Key result
The automated machine learning model identified scoliosis from total body DXA scans with a sensitivity of 86.5%, specificity of 96.9%, and an AUC of 0.80 compared to manual annotation.
Why the study?
Manual identification of scoliosis from DXA scans needed automation to enable full utilization of population-based research cohorts without relying on high-radiation radiographs.
Does an automated machine learning method accurately identify scoliosis from total body DXA scans compared to manual annotation?
Observational (n=17,000)
No
Does an automated machine learning method accurately identify scoliosis from total body DXA scans compared to manual annotation?
An automated machine learning method can accurately and reliably identify scoliosis from total body DXA scans, enabling large-scale population-based research without the need for manual annotation.
Answers are grounded in this paper and cite their sources.
Supports population scoliosis research via DXA; extends methodological options but leaves clinical adoption open.
Scoliosis is a 3D-torsional rotation of the spine, but risk factors for initiation and progression are little understood. Research is hampered by lack of population-based research since radiographs cannot be performed on entire populations due to the relatively high levels of ionising radiation. Hence we have developed and validated a manual method for identifying scoliosis from total body dual energy X-ray absorptiometry (DXA) scans for research purposes. However, to allow full utilisation of population-based research cohorts, this needs to be automated. The purpose of this study was therefore to automate the identification of spinal curvature from total body DXA scans using machine learning techniques. To validate the automation, we assessed: (1) sensitivity, specificity and area under the receiver operator curve value (AUC) by comparison with 12,000 manually annotated images; (2) reliability by rerunning the automation on a subset of DXA scans repeated 2-6 weeks apart and calculating the kappa statistic; (3) validity by applying the automation to 5000 non-annotated images to assess associations with epidemiological variables. The final automated model had a sensitivity of 86.5%, specificity of 96.9% and an AUC of 0.80 (95%CI 0.74-0.87). There was almost perfect agreement of identification of those with scoliosis (kappa 0.90). Those with scoliosis identified by the automated model showed similar associations with gender, ethnicity, socioeconomic status, BMI and lean mass to previous literature. In conclusion, we have developed an accurate and valid automated method for identifying and quantifying spinal curvature from total body DXA scans.
No takes yet. Share an insight, caveat, or question.
Jamaludin et al. (2020) conducted an observational in Scoliosis (n=17,000). Automated machine learning method for DXA scans vs. Manual DXA Scoliosis Method (DSM) annotation was evaluated on Area under the receiver operator curve (AUC) for identifying scoliosis (AUC 0.80, 95% CI 0.74-0.87). The automated machine learning model identified scoliosis from total body DXA scans with a sensitivity of 86.5%, specificity of 96.9%, and an AUC of 0.80 compared to manual annotation.
Synapse has enriched 10 related papers. Consider them for comparative context: