Transformer model predicts spine curve progression in adolescents
Predicting the Progression of Adolescent Idiopathic Scoliosis
Computer Vision and Pattern Recognition
Summary
Adolescent Idiopathic Scoliosis is a condition where the spine curves sideways during adolescence for unknown reasons. This can cause pain and worsen quickly as teens grow. The authors created a computer model that looks at spine shapes from special scans to predict how the curve will change from age 9 to 24. They trained the model on many made-up spine curves, then showed it works well on real scan data, especially after some extra training with that real data. This system can predict how scoliosis develops or if a spine remains normal over time.
What this means in practice
- •For orthopedic clinicians: Use model predictions from DXA scans to forecast scoliosis progression and guide patient treatment plans over adolescence.
- •For medical device developers: Integrate predictive modeling into spine imaging software to provide early warnings of scoliosis worsening during growth periods.$Commercial implications: Enables development of clinical decision support tools that can be marketed to hospitals and clinics managing scoliosis care.
Authors
Owen Pullen, Amir Jamaludin, Andrew Zisserman
Abstract
Adolescent Idiopathic Scoliosis is defined as a lateral curvature of the spine that develops during adolescence, without known cause. The condition can result in significant pain and disability, and often progresses rapidly during adolescence. The objective of this paper is to predict the progression of the condition in a temporal sequence from ages 9 to 24, as measured from a sequence of Dual X-ray Absorptiometry (DXA) scans. To this end, we train a transformer model that takes in the curve of the spine to predict curve progression. The model is trained using a large-scale synthetic dataset of spine curves and their time series, covering different curve types and different progression patterns. We show that the model is able to generalise from synthetic to real data by evaluating it on a dataset of real DXA scans covering multiple time points. We find that fine-tuning the model on real data gives a significant boost to performance. The model is able to accurately predict spine curve progression in both scoliosis and normal cases.