domain gap between natural and medical images for pulmonary disease detection.

Vision Transformers Transfer Learning Chest X-ray Medical Imaging

Mentors

Mentors & Collaborators

Guided by exceptional researchers across multiple institutions whose expertise and mentorship shape my work.

JCC

Dr. Jose Carbonell Caballero

Researcher, BSC Computational Biology

Barcelona Supercomputing Center · Direct collaborator on virtual cell & foundation models

TZ

Dr. Tehseen Zia

Associate Professor, COMSATS University Islamabad

Co-PI at Medical Imaging and Diagnostics Lab (MIDL) · Master's Supervisor

Publications

Selected Works

Publications will be listed here as they become available. Visit my Google Scholar profile for the most up-to-date list of research contributions.

View Google Scholar

Expertise

Skills & Technologies

Deep Learning Computer Vision PyTorch TensorFlow Python Single-Cell Analysis Medical Image Analysis Explainable AI (XAI) Mechanistic Interpretability Foundation Models Virtual Cell Modeling Transfer Learning Vision Transformers R Data Science Omics Data Analysis

Get in Touch

Let's Collaborate

Interested in research collaboration, academic discussions, or AI consulting? I'd love to hear from you.

Send Email LinkedIn