ÌÇÐÄTV

The Predictive Systems in Biomedicine (PRISM) Lab, led by Dr Fayyaz Minhas at the University of ÌÇÐÄTV, develops machine learning and artificial intelligence methods to understand complex biological and clinical systems. Our vision is to build algorithms behind the cure by building methods and applications that transform large and diverse biomedical datasets into insights that advance biological discovery and improve patient care.

The PRISM Lab is based in the Department of Computer Science and works closely with the Tissue Image Analytics Centre. We collaborate with clinicians, biologists, and industry partners worldwide to translate advances in AI into biomedical discovery, clinical decision support, and therapeutic innovation.

Research Programme

Our research develops machine learning frameworks that transform complex biomedical data into actionable scientific and clinical insight. We work across computational pathology, oncology, laboratory medicine, bioinformatics, and biomedical signal analysis, spanning scales from proteins and small molecules to cells, tissues, and patients, with a focus on diagnostic, prognostic, and predictive biomarker discovery.

We develop both core methodological advances and their application to real-world biomedical and clinical problems. This work is organised into two complementary themes:

Core Methodological Approaches for Biomedicine

• Learn: Multimodal and Spatial Representation Learning and Predictive Modelling for Biomedicine
  Developing machine learning models for predictive modelling and inference across heterogeneous data types, including images, spatial omics, genomic data, clinical records, and biomedical signals. This includes learning unified multimodal representations that capture biological structure and variability across scales. Specifically, we focus on building geometric and temporal models to capture spatial organisation, cell–cell interactions, tissue architecture, and disease progression over time, enabling the discovery of spatial and dynamic biomarkers.

• Explain: Interpretable, Causal, and Robust AI in Biomedicine
  Designing models that are transparent, reliable, and clinically meaningful, using causal representation learning, confounder-aware modelling, and principled interpretability to ensure robust and generalisable predictions across datasets, institutions, and populations.

• Deploy: Translational AI and Agentic Systems
  Developing AI systems that integrate modelling, reasoning, and decision-making into end-to-end workflows, enabling reproducible analysis pipelines and deployment in scientific and clinical settings.

Biomedical Applications and Translation

Applying these methods to high-impact problems across:

• Computational Pathology and Oncology
  Biomarker discovery, risk stratification, and prediction of disease progression and treatment response from histology and multimodal data (e.g., for colorectal cancer, analysis of fertility, etc.)

• Bioinformatics and Computational Biology (Proteins and Small Molecules)
  Modelling molecular/protein structure, interactions, and function for biomarker discovery and therapeutic insight esp. for antimicrobial discovery and protein-protein and protein-small molecule interactions.

• Laboratory Medicine
  Predictive modelling from laboratory test data and clinical records for diagnosis, monitoring, and risk prediction.

• Biomedical Signal Analysis
  Analysis of physiological and acoustic signals, including cardiac electrophysiology (e.g., ventricular tachycardia) and respiratory sounds (e.g., cough), for detection, characterisation, and clinical decision support.

For recent research talks, please see this page.

For publications and patents, please visit

Funded Projects Led by PRISM Lab (Total ÌÇÐÄTV Allocation: ~£1.5M)

1. SpyGlass: Machine Learning Guided Discovery of Correlation Structures in Cross-modal Omics Data
   GSK PhD Studentship
   Duration: Jan 2026 – Jul 2029
   Funding: £199,840 (ÌÇÐÄTV Allocation Only)
2. Kikohozi COUGH APP – AI/ML-based Cough Audio Classifier
   MRC Applied Global Health Research Programme
   Collaborators: Griffiths, Anyanwu, AKU Tanzania & Kenya
   Duration: Sep 2025 – Feb 2029
   Funding: £250,400.74 (ÌÇÐÄTV Allocation)
3. PREEMPTR: Predictive Recommendation Engine for Early Event Monitoring, Proactive Testing and Result Interpretation
   Interdisciplinary Research Spotlights (IRDF)
   Duration: Apr 2025 – Jul 2025
   Funding: £15,000
4. FERRIQ: AI-driven Mapping of Disease Trajectories in Genetic Iron Overload
   EPSRC Network+ (Future Blood Testing for Personalised Analytics)
   Duration: May 2023 – Apr 2024
   Funding: £50,000
5. Causal Modelling with Graph Neural Networks for Personalised Medicine in Computational Pathology
   EPSRC New Investigator Award (EP/W02909X/1)
   Duration: Jul 2022 – Jun 2025
   Funding: £376,202
6. Machine Learning for Early Detection of Mesothelioma from Pleural Fluid Data
   Asthma + Lung UK – Mesothelioma Early Diagnosis Programme
   Duration: Sep 2022 – 2025
   Funding: £377,054
7. Vibrational Spectroscopy for Spatial Disease Analysis
   EPSRC CLIRPath-AI (EP/W00058X/1)
   Duration: Sep 2022 – Jun 2023
   Funding: £50,000 (ÌÇÐÄTV component)
8. Breast Cancer Receptor Status Prediction using Spectroscopy and AI
   EPSRC CLIRPath-AI (EP/W00058X/1)
   Duration: Sep 2022 – Jun 2023
   Funding: £50,000 (ÌÇÐÄTV component)
9. PRISM: Identification of Pre-neoplastic Signatures in Mesothelioma
   Cancer Research UK Early Detection Innovation Award
   Duration: Jun 2020 – Mar 2023
   Funding: £97K total (£50K ÌÇÐÄTV)
10. GSK-funded PhD Studentship
    GSK
    Duration: 2019 – 2023
    Funding: ~£150K (ÌÇÐÄTV)

Collaborative Projects

1. ÌÇÐÄTV Co-I in NIHR AAC AI award AI_AWARD02688 on "COBIx: Multi-site validation of automated AI tool for screening of large bowel endoscopic biopsy slides" 01/2023-01 to 12/2026 (3 years) (Total funding amount £ 2,599,963, ÌÇÐÄTV allocation: £1,169,014.46)
2. Collaborator on Cancer Research UK Early Detection Committee -Innovation Award titled Haem-AI "Advanced early detection of myeloproliferative neoplasms (MPN) using digital image analysis, computational pathology and machine learning" (Ref: C68644/A30721 dated April 1 2020, duration 1 year). Role: Listed Co-Investigator/Collaborator (total allocation: £100K).
3. Co-I and Collaborator on Wellcome Institutional Strategic Support Fund Multi-modal Data Integration grant on "Coupling Spatial Transcriptomics, Histology and Proteomics to deconvolve multicellular spatial spots" Led by University of Manchester (June ‘22 to Dec ‘22, Total allocation: £24K)
4. Pathology image data Lake for Education, Analytics, and Discovery (PathLAKE and PathLAKE Plus Follow-up) 01 Jan 2019 (duration: 51 months): (ÌÇÐÄTV allocation £2,275,277)
5. PathLAKE+ ÌÇÐÄTV Grant Amount £313,917 funded by Innovate UK, Role: Researcher Assistant Professor and Co-I in PathLAKE+ .
6. TIAToolbox (Led by Shan Raza) An open‑source computational pathology library

Prospective Students: If you are aprospective PhD, Masters or undergraduate studentinterested in working in the PRISM lab, please click here for details.Link opens in a new window

Email Contact:fayyaz.minhas@warwick.ac.uk

Current Members

Lead Investigator

• Dr. Fayyaz Minhas

Post-Doctoral Fellows

• Dr. Mark Eastwood
• Dr. Ruqayya Awan
• Dr. Haoran Ni (via CAMaCS)
• Dr. Julio Hurtado (via CAMaCS)

Current PhD Students

• Piotr Keller
• George Wright
• Anja Estermann
• Agrima Agarwal (Co-Supervision with Prof. Emma McPherson)
• Joseph Mayer (Co-Supervision with Prof. Tarvinder Dhanjal)

Past PhD Students

• Dr. Muhammad Dawood
• Dr. Srijay Deshpande
• Dr. John Pocock
• Dr. Rawan AlBusayli
• Dr. Adiba Yaseen
• Dr. Amina Asif
• Dr. Wajid Abbasi
• Dr. Sadaf Gull
• Dr. Hira Kamal

Other Alumni

• Ammar Khairi (Research Assistant)