3D Radiological Outcomes and Quality of Life of Patients With Moderate Idiopathic Scoliosis Treated With Anterior Vertebral Growth Modulation Versus Bracing: Two-Year Follow-up.

STUDY DESIGN: Observational cohort study. OBJECTIVE: To test the hypothesis that anterior vertebral body growth modulation (AVBGM) achieves 3D deformity correction after 2-year follow-up while brace treatment limits curve progression for moderate idiopathic scoliosis (30-50°). SUMMARY OF BACKGROUND DATA: For idiopathic scoliosis, bracing and AVBGM have overlapping indications in skeletally immature patients with moderate scoliosis curve angles, creating a grey zone in clinical practice between them. The relative 3D deformity control performance over a 2-year period between these fusionless treatments is still uncertain. METHODS: A retrospective review of a prospective idiopathic scoliosis patients database, recruited between 2013 and 2018 was performed. Inclusion criteria were skeletally immature patients (Risser 0-2), with Cobb angles between 30° and 50° and a 2-year follow-up after bracing or AVBGM. 3D radiological parameters and health related quality of life (HRQoL) scores were evaluated. Unpaired t test was used. RESULTS: Thirty nine patients (12.7 ± 1.3 y.o.) with Cobb angles more than or equal to 30° treated with brace and 41 patients (11.8 ± 1.2 y.o.) with presenting Cobb angles less than or equal to 50° who received AVBGM were reviewed. The statistical analysis of 3D deformity measurements showed that at 2-year follow-up, only the 3D spine length and both sides apical vertebral heights changed significantly with brace treatment. While AVBGM treatment achieved statistically significant correction differences in thoracic and lumbar Cobb angles, TrueKyphosis, 3D spine length, and selective left apical vertebral height ( P < 0.05). 35% of brace patients had a curve progression of more than 5° at final follow-up while it was 0% for AVBGM. HRQoL assessment showed no statistically significant differences between pre and post SRS-22 total scores for each group ( P > 0.05). CONCLUSION: Even though these two cohorts are not fully comparable, bracing seems to control progression for a significant portion of patients with moderate scoliosis curves, while AVBGM significantly corrected and maintained 3D deformity parameters at 2-year follow-up.

Automatic bone maturity grading from EOS radiographs in Adolescent Idiopathic Scoliosis.

Adolescent Idiopathic Scoliosis (AIS) is a deformation of the spine and it is routinely diagnosed using posteroanterior and lateral radiographs. The Risser sign used in skeletal maturity assessment is commonly accepted in AIS patient's management. However, the Risser sign is subject to inter-observer variability and it relies mainly on the observation of ossification on the iliac crests. This study proposes a new machine-learning-based approach for Risser sign skeletal maturity assessment using EOS radiographs. Regions of interest including right and left humeral heads; left and right femoral heads; and pelvis are extracted from the radiographs. First, a total of 24 image features is extracted from EOS radiographs using a ResNet101-type convolutional neural network (CNN), pre-trained from the ImageNet database. Then, a support vector machine (SVM) algorithm is used for the final Risser sign classification. The experimental results demonstrate an overall accuracy of 84%, 78%, and 80% respectively for iliac crests, humeral heads, and femoral heads. Class activation maps using Grad-CAM were also investigated to understand the features of our model. In conclusion, our machine learning approach is promising to incorporate a large number of image features for different regions of interest to improve Risser grading for skeletal maturity. Automatic classification could contribute to the management of AIS patients.

A freehand ultrasound framework for spine assessment in 3D: a preliminary study.

X-ray imaging is currently the gold standard for the assessment of spinal deformities. The purpose of this study is to evaluate a freehand 3D ultrasound system for volumetric reconstruction of the spine. A setup consisting of an ultrasound scanner with a linear transducer, an electromagnetic measuring system and a workstation was used. We conducted 64 acquisitions of US images of 8 adults in a natural standing position, and we tested three setups: 1) Subjects are constrained to be close to a wall, 2) Subjects are unconstrained, and 3) Subjects are constrained to performing fast and slow acquisitions. The spinous processes were manually selected from the volume reconstruction from tracked ultrasound images to generate a 3D point-based model depicting the centerline of the spine. The results suggested that a freehand 3D ultrasound system can be suitable for representing the spine. Volumetric reconstructions can be computed and landmarking can be performed to model the surface of the spine in the 3D space. These reconstructions promise to generate computer-based descriptors to analyze the shape of the spine in the 3D space.Clinical Relevance- We provide clinicians with a protocol that could be integrated in clinical setups for the assessment and monitoring of AIS, based on US image acquisitions, which constitutes a radiation-free technology.

Convolutional Neural Networks for Automatic Risser Stage Assessment.

PURPOSE: To develop an automatic method for the assessment of the Risser stage using deep learning that could be used in the management panel of adolescent idiopathic scoliosis (AIS). MATERIALS AND METHODS: In this institutional review board approved-study, a total of 1830 posteroanterior radiographs of patients with AIS (age range, 10-18 years, 70% female) were collected retrospectively and graded manually by six trained readers using the United States Risser staging system. Each radiograph was preprocessed and cropped to include the entire pelvic region. A convolutional neural network was trained to automatically grade conventional radiographs according to the Risser classification. The network was then validated by comparing its accuracy against the interobserver variability of six trained graders from the authors' institution using the Fleiss κ statistical measure. RESULTS: Overall agreement between the six observers was fair, with a κ coefficient of 0.65 for the experienced graders and agreement of 74.5%. The automatic grading method obtained a κ coefficient of 0.72, which is a substantial agreement with the ground truth, and an overall accuracy of 78.0%. CONCLUSION: The high accuracy of the model presented here compared with human readers suggests that this work may provide a new method for standardization of Risser grading. The model could assist physicians with the task, as well as provide additional insights in the assessment of bone maturity based on radiographs.© RSNA, 2020.

Prediction of spinal curve progression in Adolescent Idiopathic Scoliosis using Random Forest regression.

BACKGROUND: The progression of the spinal curve represents one of the major concerns in the assessment of Adolescent Idiopathic Scoliosis (AIS). The prediction of the shape of the spine from the first visit could guide the management of AIS and provide the right treatment to prevent curve progression. METHOD: In this work, we propose a novel approach based on a statistical generative model to predict the shape variation of the spinal curve from the first visit. A spinal curve progression approach is learned using 3D spine models generated from retrospective biplanar X-rays. The prediction is performed every three months from the first visit, for a time lapse of one year and a half. An Independent Component Analysis (ICA) was computed to obtain Independent Components (ICs), which are used to describe the main directions of shape variations. A dataset of 3D shapes of 150 patients with AIS was employed to extract the ICs, which were used to train our approach. RESULTS: The approach generated an estimation of the shape of the spine through time. The estimated shape differs from the real curvature by 1.83, 5.18, and 4.79° of Cobb angles in the proximal thoracic, main thoracic, and thoraco-lumbar lumbar sections, respectively. CONCLUSIONS: The results obtained from our approach indicate that predictions based on ICs are very promising. ICA offers the means to identify the variation in the 3D space of the evolution of the shape of the spine. Another advantage of using ICs is that they can be visualized for interpretation.

Dynamic ensemble selection of learner-descriptor classifiers to assess curve types in adolescent idiopathic scoliosis.

While classification is important for assessing adolescent idiopathic scoliosis (AIS), it however suffers from low interobserver and intraobserver reliability. Classification using ensemble methods may contribute to improving reliability using the proper 2D and 3D images of spine curvature features. In this study, we present two new techniques to describe the spine, namely, leave-one-out and fan leave-one-out. Using these techniques, three descriptors are computed from a stereoradiographic 3D reconstruction to describe the relationship between a vertebra and its neighbors. A dynamic ensemble selection method is introduced for automatic spine classification. The performance of the method is evaluated on a dataset containing 962 3D spine models categorized according to three curve types. With a log loss of 0.5623, the dynamic ensemble selection outperforms voting and stacking ensemble learning techniques. This method can improve intraobserver and interobserver reliability, identify the best combination of descriptors for characterizing spine curve types, and provide assistance to clinicians in the form of information to classify borderline curvature types. Graphical abstract ᅟ.

Reliability and accuracy analysis of a new semiautomatic radiographic measurement software in adult scoliosis.

STUDY DESIGN: Radiographic software measurement analysis in adult scoliosis. OBJECTIVE: To assess the accuracy as well as the intra- and interobserver reliability of measuring different indices on preoperative adult scoliosis radiographs using a novel measurement software that includes a calibration procedure and semiautomatic features to facilitate the measurement process. SUMMARY OF BACKGROUND DATA: Scoliosis requires a careful radiographic evaluation to assess the deformity. Manual and computer radiographic process measures have been studied extensively to determine the reliability and reproducibility in adolescent idiopathic scoliosis. Most studies rely on comparing given measurements, which are repeated by the same user or by an expert user. A given measure with a small intra- or interobserver error might be deemed as good repeatability, but all measurements might not be truly accurate because the ground-truth value is often unknown. Thorough accuracy assessment of radiographic measures is necessary to assess scoliotic deformities, compare these measures at different stages or to permit valid multicenter studies. METHODS: Thirty-four sets of adult scoliosis digital radiographs were measured two times by three independent observers using a novel radiographic measurement software that includes semiautomatic features to facilitate the measurement process. Twenty different measures taken from the Spinal Deformity Study Group radiographic measurement manual were performed on the coronal and sagittal images. Intra- and intermeasurer reliability for each measure was assessed. The accuracy of the measurement software was also assessed using a physical spine model in six different scoliotic configurations as a true reference. RESULTS: The majority of the measures demonstrated good to excellent intra- and intermeasurer reliability, except for sacral obliquity. The standard variation of all the measures was very small: ≤ 4.2° for Cobb angles, ≤ 4.2° for the kyphosis, ≤ 5.7° for the lordosis, ≤ 3.9° for the pelvic angles, and ≤5.3° for the sacral angles. The variability in the linear measurements (distances) was <4 mm. The variance of the measures was 1.7 and 2.6 times greater, respectively, for the angular and linear measures between the inter- and intrameasurer reliability. The image quality positively influenced the intermeasurer reliability especially for the proximal thoracic Cobb angle, T10-L2 lordosis, sacral slope and L5 seating. The accuracy study revealed that on average the difference in the angular measures was < 2° for the Cobb angles, and < 4° for the other angles, except T2-T12 kyphosis (5.3°). The linear measures were all <3.5 mm difference on average. CONCLUSION: The majority of the measures, which were analyzed in this study demonstrated good to excellent reliability and accuracy. The novel semiautomatic measurement software can be recommended for use for clinical, research or multicenter study purposes.

Preliminary evaluation of a computer-assisted tool for the design and adjustment of braces in idiopathic scoliosis: a prospective and randomized study.

STUDY DESIGN: Prospective and randomized clinical study. OBJECTIVES: To evaluate the correction of the spine obtained using a 3-dimensional visualization software tool developed to assist the design and adjustment of braces compared with the correction obtained with the conventional method in a cohort of subjects with adolescent idiopathic scoliosis (AIS). SUMMARY OF BACKGROUND DATA: The optimal design and adjustment of trim lines, pad placement, and areas of relief for the Boston brace system in AIS are currently done using clinical examination and coronal radiographs. Correction of spinal curves in the coronal plane has been achieved with this technique, but 3-dimensional correction has yet to be demonstrated. METHODS: Forty-eight consecutive subjects with AIS requiring treatment with a Boston brace were prospectively entered in the study. For 24 patients (test group), brace design and adjustment was obtained using the computer-assisted tool combining surface topography, surface pressure measurement, and 3-dimensional reconstructions of the trunk, while design and adjustment for the remaining subjects (control group) was done in the conventional manner. Immediate in-brace correction of the spine at the initial visit was compared in both groups. RESULTS: Both groups were comparable in terms of age, sex, curve type, and average deformity in both the coronal and sagittal planes. The average prebrace thoracic deformity was 35 degrees of Cobb angle, while the average lumbar curve was 32 degrees in the test group and 35 degrees in controls. A statistically and clinically significant improvement in correction of coronal curves and of curves in the plane of maximal deformity was found for both thoracic and lumbar curves in both groups, but the improvement was significantly greater in the test group. The average in-brace correction in the test group was 12 degrees +/- 7 degrees compared with 7 degrees +/- 5 degrees in the control group for thoracic curves, while the average in-brace correction in the test group was 10 degrees +/- 5 degrees compared with 6 degrees +/- 5 degrees in the control group for lumbar curves. Similar average corrections were detected in the plane of maximal deformity. In addition, a significant improvement in the orientation of the plane of maximum deformity from 37 degrees to 23 degrees for lumbar curves was noted only in the test group, indicating that true 3-dimensional correction by the brace was obtained in this group. CONCLUSION: It is possible to improve the design and adjustment of braces in AIS and to achieve 3-dimensional correction of scoliotic curves with the use of a computer-assisted tool allowing 3-dimensional visualization of the spinal curves and the external shape of the trunk.