Deep learning-based precision phenotyping of spine curvature identifies novel genetic risk loci for scoliosis in the UK Biobank - PubMed
4 hours ago
- #Deep Learning
- #Genome-Wide Association Study
- #Spinal Curvature
- Deep learning was used to analyze spinal curvature from DXA scans in 57,588 UK Biobank participants, providing a quantitative measure of scoliosis severity that correlated well with clinical assessments.
- A genome-wide association study based on the imaging phenotype identified two novel genetic loci associated with scoliosis in a European population, located in SEM1/SHFM1 and near an lncRNA between EDEM1 and GRM7 on chromosome 3.
- The quantitative imaging GWAS revealed significant genetic correlations with previous ICD-10-based studies and detected more significant loci than a much larger case-control dataset, highlighting its enhanced power to uncover genetic risk factors for scoliosis.