SCIseg · Traumatic Spinal Cord Injury: The Need to Classify Disease Severity
Horizon Europe — Marie Skłodowska-Curie Actions
- Duration
- 2023-05-01 → 2027-04-30
- EU contribution
- €269,047
- Participants
- 2
- Scheme
- HORIZON-TMA-MSCA-PF-GF
Lines connect the coordinator with its partners.
Results in brief
Traumatic Spinal Cord Injury: The Need to Classify Disease Severity
This project aims to develop an automatic and reproducible analysis tool for magnetic resonance imaging (MRI) images based on machine learning and deep learning techniques, with the goal of extracting clinically relevant metrics to improve the management of patients with traumatic spinal cord injury (SCI). Each year, traumatic SCI affects between 250,000 and 500,000 individuals worldwide, often resulting from motor vehicle accidents, falls, or sports-related injuries. Traumatic SCI frequently causes severe neurological impairments, leading to a significant reduction in patients’ quality of life and imposing a substantial economic burden on healthcare systems. Although MRI examination is routinely performed in patients with traumatic SCI, its full potential remains underutilized due to the complexity of image analysis and the variability of MRI data across institutions. Deep learning, a field of artificial intelligence (AI), offers promising solutions by enabling automatic annotation, known as segmentation, of structures such as the spinal cord or lesions. This helps reduce inter-rater variability and supports the analysis of large, multi-center SCI cohorts. Quantitative MRI biomarkers derived using deep learning-based methods have already demonstrated strong correlations with clinical measures. Despite these advantages, deep learning applications in the context of traumatic SCI remain underexplored, and no open-source tools are currently available. To address this gap, the project focuses on developing an automatic, reproducible pipeline for MRI processing and metric extraction to support clinical decision-making in traumatic SCI.
Data: CORDIS, © European Union
Project objective
Traumatic spinal cord injury (tSCI) markedly reduces patients’ quality of life and economically burdens health systems. Neurological examinations and clinical magnetic resonance imaging (MRI) scans are currently insufficient for the proper classification of the tSCI baseline level (i.e., severity). Although MRI scans are routinely employed in tSCI patients, the MRI potential is not fully utilised due to the complexity of the analysis and diversity of MRI data across hospitals. The aim of this project is to propose a fully automatic and reproducible analysis tool that could be run by clinicians to improve the clinical management of tSCI patients. First, deep learning models for automatic spinal cord and lesion segmentation from MRI images will be developed to go beyond the currently used error-prone and time-consuming manual segmentations. The models will be trained on a multi institutional MRI dataset to be robust to MRI data heterogeneity across hospitals. Then, quantitative measures of the tSCI severity will be automatically computed from the segmented structures (i.e., spinal cord and lesions) and employed within the statistical model to predict tSCI severity. Finally, the developed methodology will be translated to the real-world healthcare system and tested on a prospectively acquired dataset of tSCI patients. Importantly, deep learning models, analysis pipeline, and statistical model will be seamlessly integrated into the current state-of-the-art ecosystem for spinal cord MRI data analysis and made publicly available to facilitate open science and reproducibility across hospitals. The project will create the first step in the improvement of care and clinical management in millions of patients with tSCI worldwide. In the longer term, after demonstrating the clinical relevance of the proposed tools, we assume that advanced MRI-based methods will be adopted by the larger clinical community for more personalised care.
Original text from CORDIS.
Participants
Links
- View on CORDIS
- DOI: 10.3030/101107932
- https://ec.europa.eu/research/participants/documents/downloadPublic?documentIds=080166e503a24614&appId=PPGMS
Data: CORDIS, © European Union