A recent study published in JAMA Network Open suggests that a smartphone app utilizing a Deep Learning Model can be effective in managing adolescent idiopathic scoliosis (AIS). The app, called ScolioNets, uses a validated deep learning model to classify the severity and curve type of AIS and to identify progression.
Related Articles
The research, conducted by Teng Zhang, Ph.D., from the University of Hong Kong, and colleagues, involved analyzing radiographs and smartphone photographs of the backs of adolescent patients at spine clinics. The app’s deep learning model was developed and validated in a training cohort and then integrated into the AlignProCARE open platform app.
The study included a total of 1,780 patients in the training data set, with 378 patients and 376 follow-ups in the prospective testing data sets. The results showed that the model recommended follow-up and considering surgery with high accuracy, as indicated by area under the receiver operating characteristic curves (AUCs) of 0.839 and 0.902, respectively. The model also demonstrated good ability to differentiate between different curve types (AUCs of 0.777, 0.760, and 0.860 for thoracic, thoracolumbar or lumbar, and mixed curve types, respectively). Additionally, the model successfully identified participants with or without curve progression (AUC of 0.757).
The deep learning model proved to be more effective in recognizing severities and curve types compared to both a senior and junior surgeon. The authors of the study highlight that the app’s ability to continuously monitor scoliosis without additional radiation exposure and with minimal cost could lead to prompt interventions when progression is detected.
In conclusion, the use of a smartphone app with a deep learning model shows great promise for managing adolescent idiopathic scoliosis. It provides a non-invasive and cost-effective way to monitor the condition and trigger appropriate interventions when necessary.
The post Smartphone App Using Deep Learning Model Helps Manage Adolescent Idiopathic Scoliosis appeared first on TS2 SPACE.
