View Proposal

Proposer
Marta Vallejo
Title
Enhancing Machine Learning Models to Analyse Immune Profiles in ALS Tissue (in collaboration with Aberdeen University)
Goal
Description
Amyotrophic Lateral Sclerosis (ALS) is a progressive neurodegenerative disease. Recent work has revealed that ALS patients can be grouped based on immune activity levels into two profiles: * NPS1: High adaptive immune response * NPS2: Low adaptive immune response This project builds on an ongoing collaboration with the Gregory Lab at the University of Aberdeen (https://gregorylaboratory.com/), which has used immunohistochemistry (IHC) on ALS brain and spinal cord tissues to explore these immune profiles. For generating the dataset, two key markers were used (POM121 (PA5-36498) and a novel in-house sCTLA-4 antibody (JMW-3B3), developed at Aberdeen). Image analysis was performed using QuPath, with features extracted via superpixel and nuclear segmentation. A simple linear model has already been implemented. The next step is to extend this model using more advanced techniques. If successful, this project could identify interpretable imaging biomarkers that distinguish immune profiles in ALS tissue, contributing to personalised disease classification and supporting future clinical decision-making or therapeutic targeting. Aims - Improve the baseline linear model using more robust and interpretable methods - Investigate which features (e.g. nuclear shape, intensity, spatial patterns) best separate NPS1 and NPS2 - Explore cross-region patterns (e.g. spinal cord vs. motor cortex) - Optionally: test models for sCTLA-4 prediction or unsupervised clustering Tools & Skills - Python - Scikit-learn / XGBoost / PyTorch (for modelling) - QuPath feature understanding (dataset already extracted) - Experience with pandas, NumPy, and plotting libraries is a plus Support & Collaboration - Direct contact with researchers from the University of Aberdeen and the Gregory Lab - Support from supervisors involved in the clinical publication effort - Opportunity to co-author a peer-reviewed journal paper if results are strong Who is this for? - Students interested in machine learning in healthcare or medical imaging - Those keen to work on a real dataset with clear clinical relevance and publication potential - Ideal for Honours or MSc-level individual projects
Resources
Dataset: You will be provided with 41 ALS samples: 1.- Spinal cord: 20 cases (10 NPS1, 10 NPS2) 2.- Motor cortex: 21 cases (10 NPS1, 10 NPS2, 1 unclassified) 3.- Image-derived features: extracted from IHC-stained tissue using QuPath 4.- Metadata including region, immune profile, and antibody marker
Background
Url
Difficulty Level
Variable
Ethical Approval
Full
Number Of Students
2
Supervisor
Marta Vallejo
Keywords
machine learning, deep learning, healthcare
Degrees
Bachelor of Science in Computer Science
Master of Science in Artificial Intelligence
Master of Science in Artificial Intelligence with SMI
Master of Science in Computing (2 Years)
Master of Science in Data Science
Bachelor of Science in Computing Science
Postgraduate Diploma in Artificial Intelligence
Bachelor of Science in Statistical Data Science
BSc Data Sciences
Back to List