Master Thesis: Multimodal Deep Learning for Predicting Osteoarthritis Pain Trajectories

11.07.2025, Abschlussarbeiten, Bachelor- und Masterarbeiten

Overview

Knee osteoarthritis (OA) affects over 650 million people globally. Although radiographic severity is commonly used for assessment, pain is the most impactful symptom and directly affects patients’ function and quality of life. Pain progression varies widely, indicating the presence of distinct pain trajectories.

This thesis aims to use deep learning with multimodal data to identify and predict individual OA pain progression patterns.

Objectives

• Identify Pain Trajectories: Use clustering techniques on 9-year longitudinal pain data to identify distinct progression patterns.
• Develop a Multimodal Prediction Model: Design a multimodal deep learning architecture combining baseline MRI, X-ray, and clinical data to predict pain trajectory membership.
• Clinical Interpretability: Identify the most relevant clinical and imaging features influencing pain trajectories.
(Tasks can be adapted based on student interests)

Prerequisites

• Advanced knowledge of deep learning with imaging data
• Strong Python and PyTorch/TensorFlow skills
• Experience with medical data and multimodal models is a plus, but not required

What We Offer

• Access to public and private OA datasets
• High-performance computing resources
• A collaborative and interdisciplinary research group consisting of doctors and computer scientists
• Workspace at the Department of Orthopedics and Sports Orthopedics at the University Hospital and the Institute for AI and Informatics in Medicine
• Potential for scientific publications
• Opportunity to contribute to clinically meaningful research in OA management

How to Apply

Preferred starting date: September 2025 (flexible)

Send an email to amna.gillani@tum.de with your CV and transcript.

References

• Panfilov, Egor, et al. "End-To-End Prediction of Knee Osteoarthritis Progression With Multimodal Transformers." IEEE Journal of Biomedical and Health Informatics (2025).
• Lee, Jinhee J., et al. "An ensemble clinical and MR-image deep learning model predicts 8-year knee pain trajectory: Data from the osteoarthritis initiative." Osteoarthritis Imaging 1 (2021): 100003.
• Nguyen, Huy Hoang, et al. "Clinically-inspired multi-agent transformers for disease trajectory forecasting from multimodal data." IEEE Transactions on Medical Imaging 43.1 (2023): 529-541.
• Tiulpin, Aleksei, et al. "Multimodal machine learning-based knee osteoarthritis progression prediction from plain radiographs and clinical data." Scientific Reports 9.1 (2019): 20038.

Kontakt: amna.gillani@tum.de