This GitHub project aims to develop models for detecting and classifying degenerative spine conditions using lumbar spine MRI images, simulating the diagnostic performance of radiologists.
Low back pain is a prevalent health issue globally, impacting millions each year. Often, it is symptomatic of degenerative spine conditions like spondylosis, characterized by disc degeneration and spinal canal narrowing. Magnetic Resonance Imaging (MRI) provides critical insights into these conditions, aiding in accurate diagnosis and treatment planning.
This project, a collaboration between Radiological Society of North America (RSNA) and American Society of Neuroradiology (ASNR), focuses on leveraging artificial intelligence to enhance the detection and classification of degenerative spine conditions using lumbar spine MRI images. The goal is to develop models that can accurately identify five specific conditions across different intervertebral disc levels.
The dataset for this project is meticulously curated, sourced from eight global sites and spanning five continents. It includes severity scores for each condition at various disc levels. This comprehensive dataset promises to standardize classification methods and facilitate the development of automated diagnostic tools for degenerative lumbar spine conditions.
Tyler Richards, Jason Talbott, Robyn Ball, Errol Colak, Adam Flanders, Felipe Kitamura, John Mongan, Luciano Prevedello, Maryam Vazirabad.. (2024). RSNA 2024 Lumbar Spine Degenerative Classification. Kaggle. https://kaggle.com/competitions/rsna-2024-lumbar-spine-degenerative-classification