SpineNet™
SpineNet™ is a validated AI platform that automates spine MRI analysis, replacing subjective, reader-dependent grading with consistent, reproducible, and scalable quantitative assessment.
SpineNet™ has been run on
27,000+ MRIs to date.
SpineNet™
SpineNet™ automates expert-level spine MRI interpretation. It eliminates reader variability and produces structured per-vertebra and per-disc data for clinical care, research, device development, registries, trials, and surgical planning.
The system detects and labels vertebrae on sagittal MRI sequences, grades multiple disc pathologies, and works on standard lumbar, cervical, and whole-spine scans without special protocols.
Developed and validated over a decade with peer-reviewed, award-winning research at the University of Oxford, SpineNet™ shows robust real-world performance.
[1] SpineNet: Automated Classification and Evidence Visualisation (Medical Image Analysis 41:63–73), 2017
https://pubmed.ncbi.nlm.nih.gov/28756059/
[2] Convolutional Approach to Vertebrae Detection and Labelling in Whole Spine MRI (MICCAI 2020)
https://dl.acm.org/doi/10.1007/978-3-030-59725-2_69
[3] Automated Detection, Labelling and Radiological Grading of Clinical MR Scans (Nature Scientific Reports) 2022 https://www.nature.com/articles/s41598-024-64580-w