3D Markerless asymmetry analysis in the management of adolescent idiopathic scoliosis

Abstract

BackgroundThree dimensional (3D) markerless asymmetry analysis was developed to assess and monitor the scoliotic curve. While the developed surface topography (ST) indices demonstrated a strong correlation with the Cobb angle and its change over time, it was reported that the method requires an expert for monitoring the procedure to prevent misclassification for some patients. Therefore, this study aimed at improving the user-independence level of the previously developed 3D markerless asymmetry analysis implementing a new asymmetry threshold without compromising its accuracy in identifying the progressive scoliotic curves.MethodsA retrospective study was conducted on 128 patients with Adolescent Idiopathic Scoliosis (AIS), with baseline and follow-up radiograph and surface topography assessments. The suggested “cut point” which was used to separate the deformed surfaces of the torso from the undeformed regions, automatically generated deviation patches corresponding to scoliotic curves for all analyzed surface topography scans.ResultsBy changing the “cut point” in the asymmetry analysis for monitoring scoliotic curves progression, the sensitivity for identifying curve progression was increased from 68 to 75%, while the specificity was decreased from 74 to 59%, compared with the original method with different “cut point”.ConclusionsThese results lead to a more conservative approach in monitoring of scoliotic curves in clinical applications; smaller number of radiographs would be saved, however the risk of having non-measured curves with progression would be decreased.

Combining 3D Structured Light Imaging and Spine X-ray Data Improves Visualization of the Spinous Lines in the Scoliotic Spine

“…Studies using 3D ST have often been used to find a measurement method that produces results parallel to the Cobb angle measured by conventional radiography [37,[42][43][44][45][46][47][48][49][50][51][52]. Typically, the results of this approach are less than moderately accurate due to the different measures.…”

Decreased Vertical Trunk Inclination Angle and Pelvic Inclination as the Result of Mid-High-Heeled Footwear on Static Posture Parameters in Asymptomatic Young Adult Women

“…Radiographic measurements of SVA, PT, LL, SS, or PI are based on radiographic points assessed on the full spine radiograms [72], which are inaccessible in surface measurements. Parameters extracted by means of recently developed non-invasive optical systems (including moiré projection method and surface topography) are frequently novel and specific for each device [12,14,47,82,88,99,101,102,103]. Some of the indexes are recommended by the SOSORT [60].…”

Performance of Surface Topography Systems in Scoliosis Management: A Narrative Review

“…34 In a later study by the same team, the accuracy of the classification tree was improved after adding more training data (n = 128) and improving the algorithm for finding the best plane of symmetry. 54,55 Other machine learning methods were examined, and the K-nearest neighborhood and neural network were found to be the most effective methods for predicting the curve severity. 56 Frerich et al 41 used DIERS Formetric 4D system and defined ST parameters, including the scoliosis angle that was correlated to the Cobb angle in the lumbar (r = 0.76, prediction deviation = 9.42°) and thoracic (r = 0.87, prediction deviation = 6.98°) regions.…”