Predicting curve progression for adolescent idiopathic scoliosis using random forest model

Abstract

Background Adolescent Idiopathic Scoliosis (AIS) is a three-dimensional (3D) spinal deformity characterized by coronal curvature and rotational deformity. Predicting curve progression is important for the selection and timing of treatment. Although there is a consensus in the literature regarding prognostic factors associated with curve progression, the order of importance, as well as the combination of factors that are most predictive of curve progression is unknown. Objectives (1) create an ordered list of prognostic factors that most contribute to curve progression, and (2) develop and validate a Machine Learning (ML) model to predict the final major Cobb angle in AIS patients. Methods 193 AIS patients were selected for the current study. Preoperative PA, lateral and lateral bending radiographs were retrospectively obtained from the Shriners Hospitals for Children. Demographic and radiographic features, previously reported to be associated with curve progression, were collected. Sequential Backward Floating Selection (SBFS) was used to select a subset of the most predictive features. Based on the performance of several machine learning methods, a Random Forest (RF) regressor model was used to provide the importance rank of prognostic features and to predict the final major Cobb angle. Results The seven most predictive prognostic features in the order of importance were initial major Cobb angle, flexibility, initial lumbar lordosis angle, initial thoracic kyphosis angle, age at last visit, number of levels involved, and Risser "+" stage at the first visit. The RF model predicted the final major Cobb angle with a Mean Absolute Error (MAE) of 4.64 degrees. Conclusion A RF model was developed and validated to identify the most important prognostic features for curve progression and predict the final major Cobb angle. It is possible to predict the final major Cobb angle value within 5 degrees error from 2D radiographic features. Such methods could be directly applied to guide intervention timing and optimization for AIS treatment.

3D Back Contour Metrics in Predicting Idiopathic Scoliosis Progression: Retrospective Cohort Analysis, Case Series Report and Proof of Concept

“…Alfraihat et al developed a random forest model to find the most important prognostic markers for progression of the curve angle and to predict Cobb angle [21]. The final random forest model is complex with 15 predictors: initial major Cobb angle, initial lumbar lordosis angle, initial thoracic kyphosis angle, age at initial follow-up, age at final follow-up, initial apical wedge angle, the time span between age at initial and final followups, flexibility, axial rotation, Lenke type, gender, brace status, apex location, number of levels involved in the curve, and Risser sign.…”

“…By using the backward variable selection, their model contained five significant predictors: skeletal maturity, curve classification, Cobb angle, the plane of maximal curvature, and 3D disk wedging (T3-T4, T8-T9). Recently, using machine learning methods and the random forest model, there have been attempts to evaluate curve dynamics over time [21,22]. However, all the methods mentioned above make the prediction based mainly on the radiological variables.…”

Patient-specific finite element modeling of scoliotic curve progression using region-specific stress-modulated vertebral growth

“…Eighth, genetic factors were not integrated in the current FE modeling approach. Numerous factors have been shown to correlate with AIS development and progression, and therefore future modeling framework development may benefit by integrating such relationships [ 53 , 54 ]. Ninth, no sex-specific stress sensitivity or rate-of-change of stress sensitivity has been established [ 55 ].…”

In vitro coupled motions of the whole human thoracic and lumbar spine with rib cage

“…[11][12][13] To the authors' knowledge, no quantitative data exist on in vitro region-specific coupled motions of the whole human TS and LS (T1-L5) with rib cage. Such kinematics data may be used to validate computational models of the spine [18][19][20] that predict the performance of implants and surgical interventions, and to better understand spinal biomechanics in deformity, [21][22][23] degeneration, and trauma [24][25][26][27][28] as well as the biofidelity of surrogate animal spine models. 29 Therefore, the objective of this in vitro biomechanical study is to quantify the coupled motions of the whole human TS and LS with rib cage.…”