Background/Objectives: The Spinal Appearance Questionnaire (SAQ) assesses the self-perception of appearance of adolescents with idiopathic scoliosis (AIS). Due to originally scoring too many domains and producing prevalent ceiling effects when treated conservatively, the SAQv1.1, has been introduced. This study aimed to compare the test–retest reliability, convergent validity, and ceiling and floor effects of the two versions of the SAQ in patients with AIS. Methods: Conservatively treated females with AIS, aged 10–18 years old, were consecutively recruited from a scoliosis clinic. The Scoliosis Research Society-22 refined (SRS-22r), SAQ, and SAQv1.1 were collected, in English, with radiographic measurements (Cobb angle, Coronal Balance, and Vertebral Rotations). Nine domain scores were obtained from the original SAQ. Appearance, Expectations, and Total scores were calculated for SAQ v.1.1. Questionnaires were re-administered electronically after one week. Results: One hundred females, aged 13.9 ± 1.8 years with curve angles of 28.8° ± 13.9°, were included. The test–retest reliability for SAQ varied between domains (ICC3,1 = 0.72 to 0.94). The Total, Appearance, and Expectation ICCs3,1 for the SAQv1.1 were 0.92, 0.94, and 0.86, respectively. Convergent validity was demonstrated between seven SAQ domains and the SRS-22r Total and Cobb angle (|r| = 0.32 to 0.59). The SAQv1.1 Total correlated with the SRS-22r Total (r = −0.50) and with the Cobb angle (r = 0.56). All SAQ domains presented ceiling (Curve = 11% to Kyphosis = 68%) and floor effects (Chest = 8% and Waist = 4%). The SAQv1.1 Total and Appearance had low ceiling effects (≤5%), while Expectations presented both ceiling (14%) and floor effects (10%). Conclusions: The SAQv1.1 is recommended because of its stronger reliability, superior convergent validity, and fewer ceiling and floor effects in AIS.

PURPOSE:Exercise stress tests (EST) are used to evaluate functional capacity before cardiac rehabilitation (CR). The prognostic value of EST to understand health outcomes in the year following an EST after participating in CR is not well understood. METHODS:We reviewed charts of 488 patients (females, n = 64) who underwent an EST within 100 days of starting CR. The outcomes of revascularization, rehospitalization, mortality, and the combination of revascularization or mortality were compared between patients with positive EST (n = 38), negative EST (n = 123), and nondiagnostic EST (n = 327). RESULTS:The positive EST group displayed significantly higher risk for revascularization or mortality within 1 year post-EST (HR = 13.6: 95% CI, 1.6-119.4; P = .018) compared with the negative EST group. Having a prior coronary artery bypass graft surgery independently increased the hazard of rehospitalization at 1 year (HR = 5.1: 95% CI, 1.9-13.5; P = .001) and the composite endpoint at 1 year (HR = 5.7: 95% CI, 1.6-19.7; P = .006) post-EST. The risk for rehospitalization in females (HR = 2.3: 95% CI, 1.0-5.3; P = .050) was greater than for males at 1 year, warranting future investigation into sex-based differences of long-term health outcomes post cardiac event. CONCLUSIONS:The utility of EST for patients referred to CR has been debated, but our results demonstrate that EST may provide valuable information for prognosis and decision-making in cardiac care pathways.

This study aimed to identify and validate prognostic factors to predict scoliosis progression. One hundred sixty-two girls, aged 13.5 ± 1.7 years old, diagnosed with idiopathic scoliosis were recruited. One hundred were used for model development, and 62 for testing, in which the number of progression cases was 25 and 11, respectively. All participants were scanned by an ultrasound (US) system for two consecutive visits (baseline and follow-up). The baseline parameters included (a) demographic: age, body mass index (BMI), and menarche status; (b) radiographic: X-ray Cobb angle, number-of-curve (NOC), and Risser sign; (c) ultrasonic: US Cobb, maximum axial vertebral rotation (AVR), Cobb angle at the plane of maximum curvature (PMC), kyphotic angle (KA) and reflection coefficient (RC) index. For the follow-up visit, only the US Cobb was recorded. The demographic and X-ray parameters were extracted from the scoliosis clinical records. The US parameters were measured by trained raters with good measurement reliability. The prediction model was developed using logistic regression analysis. The final predictors were US Cobb change, RC index, and NOC, and the probability of curve progression risk was Log (p/1-p) = -1.40 + 0.28(US Cobb change)-39.45(RC) + 1.36 (NOC). The model achieved sensitivity, specificity, and accuracy of 90

To determine the association of the ultrasound reflection coefficient index and other clinical parameters to predict curve progression in children with AIS. Sixty-six females (13.9 ± 1.5 years old) under observation with baseline Cobb angle (24.4 ± 10.4°) consented and participated. Besides the standard clinical procedures, all participants were scanned by an ultrasound (US) imager in a standing position. All participants had been followed, and the average follow-up visit was 7.5 ± 3.1 months. Five parameters were investigated, including the Cobb angle and the Risser sign measured from the radiographs, and the kyphotic angle (KA), the axial vertebral rotation (AVR), and the reflection coefficient (RC) index measured from the ultrasonographs. The curve was considered progressed when the Cobb angle measured on two consecutive radiographs was increased by more than 5°. Among the 5 parameters, the KA, RC index, and Risser sign were found to be significantly associated with progression (p < 0.05). Children who had KA ≤ 38° and RC ≤ 0.06 showed higher chances of progression (62

BACKGROUND:The increasing use of remote patient care for individuals with scoliosis has created opportunities for the development and utilization of different tools and methods of evaluation. The aim of this study was to determine the clinical postural variables evaluated in patients with scoliosis through telehealth. METHODS:Systematic searches were conducted by two independent reviewers in the PubMed, Scopus, EMBASE, and IEEE databases using the MeSH terms "scoliosis" and "telemedicine" and their synonyms. This study followed the MOOSE guideline (Meta-analysis of Observational Studies in Epidemiology). The methodological quality assessment was performed using the Critical Appraisal Skills Program (CASP). FINDINGS:136 studies were found in the databases, and 20 were found through manual search of references. After removing duplicates (n = 41), 81 studies were excluded, leaving 14 for full-text reading. Of these, six studies were included for qualitative synthesis, with a total of 1002 subjects. Were identified: angle of trunk rotation (ATR), angle between body segments (in development), movement - oscillations per second (Hz/s), impedance (Ohms/mm), synchronous postural observation, and aesthetic deformity. Only two studies provided validity and reproducibility coefficients, however, it was not possible to conduct a meta-analysis because the presented coefficients were different. INTERPRETATION:The angle of trunk rotation, angle between body segments, movement - oscillations per second (Hz/s), impedance (Ohms/mm), synchronous postural observation, and aesthetic deformity were the clinical postural variables evaluated through telehealth. Further research is needed to focus on the development, validation, and reproducibility of assessment instruments for patients with scoliosis through telehealth.

Clinicians monitor scoliosis progression using multiple radiographs during growth. During imaging, arms must be elevated to visualize vertebrae, possibly affecting sagittal alignment. This study aimed to determine the arm position that best represents habitual standing (and possibly allowing hand-based skeletal maturity assessment) to obtain frontal and lateral stereo-radiographs as measured using frontal, sagittal, and transverse angles. Females with and without, and males with Adolescent Idiopathic Scoliosis (AIS) were recruited consecutively. Patients were scanned using 3D Ultrasound imaging (3DUS), in 10 arm positions; habitual standing, arms supported anteriorly at 60° flexion, fingers to clavicle, chin, zygomatic, and eyebrows, arms abducted 90°, hands on wall, on blocks, and unsupported. Axial vertebral rotation (AVR) differences, frontal, and sagittal curve angles were measured. Repeated measures ANOVAs with Sidak post-hoc tests compared positions. Ninety females with and without AIS with mean age, and height of 17 ± 4 years, and 162 ± 6 cm, and ten males with AIS of 16 ± 3 years, and 174 ± 11 cm, respectively, were included. Female AIS single-curve showed larger curves in standing in all positions excluding hands on blocks (p > 0.05). Sagittal parameters showed decreases in kyphosis in arms abducted 90° and increases in lordosis in fingers to cheeks/eyebrows (p > 0.05). AVR twist was not significantly affected by position. Male AIS showed comparable results to females, but no significant differences were detected. No position represented habitual standing for all groups. When arms are raised, decreases in curve angle were shown in single-curve patients, kyphosis decreased, and lordosis increased in all groups. Most accurate positioning for all parameters was in fingers to clavicle/chin position.

Study Design. Cross-sectional study. Objective. The objective of this study was to quantify whether interapical distance and coronal balance measured on 3D ultrasound (3DUS) images differed among ten standing positions of participants with and without adolescent idiopathic scoliosis (AIS). Summary of Background Data. Apical vertebral translation (AVT) is an indicator of clinical symptoms and treatment outcomes in AIS. Stereo-radiography simultaneously captures a frontal and lateral image, yet patients are required to elevate their arms during the examination and varied positioning may impact coronal plane measurements. Ultrasound can assess these measurements without exposing participants to radiation. Methods. Females with and without AIS were recruited from a scoliosis clinic, and email advertisements, respectively. Participants underwent 3DUS scans in ten positions: standing; arms anteriorly supported in 60° of shoulder flexion; fingers to clavicles, chin, zygomatic processes, and eyebrows; shoulders abducted at 90° with hands open and thumbs on shoulders; hands on anterior wall with and without blocks; and hands unsupported. AVT and coronal balance measurements were obtained using custom software. Positions and groups were compared using repeated measures ANOVAs with Sidak pairwise comparisons. Results. Fifty-nine females had a mean age, height, and weight of 17.5±4.9 years, 162.8±5.8 cm, and 56.2±10.6 kg, respectively. Seventeen single and 14 double curve participants were included with mean maximum curve angles of 26.4°±3.6° and 25.2°±3.1°, respectively. Overall, there were no statistically significant differences in interapical distance or coronal balance among the ten positions, as well as in coronal balance between the groups ( P >0.05). The interapical distance of participants without AIS significantly differed from those with single curves in eight positions and from those with double curves in all ten positions. Conclusions. The positions evaluated may be interchangeable for coronal plane lateral deviation measurements of females with AIS.

This systematic review aimed to identify predictors of brace treatment outcomes for adolescents or adults with idiopathic scoliosis. Four databases including MEDLINE, EMBASE, Web of Science, and CINAHL were searched. Free text and indexed terms identifying the populations, predictions analyses and key outcomes were combined to search the literature. Pairs from eight independent reviewers conducted abstracts and full-text screening, and data extraction. The Quality in Prognostic Studies (QUIPS) tool was used to assess the risk of bias (ROB). Strength of evidence summary statements were formulated based on the risk of bias and the consistency of the research findings. The search found 2224 articles. After screening, seven articles were included. Only one article showed low ROB, while the others showed moderate ROB. All articles reported on patient-related outcome measurements (PROMS) of quality of life (QOL). Only one parameter achieved limited strength of evidence; shorter treatment time from one study predicted better long-term SRS-22 total scores. For other predictors, the level of evidence was unclear. Other predictors of long-term outcomes (> 1 year follow-up), from only 1 study on each outcome, were: larger Cobb angle predicted worse Spinal Appearance Questionnaire (SAQ) chest scores and worse depression; higher age predicted better SAQ curve scores, larger apical translations predicted worse SAQ shoulders and chest scores; a passive introverted personality or an active outgoing (MPI) character type predicted worse SRS-22 satisfaction; higher BMI predicted better SAQ curve, Rolland-Morris questionnaire (RMQ) lumbosacral pain, Quebec Back Pain Disability Scale (QDS) moving scores, and worse SRS-22 total; larger vital capacity predicted better QDS score; longer bracing (total) predicted worse depression; negative parental attitudes predicted worse depression; higher Strengths and Difficulties Questionnaire emotional symptoms, peer problems, prosocial behavior, and total scores predicted worse depression. Poor compliance from one short-term follow-up study predicted worse change of brace questionnaire (BRQ) for health perception, pain, physical and emotional functioning, and total scores. Moderate evidence from two studies with low and moderate RoB showed that age and Cobb angle did not predict long-term total SRS-22 score for prediction. Eleven parameters predicted bracing outcomes, but most studies presented moderate risk of bias. Only one parameter, longer treatment time, with limited strength of evidence was predictive of better long-term SRS-22 total scores. Since most findings still present an unclear level of evidence, common weaknesses were identified to encourage design of high-quality studies predicting bracing outcomes.

Paraspinal muscle imbalances in teenagers with adolescent idiopathic scoliosis (AIS) are well-documented, but their metabolic characteristics remain unclear. This study assessed oxygen recovery asymmetry (TrAsy) in paraspinal muscles during isometric trunk extension in teenagers with and without AIS and explored its correlation with spinal curvature. Fifty-one AIS participants with primary right thoracic curves (40 females; 13.5 ± 1.7 years; thoracic Cobb angles: 22.9°±6.8°; 28 mild [Cobb angles 10°–24°], 23 moderate-to-severe [≥ 25°]) and 51 non-AIS controls (33 females; 13.2 ± 1.7 years) performed prone isometric trunk extensions. Bilateral paraspinal oxygen recovery times (Tr) at T9 and L3 were measured using near-infrared spectroscopy. TrAsy (between-side Tr difference) was analyzed using mixed-design ANOVA with group (AIS vs. controls) and side (convex vs. concave) factors; Welch’s ANOVAs assessed TrAsy differences by curve severity and location. The association between thoracic Cobb angles and TrAsy was evaluated using multiple linear regression, adjusting for covariates. AIS cases (Cobb angles: 22.9°±6.8°) showed significantly longer Tr on the concave side at T9 (74.6s ± 10.1s vs. 50.1s ± 16.8s; p < 0.001), with no significant between-side difference at L3 in both groups analyzed collectively (p = 0.31). Tr was longer at T9 than L3 (p < 0.001) Single curves exhibited greater TrAsy than double curves at T9 (mean difference: 25.0s ± 8.60s; p = 0.004), with moderate-severe AIS showing larger concave-side Tr. The regression model showed that greater TrAsy was associated with larger thoracic Cobb angles (R2 = 0.46, p < 0.001). Teenagers with AIS showed greater TrAsy at T9, predominantly on the concave side, which correlated with curve severity. Further research should investigate causal relation between metabolic response and curvature progression.

OBJECTIVE:We aimed to establish the intra- and inter-evaluator reliability of apical vertebral translation (AVT), inter-apical distance and coronal balance measurements from three-dimensional ultrasound images (3DUS). METHODS:Female volunteers with and without adolescent idiopathic scoliosis underwent 3DUS scans in 10 positions: standing; arms anteriorly supported at 60° of shoulder flexion; fingers to clavicles, chin, zygomatic process and eyebrows; shoulders abducted 90° with hands open and thumbs on shoulders; hands on anterior wall with and without blocks; and hands unsupported. Custom software was utilized to obtain measurements. Intra-evaluator reliability ICC2,1 with standard error of measurement (SEM) were obtained for measurements extracted twice by a blinded evaluator (1 wk apart) on scans with pre-marked laminae. Inter-evaluator reliability compared the first measurement occasion by three evaluators. Intra-evaluator reliability was also tested for standing, chin and abducted positions, by an evaluator first measuring pre-marked laminae, comparing (1+ y later) to landmarking laminae and repeating measurements. RESULTS:Forty-four females had a mean age, height and weight of 16.8 ± 4.4 y, 163.1 ± 5.9 cm and 56.1 ± 10.8 kg, respectively. Fourteen single and 13 double curve participants had mean maximum curve angles of 29.2° ± 4.4 and 25.5° ± 3.4, in standing, respectively. AVT, inter-apical distance and coronal balance measurements satisfied the criteria for individual use (ICC > 0.90) for both intra-evaluator scenarios (SEM below 0.89 mm for pre-marked laminae or <4.97 mm when re-marking laminae) and for the inter-evaluator reliability (SEM < 1.12 mm). CONCLUSION:Reliable coronal plane vertebra displacement measurements can be obtained from 3DUS images in multiple standing positions.

ABSTRACT The aim of this study was to present the development and assessment of the measurement accuracy of variables obtained with the Digital Imaged-based Postural Assessment (DIPA©) Capture app and Analysis software as part of a telehealth assessment protocol for idiopathic scoliosis. To enable the correct capture of photographic images for further analysis, the development of an application, called DIPA© Capture, was performed. A photo is automatically obtained only when the smartphone is aligned. The sample of this prospective study was composed of consecutive images of a real plumb line (RPL) taken by using the DIPA© Capture app. Once the photo was captured the app automatically drew a virtual plumb line (VPL). The inclination of the RPL and VPL was measured using the DIPA-S© eHealth Analysis software. A total of 50 images using the App comprised this sample. The median (min-max) inclination angle of the real and virtual plumb lines was 89.8° (88.6°-90°) with a 1.4° range and 90° (89.4°-90°) with a 0.6° range, respectively. The mean difference between the inclination of the two plumb lines (RPL - VPL) was very small -0.1°±0.04° (p= 0.017). The RMS error was 0.3°. The DIPA© Capture app and Analysis software for image acquisition and measurement was developed and is ready for testing with patients. The app accurately captures the alignment of the smartphone during the image acquisition and the software shows adequate measurement accuracy for future assessment of idiopathic scoliosis by photogrammetry. Level of Evidence III; Diagnostic Studies - Investigation of a Diagnostic Test.

The impact of exercise-specific face masks (ESFMs) in aerobically fit individuals on physiological, perceptual, respiratory, and performance responses remains unclear. How ESFMs mitigate exercise-induced bronchoconstriction (EIB) is also unknown. Thus, this study aimed to determine how an ESFM altered within-exercise physiological, perceptual, respiratory, and performance responses to graded treadmill exercise. Twenty-four individuals (11 females) completed a discontinuous graded exercise test on a treadmill under two conditions (ESFM and unmasked). Physiological, respiratory function, and perceptual measures were assessed. Performance was determined by time to exhaustion. Statistical analyses included linear mixed-effects modeling, repeated measures analysis of variance, and pairwise comparisons using an alpha value of 0.05. ESFM use significantly impaired performance (median = −150.5 s) and decreased arterial oxygen saturation at maximal intensity (mean = −3.7%). Perceptions of air hunger and work of breathing were elevated across submaximal and maximal intensities. Perceived exertion and breathing discomfort were significantly elevated submaximally but not maximally. Spirometry measures were not significantly different at termination but were significantly improved at submaximal intensities in participants with and without EIB. ESFM use in fit individuals increased perceptual discomfort, impaired performance, and augmented arterial desaturation. Respiratory function improvements were observed but were accompanied by adverse perceptual sensations. Despite this, performance impairments may limit the real-world utility of ESFMs for athletes.

Measuring the kyphotic angle (KA) and lordotic angle (LA) on lateral radiographs is important to truly diagnose children with adolescent idiopathic scoliosis. However, it is a time-consuming process to measure the KA because the endplate of the upper thoracic vertebra is normally difficult to identify. To save time and improve measurement accuracy, a machine learning algorithm was developed to automatically extract the KA and LA. The accuracy and reliability of the T1-T12 KA, T5-T12 KA, and L1-L5 LA were reported. A convolutional neural network was trained using 100 radiographs with data augmentation to segment the T1-L5 vertebrae. Sixty radiographs were used to test the method. Accuracy and reliability were reported using the percentage of measurements within clinical acceptance (≤9°), standard error of measurement (SEM), and inter-method intraclass correlation coefficient (ICC2,1). The automatic method detected 95 % (57/60), 100 %, and 100 % for T1-T12 KA, T5-T12 KA, and L1-L5 LA, respectively. The clinical acceptance rate, SEM, and ICC2,1 for T1-T12 KA, T5-T12 KA, and L1-L5 LA were (98 %, 0.80°, 0.91), (75 %, 4.08°, 0.60), and (97 %, 1.38°, 0.88), respectively. The automatic method measured quickly with an average of 4 ± 2 s per radiograph and illustrated how measurements were made on the image, allowing verifications by clinicians.

We read an interesting sectorial literature analysis conducted by Coraci et al [1] to summarize the characteristics of outcome measures reported in peer-reviewed articles related to rehabilitation and AIS over the last decade. Out of 127 relevant papers identified from PubMed, the most frequently used outcome measure was Cobb angles from radiographs. Other commonly assessed outcomes included health-related quality of life, surface volume, respiration, muscular function, and electrophysiological assessments.

Clinicians monitor scoliosis progression using radiographs during growth. Three standing positions were assessed because the arms must be elevated to visualize all vertebrae on radiographs, possibly affecting sagittal parameters. 3D Ultrasound (3DUS) is safe to assess positions without radiation, but its test–retest reliability has not been established for common radiograph positions. The aim was to determine the test–retest reliability of frontal, sagittal, and transverse measurements obtained from three positions using 3DUS imaging. Participants with AIS were recruited from a clinic in Alberta. Participants underwent 3DUS scans in habitual standing, fingers to clavicle, and hands on wall. Participants were re-scanned in the same positions 20-min following first scans. Custom software obtained measurements. Test–retest reliability (ICC3,1) with standard error of measurement (SEM) was assessed by one evaluator who was blinded to the test measurement when completing retest. Forty-three participants with AIS had an age, height, and weight of 15 ± 2 years, 164 ± 10 cm, and 54 ± 11 kg, respectively. Standing curve angle was 23 ± 11°. Habitual standing max curve angle, T4/T5 kyphosis, and AVR twist satisfied criteria for individual use (ICC2,1 > 0.90). Fingers to clavicle AVR twist satisfied criteria for individual use, and all hands on wall measurements satisfied criteria for individual use (ICC > 0.90). All other parameters satisfied criteria for research use (ICC > 0.70). Range of SEM for curve angle, whole kyphosis, T4/T5 kyphosis, lordosis, and AVR twist were 3.0–4.8°, 4.0–5.1°, 3.6–5.3°, 3.9–5.6°, and 1.6–2.2°, respectively. 3DUS produces reliable frontal, sagittal, and transverse measurements for research use in three standing positions. Hands on wall produces reliable measurements for clinical practice.

OBJECTIVE: To summarize the development and validation of machine learning (ML) models using data from the first clinical visit to predict ensuing curve progression in individuals with adolescent idiopathic scoliosis. DESIGN: Systematic review. LITERATURE SEARCH: Five databases were searched from inception to March 31, 2023. STUDY SELECTION CRITERIA: Studies had to include teenagers (10-17 years) with adolescent idiopathic scoliosis and internally/externally validated ML models for predicting curve progression. DATA SYNTHESIS: We used the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) to guide data extraction, the International Journal of Medical Informatics (IJMEDI) checklist for quality assessments, and the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system for evidence certainty. RESULTS: Seventeen studies with 37 ML models (3701 participants) were included. The models predicted risk, threshold progression, Cobb angles, and curve variations. The IJMEDI checklist identified critical shortcomings in data understanding (60%), preparation (82%), validation (95%), and deployment (95%). Thirty-four ML models showed fair to excellent accuracy, with the area under the receiver operating characteristic curve ranging from 0.70 to 0.93 in most categories. CONCLUSION: While most models achieved moderate to good accuracy, the certainty of evidence was very low, rendering most models not clinically deployable and lacking external validation. Therefore, these findings should be interpreted with caution. Future research should address these limitations to improve the clinical applicability and external validity. JOSPT Open 2024;2(3):202-224. Epub 17 April 2024. doi:10.2519/josptopen.2024.0718

Understanding idiopathic scoliosis (IS) natural history during growth is essential for shared decision-making between patients and physicians. We developed a retrospective model with the largest available sample in the literature and we aimed to investigate if using three peri-pubertal growth periods provides better prediction than a unique model. Secondary analysis of a previous study on IS natural history data from radiographs before and at the first consult. Three groups: BEFORE (age 6–10), AT (age 11-Risser 2) and AFTER (from Risser 3) the pubertal growth spurt. Available predictors: Cobb angle, curve type, sex, observation time, and Risser score. We used linear mixed-effects models to predict future Cobb angles in each group. We internally validated prediction accuracy with over 100 patients per group (3 to 5-fold cross-validation). We included 1563 participants (275 BEFORE, 316 AFTER, 782 females and 190 males AT). Curves increased over time mostly in AT, importantly in BEFORE, but also in AFTER. All models performed better than the general one. In BEFORE, 74.2

Spinal alignment can be assessed without radiation using three-dimensional ultrasound imaging (3DUS). Reliable measurements could inform the ideal arm position for scoliosis radiographs. This study determined the inter-evaluator reliability of axial vertebral rotation (AVR) measurements and sagittal curve angles in healthy females from 3DUS spinal images in standing with two common radiograph arm positions: Chin and abduction (Abd). Three novice evaluators (R1, R2, and R3) measured images once. The Bland-Altman bias and limits of agreement (LOA) were calculated. T5-T12 kyphosis measurements met our bias and LOA error threshold only for R1 vs R2 in the Chin position [-0.53 degrees (-10.7,9.7)]. The distance between the bias and the LOA also met our acceptable threshold of 5 degrees for all mean AVRTwist, and all but one AVRTwist measurements (R1-R2_Chin = 5.9 degrees). Three lordosis measurements did not meet our acceptable threshold (R1-R2_Chin, R1-R3_Chin, and R2-R3_Abd). The inter-evaluator reliability of novice evaluators was adequate and did not differ significantly between positions.

Ultrasonography for scoliosis is a novel imaging method that does not expose children with adolescent idiopathic scoliosis (AIS) to radiation. A single ultrasound scan provides 3D spinal views directly. However, measuring ultrasonograph parameters is challenging, time-consuming, and requires considerable training. This study aimed to validate a machine learning method to measure the coronal curve angle on ultrasonographs automatically. A total of 144 3D spinal ultrasonographs were extracted to train and validate a machine learning model. Among the 144 images, 70 were used for training, and 74 consisted of 144 curves for testing. Automatic coronal curve angle measurements were validated by comparing them with manual measurements performed by an experienced rater. The inter-method intraclass correlation coefficient (ICC2,1), standard error of measurement (SEM), and percentage of measurements within clinical acceptance (≤ 5°) were analyzed. The automatic method detected 125/144 manually measured curves. The averages of the 125 manual and automatic coronal curve angle measurements were 22.4 ± 8.0° and 22.9 ± 8.7°, respectively. Good reliability was achieved with ICC2,1 = 0.81 and SEM = 1.4°. A total of 75