Abstract
Aims
This systematic review aims to identify 3D predictors derived from biplanar reconstruction, and to describe current methods for improving curve prediction in patients with mild adolescent idiopathic scoliosis.
Methods
A comprehensive search was conducted by three independent investigators on MEDLINE, PubMed, Web of Science, and Cochrane Library. Search terms included “adolescent idiopathic scoliosis”,“3D”, and “progression”. The inclusion and exclusion criteria were carefully defined to include clinical studies. Risk of bias was assessed with the Quality in Prognostic Studies tool (QUIPS) and Appraisal tool for Cross-Sectional Studies (AXIS), and level of evidence for each predictor was rated with the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach. In all, 915 publications were identified, with 377 articles subjected to full-text screening; overall, 31 articles were included.
Results
Torsion index (TI) and apical vertebral rotation (AVR) were identified as accurate predictors of curve progression in early visits. Initial TI > 3.7° and AVR > 5.8° were predictive of curve progression. Thoracic hypokyphosis was inconsistently observed in progressive curves with weak evidence. While sagittal wedging was observed in mild curves, there is insufficient evidence for its correlation with curve progression. In curves with initial Cobb angle < 25°, Cobb angle was a poor predictor for future curve progression. Prediction accuracy was improved by incorporating serial reconstructions in stepwise layers. However, a lack of post-hoc analysis was identified in studies involving geometrical models.
Conclusion
For patients with mild curves, TI and AVR were identified as predictors of curve progression, with TI > 3.7° and AVR > 5.8° found to be important thresholds. Cobb angle acts as a poor predictor in mild curves, and more investigations are required to assess thoracic kyphosis and wedging as predictors. Cumulative reconstruction of radiographs improves prediction accuracy. Comprehensive analysis between progressive and non-progressive curves is recommended to extract meaningful thresholds for clinical prognostication.
Cite this article: Bone Jt Open 2024;5(3):243–251.
Take home message
The torsion index and apical vertebral rotation are good 3D predictors of curve progression.
3D Cobb angle, thoracic kyphosis, and sagittal wedging are weaker predictors that require further investigation.
Serial spinal reconstructions and inclusion of growth extrapolation are needed to provide better predictive model accuracy.
Introduction
Adolescent idiopathic scoliosis (AIS) is a complex condition that requires regular follow-up monitoring and casts significant psychosocial pressure on its patients.1-6 Prediction of curve progression can reduce unnecessary consultations and bracing in non-progressive patients, while allowing earlier intervention and proper prognostication to progressive patients.7,8
As a 3D deformity, AIS is characterized by the lateral spinal curvature in the frontal plane, a disturbance of physiological spinal curvatures in the sagittal plane, and an axial rotation of the vertebrae in the transverse plane.9-14 Despite hypokyphosis and axial rotation being recognized as important factors in curve development, patients undergoing conservative management are usually only assessed using 2D Cobb angle and bone age for prediction of curve progression. In the recent literature, more specialized centres have been characterizing spinal deformity in axial and sagittal planes to improve accuracy in predicting curve progression.15-21
To assess rotation in larger curves, the Nash-Moe22 method has been extensively used, but it is limited by low accuracy and replicability.23,24 Meanwhile, 3D reconstruction from CT scans are not routinely performed due to exposure to ionizing radiation.25,26 In recent years, 3D reconstruction of biplanar radiographs has been increasingly validated for its accuracy and reproducibility.27,28 It should be noted that 3D in the context of biplanar reconstruction also refers to the ability to derotate vertebral segments to obtain segmental kyphosis, wedging, and intervertebral rotation in the patient plane.29-31 While providing extensive quantitative data, commercially available programs for biplanar reconstruction still require considerable manual effort in mapping spinal landmarks prior to the automated measurement sequence.32 In recent years, we have also seen a rise in transdisciplinary studies using machine learning on clinical data to develop in-house programs for predicting curve progression,7,16,33-36 which involves specialized terminology that may be challenging to digest.
To extract useful clinical points from the diverse range of existing studies, this systematic review aims to identify and investigate 3D parameters derived from biplanar reconstruction as predictors of curve progression. The focus is on nonoperative AIS patients, especially to stratify risk of progression at early visits. In addition, the review aims to summarize current techniques to improve predictive accuracy using machine learning.
Methods
Search strategy and selection criteria
The literature search and reporting of study results were conducted according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement.37 Three independent investigators (HTSW, DLLW, CHST) performed an extensive search on the following databases: PubMed, Web of Science, MEDLINE, and Cochrane Library. All fields were searched in the databases using the following keywords: (adolescent idiopathic scoliosis) AND ((biplanar) OR (3D) OR (three-dimensional)) AND ((progression) OR (alignment) OR (prognosis)). The search was limited to publications from 1 January 1996 to 31 December 2023 to exclude obsolete techniques in generating 3D spinal models. The full search strategy can be seen in Supplementary Table i. Potentially relevant abstracts were screened based on the inclusion and exclusion criteria (Table I), and full-text articles were obtained for eligible results. The references of each included article were screened for any other pertinent articles (see Supplementary Table ii). Any discrepancies in the final decision of inclusion were settled through discussion with all authors.
Table I.
Inclusion and exclusion criteria.
| Inclusion criteria | Exclusion criteria |
|---|---|
|
|
Data extraction and critical appraisal
The primary outcome of this systematic review was the efficacy of 3D parameters derived from biplanar radiographs as predictors of curve progression in nonoperative AIS, which was reported using statistical measures including sensitivity and specificity, positive predictive value (PPV), negative predictive value (NPV), area under the curve (AUC), root mean square error (RMSE), and R-squared (r2). The secondary outcome was to summarize methods to improve prediction analysis involving geometrical models.
The 3D parameters derived from biplanar radiographs include Cobb angle, coronal tilt, thoracic kyphosis (TK), lumbar lordosis (LL), apical vertebral rotation (AVR), intervertebral rotation at the upper junctional zone (upper IAR), intervertebral rotation at the lower junctional zone (lower IAR), angle of the plane of the maximum curvature (POMC), torsion index (TI), hypokyphosis index, and vertebral and/or disc wedging in the frontal and sagittal planes. Detailed descriptions of the included parameters are shown in Supplementary Tables iv to vi.
Other information regarding the study design, sample size, patient population, predictors identified, risk of bias, and level of evidence can also be viewed in the Supplementary Material.
Risk of bias
Three independent reviewers (HTSW, DLLW, CHST) assessed the risk of bias for the included longitudinal studies using the six domains of the Quality in Prognostic Studies (QUIPS).38 For retrospective studies, bias due to attrition is not applicable and therefore not assessed. The QUIPS risk of bias for these studies is detailed in the Supplementary Table iii. Cross-sectional studies were assessed using the Appraisal tool for Cross-Sectional Studies (AXIS).39 Due to lack of a scoring system, overall results will be described using mean number of items achieved and any notable underperformance in particular items will be reported. Any discrepancy was discussed with all authors until a consensus was reached.
Grading of evidence
The three reviewers assessed the quality of evidence of the outcomes according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.40 All included studies in this review were observational studies and were thus initially assigned as having a low level of evidence, according to GRADE guidelines.40 Downgrading of quality of evidence was done according to the five domains in the GRADE guidelines: risk of bias,41 imprecision,41,42 indirectness,43 inconsistency,41 and publication bias.44 Meanwhile, the quality of evidence was upgraded based on large magnitude of effect, dose-response gradient, and plausible confounding that can increase confidence in estimated effects.45 The overall quality of evidence is detailed in the Supplementary Tables vii to viii.
Search results
The search results are illustrated in the PRISMA flowchart (Figure 1). A total of 915 articles were yielded from the initial search, of which 183 articles were from MEDLINE, 299 articles from Web of Science, 37 articles from Cochrane library, and 396 articles from PubMed. Of the 915 articles, there were 538 duplicated articles, and 377 unique articles were screened for the inclusion and exclusion criteria. As a result, a total of 31 articles were included in the final study for further analysis.
Fig. 1
PRISMA (preferred reporting items for systematic reviews and meta-analyses) flowchart illustrating selection process of articles.
Among the 31 articles included, 16 were cross-sectional studies, 13 were retrospective cohort studies, one was a prospective cohort study, and another a prospective case-control study. Sample sizes ranged from ten to 321 AIS subjects. Overall, nine studies included patients with mild curves (< 20°) exclusively, while the remaining studies included moderate (20° to 40°) or severe curves (> 40°). The mean age of subjects across studies was 13.3 years (10 to 18) and the length of follow-up ranged from three months to eight years. Progression was defined by most studies as interval increase in Cobb angle > 5°, or the initiation of brace treatment as determined by an orthopaedic specialist. The overall risk of bias for included studies was low. For cross-sectional studies, the mean number of AXIS items accomplished was 19.2 (standard deviation (SD) 1.0; 17 to 20), with cohort base being the most frequently violated item, followed by eligibility criteria.
Results
Coronal plane
For patients with mild curves, there is moderate evidence from five studies supporting 3D Cobb angle as a weak predictor for differentiating the risk of curve progression. Wang et al19 and Nault et al46 reported no significant difference in initial 3D Cobb angle comparing progressive and non-progressive groups. The two studies each had considerable sample sizes compared to 3D reconstruction studies in the literature, with 490 subjects and 172 subjects, respectively. Vergari et al16,17 also found initial 3D Cobb angle to be similar (mean difference < 3°) between progressive and non-progressive groups, but did not present results of statistical comparison. Aside from 3D Cobb angle, Nault et al47 also reported coronal apical disc wedging as a statistically significant predictor of final Cobb angle (β = 0.820; p = 0.016). However, the effect is small, and prior study by the same group of authors found no difference in initial coronal apical disc wedging between progressive and non-progressive groups.46 Almansour et al48 used coronal tilt to characterize segmental lateral displacement in the frontal plane. However, changes in coronal tilt after bracing largely reflected reduction in Cobb angle and provided little additional information.
Sagittal plane
In our included studies, 3D thoracic kyphosis (3D TK) and sagittal wedging were the most frequently reported sagittal parameters. There is weak evidence supporting 3D TK as a predictor of curve progression from three studies.18,46,49 In a study of 172 patients, Nault et al18 reported weak correlation between 3D T4-T12 TK at first visit and final 3D Cobb angle (r = −0.288, p = 0.01). Another study by the same group of authors46 reported that patients with progressive curves had lower initial 3D T4-T12 TK compared to the non-progressive group (mean 20.6° vs 25.0°; p = 0.02). Conversely, Wang et al19 reported no significant difference in initial 3D T4-T12 TK between two groups (22.3 (SD 8.5) vs 21.4 (SD 9.2), p = 0.635).
Sagittal vertebral wedging was reported in five studies.49-53 Begon et al50 reported that sagittal vertebral wedging is present in mild curves. Scherrer et al51 reported that vertebral wedging at thoracic apices was associated with increase in Cobb angle. However, statistical evidence supporting its use in predicting curve progression is lacking. A ‘hypokyphosis index’ was mentioned in two studies,49,53 which was a function of wedging of the apical vertebra compared to normal controls. While its replicability is low, the hypokyphosis index was reported to increase predictive accuracy for curve progression. Lastly, none of our included studies reported spinopelvic parameters as curve predictors of significance.
Axial plane
There is moderate evidence from five studies supporting axial rotation and/or TI as good predictors of curve progression (Table II). Among our included studies, AVR, intervertebral axial rotation (IAR), and TI were the most reported parameters representing axial rotation. In sterEOS 3D (EOS Imaging, France), axial rotation is calculated for each vertebra after adjusting for pelvic rotation.49,54-58 IAR was typically reported for both upper and lower junctions.59,60 The TI was generally defined as the mean of the two sums of IAR from the lower junction to the apex, and from the apex to the upper junction, as described by Steib et al.18,46,61,62 Several studies used geometrical modelling to obtain torsion,16,49,63,64 and Skalli et al49 incorporated torsion into the ‘severity index’ for predicting progression.
Table II.
Comparison of axial parameters between progressive and non-progressive groups.
| Mean Torsion index, ° (SD) | Mean apical vertebral rotation, ° (SD) | |||||
|---|---|---|---|---|---|---|
| Author | Progressive | Non-progressive | p-value | Progressive | Non-progressive | p-value |
| Courvoisier et al65 | 7 (2) | 3 (1) | < 0.001† | 9 (3) | 4 (2) | < 0.001† |
| Wang et al19 | 6 (3) | 3 (2) | 0.020‡ | 7 (5) | 4 (3) | 0.006‡ |
| Nault et al46 | 4.5* | 3.1* | 0.02‡ | 8.1* | 5.7* | 0.006‡ |
| Vergari et al16 | 4.1 (2.1) | 5.6 (2.8) | N/A | 7.6 (4.1) | 6.1 (3.6) | N/A |
| Skalli et al49 | 6 (3 | 4 ± 2 | < 0.05‡ | 7 (4) | 6 (4) | N/A |
-
Different authors reported the means and standard deviations in different decimal places.
-
*
Nault et al did not report SDs.
-
†
< 0.001.
-
‡
< 0.05.
-
N/A, not available; SD, standard deviation.
Among the transverse parameters, Courvoisier et al65 reported that TI had the highest predictive value (Figure 2) for progression compared to AVR and IAR. When a TI of 3.7 was used as a cut-off, prediction of progression had a sensitivity and specificity of 81% (AUC 0.85 (0.77 to 0.94)). Wang et al19 analyzed radiographs of patients at the first visit, and reported that while the progressive group had similar Cobb angle, TK, and LL with the non-progressive group, the progressive group had higher AVR and torsion.
Fig. 2
Summary of important outcomes from the current review. #Torsion index refers to the mean intervertebral rotation within the scoliotic segment. *Hypokyphosis was quantified by 3D T5-T12 thoracic kyphosis as well as sagittal wedging of the vertebra and intervertebral disc, respectively. ^Cumulative reconstruction refers to including all past reconstructions and intermediate output layers in every input layer of the predictive model, as opposed to sequential layering, in which only the most recent spinal reconstruction is included in the input layer.
Several cross-sectional studies that were included also analyzed differences between Lenke curve types. Despite the lack of serial data, the differences in 3D spinal deformity between curve types may offer insights into the pathomechanism of curve progression. Karam et al53 reported that thoracic curves, which typically have larger Cobb angles, had the highest TI, while TL curves had the highest AVR. Conversely, Courvoisier et al65 found that TI was comparable for T and L curves.
Prediction analysis
Inclusion of cumulative reconstructions was reported to improve prediction accuracy, when compared to only using the spinal reconstruction from the most recent visit (i.e. sequential layering). García-Cano et al66 reported that the average root mean square error (RMSE) of the spinal model was improved from 10.36 mm to 8.78 mm after considering all reconstructions from prior visits. Regarding the type of prediction model, García-Cano et al66 reported using a random forest model to directly predict spinal morphology, while Kadoury et al67 used probabilistic classification model to identify progressive curves, in which the model was reported to be superior to using a support vector machine model.
Five studies used geometrical spinal models instead of conventional 3D parameters, thus the predictors were presented as clusters of 3D morphology,66 or composite ‘black-box’ models, in which relative importance of spinal parameters was not specified.16,49,67,68 Despite the lack of detailed analysis, these studies on a whole demonstrated accurate predictions, with sensitivity ranging from 88% to 92% and specificity ranging from 74% to 84%.
Discussion
3D reconstruction of biplanar radiographs allows comprehensive evaluation of the scoliotic spine and offers robust data for accurately predicting curve prediction. In this review, we have collected and summarized the key predictors of curve progression in mild curves. TI and AVR, both transverse parameters, were identified as accurate predictors with moderate evidence, while 3D TK and sagittal wedging was identified as predictors with weak evidence (Figure 2). In patients with mild curves, 3D Cobb angle was found to be a weak predictor with moderate evidence. In terms of predictive modelling, using cumulative layering of past reconstructions increases predictive accuracy.
Coronal curvature
It is well established that a larger Cobb angle predisposes to curve progression.1,7,18 A systematic review by Wong et al1 found that initial 2D Cobb angle > 25° and thoracic curves were predictive of curve progression. However, most of our included longitudinal studies involves the first radiograph at early visits, when patients are skeletally immature and have mild curves. Nault et al46 and Wang et al19 reported no significant differences when comparing initial 3D Cobb angle between progressive and non-progressive groups. The evidence supports the theory that coronal curvature is not the initial trigger of curve progression, which will be elaborated in the following sections. Nevertheless, as Cobb angle acts as a poor predictor at mild stages, the coronal curvature eventually evolves along with wedging and rotation in other planes, which adds predictive value. To evaluate the predictive power of increasing Cobb angle on growing patients, Parent et al7 compared the accuracy of predictive models based on 2D Cobb angle assessed at different consultations. Relative skeletal immaturity with larger initial 2D Cobb angle and longer duration of observation were associated with curve progression, though the cohort had a larger range of baseline Cobb angle (20° (SD 10°)). The longitudinal study design and clinically oriented reporting present as a strong framework for potential studies using 3D spinal parameters, and can pave the way for accurate predictions at first visits.
Sagittal deformity
While there is no universally accepted theory explaining the pathogenesis of AIS, anterior column overgrowth is a frequently studied phenomenon,69-75 with ongoing debate surrounding the cause-effect relationship with wedging and rotation.76,77 Our included studies also reported sagittal wedging in lower junctional vertebrae and discs, which was associated with increases in Cobb angle.49,51 Among thoracic, thoracolumbar, and lumbar curves, thoracic hypokyphosis was consistently observed in 3D studies.53,78-80 Using CT reconstruction, Schlösser et al81 reported that anterior overgrowth was observed in primary and compensatory curves, but not at junctional segments. In contrast, 2D studies have either identified hypokyphosis in thoracic curves only,82,83 or found that TK in AIS patients was equivocal to controls.84
Although further investigations are needed, the current evidence from 3D studies supports thoracic hypokyphosis and sagittal wedging as potential predictors present even in mild curves.18,49,72 This may be explained by the shift of plane of maximum curvature towards the frontal plane in hypokyphotic patients. The asymmetrical loading on the vertebral bodies due to gravity-induced torque may result in frontal wedging and progressive intervertebral rotation.75,85,86 While accurate estimation of TK in the patient plane remains challenging on plain radiography,87,88 algorithms to predict 3D TK based on 2D Cobb and 2D TK present as promising, accessible tools for clinical practice.89
Rotation and torsion
Our included studies indicated that axial rotation and TI improved overall accuracy of predicting curve progression. TI > 3.4° was strongly predictive of progression, which was likely because TI is a summative parameter taking in account of all intervertebral rotations across the scoliotic segment, and thus also captures curve types by location. Interestingly, Wang et al19 reported increase in AVR ahead of TK increase, which remains to be further investigated as the cohort only included 162 subjects. Among curve types, thoracic curves had the highest TI while thoracolumbar curves had the highest AVR, and both had significantly larger Cobb angle than lumbar curves.53 While a systematic review of 2D predictors by Wong et al1 has also found that thoracic curves and high AVR were associated with curve progression, characterizing rotation of the whole curve may improve prediction of curve progression.19,49
While conventional methods of estimating rotation in daily practice, such as Nash-Moe grading, are limited by low reproducibility,24 estimation of AVR using 2D Cobb angle and 2D TK,90 as well as fully automated measurement programmes,91 both exist as efficient solutions. Digital surface topography devices to quantify spinal and trunk rotation have also become more widely used in place of scoliometers, allowing whole-spine assessment.92-94
Prediction analysis
Machine-learning algorithms are capable of processing complex data and generating more accurate predictions compared to traditional regression models. Serial reconstructions arranged in stepwise layers were found to strongly improve prediction accuracy,7,66 as this allows better extrapolation of growth trajectories. While random forest model and probabilistic classification model were reported as useful prognostication models,66,67 more complex models, such artificial neural network models,19,34 have yet to be explored in 3D analysis. Regardless of reconstruction technique, most machine-learning programs can generate a vast amount of quantitative data. While dimensionality reduction tools aids in the extraction and refinement of statistically significant parameters, these often result in complex clusters involving combinations of coronal, sagittal, and axial deformities that are difficult to translate into clinical prognostication. Despite the ability to capture complex deformities, generalizable nomenclature is still necessary for meaningful interpretation. Comprehensive post-hoc analysis is also integral for generating cut-offs and analyzing interactions between spinal parameters, and thus should always be incorporated.
This is the first review to evaluate predictors of curve progression based on 3D reconstruction of biplanar radiographs. There were several limitations in this review. First, due to different methodologies in our included studies in terms of timeframe and reconstruction technique, a meta-analysis could not be performed. For studies using semi-automated methods for variable extraction, more comprehensive characterization of synthesized features would allow better interpretation. Second, publication bias could not be assessed, as most studies did not report effect sizes and confidence intervals. Third, no randomized controlled trials were identified during our search. Nevertheless, the predictors extracted from included studies were rigorously examined for quality of evidence.
In conclusion, TI and AVR were good predictors of curve progression, while more investigations are needed to validate 3D thoracic kyphosis and sagittal wedging as predictors; 3D Cobb angle was found to be a weak predictor. To improve predictive accuracy, machine-learning models based on serial spinal reconstructions can be used to capture the complex interactions between spinal parameters and extrapolate growth trajectories. Future research should include more comprehensive post-hoc analysis with comparison of relative importance among various parameters to facilitate interpretation. In daily practice, algorithms to predict 3D TK and AVR based on 2D parameters, as well as surface topography, can be applied to quickly assess curve morphology.
References
1. Wong LPK , Cheung PWH , Cheung JPY . Curve type, flexibility, correction, and rotation are predictors of curve progression in patients with adolescent idiopathic scoliosis undergoing conservative treatment: a systematic review . Bone Joint J . 2022 ; 104-B ( 4 ): 424 – 432 . Crossref PubMed Google Scholar
2. Zimoń M , Matusik E , Kapustka B , Durmała J , Doroniewicz I , Wnuk B . Conservative management strategies and stress level in children and adolescents with idiopathic scoliosis . Psychiatr Pol . 2018 ; 52 ( 2 ): 355 – 369 . Crossref PubMed Google Scholar
3. Wang H , Tetteroo D , Arts JJC , Markopoulos P , Ito K . Quality of life of adolescent idiopathic scoliosis patients under brace treatment: a brief communication of literature review . Qual Life Res . 2021 ; 30 ( 3 ): 703 – 711 . Crossref PubMed Google Scholar
4. Cheung PWH , Wong CKH , Cheung JPY . An insight Into the health-related quality of life of adolescent idiopathic scoliosis patients who are braced, observed, and previously braced . Spine (Phila Pa 1976) . 2019 ; 44 ( 10 ): E596 – E605 . Crossref PubMed Google Scholar
5. Wang H , Zhang T , Zhang C , et al. An intelligent composite model incorporating global / regional X-rays and clinical parameters to predict progressive adolescent idiopathic scoliosis curvatures and facilitate population screening . EBioMedicine . 2023 ; 95 : 104768 . Crossref PubMed Google Scholar
6. Birch NC , Tsirikos AI . Long-term follow-up of patients with idiopathic scoliosis: providing appropriate continuing care . Bone Joint J . 2023 ; 105-B ( 2 ): 99 – 100 . Crossref PubMed Google Scholar
7. Parent EC , Donzelli S , Yaskina M , et al. Prediction of future curve angle using prior radiographs in previously untreated idiopathic scoliosis: natural history from age 6 to after the end of growth (SOSORT 2022 award winner) . Eur Spine J . 2023 ; 32 ( 6 ): 2171 – 2184 . Crossref PubMed Google Scholar
8. Cheung JPY , Cheung PWH . Supine flexibility predicts curve progression for patients with adolescent idiopathic scoliosis undergoing underarm bracing . Bone Joint J . 2020 ; 102-B ( 2 ): 254 – 260 . Crossref PubMed Google Scholar
9. Konieczny MR , Senyurt H , Krauspe R . Epidemiology of adolescent idiopathic scoliosis . J Child Orthop . 2013 ; 7 ( 1 ): 3 – 9 . Crossref PubMed Google Scholar
10. Weinstein SL , Dolan LA , Cheng JCY , Danielsson A , Morcuende JA . Adolescent idiopathic scoliosis . Lancet . 2008 ; 371 ( 9623 ): 1527 – 1537 . Crossref PubMed Google Scholar
11. Post M , Verdun S , Roussouly P , Abelin-Genevois K . New sagittal classification of AIS: validation by 3D characterization . Eur Spine J . 2019 ; 28 ( 3 ): 551 – 558 . Crossref PubMed Google Scholar
12. Pasha S , Baldwin K . Are we simplifying balance evaluation in adolescent idiopathic scoliosis? Clin Biomech (Bristol, Avon) . 2018 ; 51 : 91 – 98 . Crossref PubMed Google Scholar
13. Kotwicki T . Evaluation of scoliosis today: examination, X-rays and beyond . Disabil Rehabil . 2008 ; 30 ( 10 ): 742 – 751 . Crossref PubMed Google Scholar
14. Wong DLL , Mong PT , Ng CY , et al. Can anterior vertebral body tethering provide superior range of motion outcomes compared to posterior spinal fusion in adolescent idiopathic scoliosis? a systematic review . Eur Spine J . 2023 ; 32 ( 9 ): 3058 – 3071 . Crossref PubMed Google Scholar
15. Amzallag-Bellenger E , Uyttenhove F , Nectoux E , et al. Idiopathic scoliosis in children and adolescents: assessment with a biplanar X-ray device . Insights Imaging . 2014 ; 5 ( 5 ): 571 – 583 . Crossref PubMed Google Scholar
16. Vergari C , Gajny L , Courtois I , et al. Quasi-automatic early detection of progressive idiopathic scoliosis from biplanar radiography: a preliminary validation . Eur Spine J . 2019 ; 28 ( 9 ): 1970 – 1976 . Crossref PubMed Google Scholar
17. Vergari C , Skalli W , Abelin-Genevois K , et al. Effect of curve location on the severity index for adolescent idiopathic scoliosis: a longitudinal cohort study . Eur Radiol . 2021 ; 31 ( 11 ): 8488 – 8497 . Crossref PubMed Google Scholar
18. Nault M-L , Beauséjour M , Roy-Beaudry M , et al. A predictive model of progression for adolescent idiopathic scoliosis based on 3D spine parameters at first visit . Spine (Phila Pa 1976) . 2020 ; 45 ( 9 ): 605 – 611 . Crossref PubMed Google Scholar
19. Wang H , Zhang T , Cheung KMC , Shea GKH . Application of deep learning upon spinal radiographs to predict progression in adolescent idiopathic scoliosis at first clinic visit . EClinicalMedicine . 2021 ; 42 : 101220 . Crossref PubMed Google Scholar
20. Wong LPK , Cheung PWH , Cheung JPY . Supine correction index as a predictor for brace outcome in adolescent idiopathic scoliosis . Bone Joint J . 2022 ; 104-B ( 4 ): 495 – 503 . Crossref PubMed Google Scholar
21. Wong LPK , Cheung PWH , Cheung JPY . Curve type, flexibility, correction, and rotation are predictors of curve progression in patients with adolescent idiopathic scoliosis undergoing conservative treatment: a systematic review . Bone Joint J . 2022 ; 104-B ( 4 ): 424 – 432 . Crossref PubMed Google Scholar
22. Nash CL , Moe JH . A study of vertebral rotation . J Bone Joint Surg Am . 1969 ; 51-A ( 2 ): 223 – 229 . PubMed Google Scholar
23. Marawar SV , Ordway NR , Auston DA , et al. Assessment of inter- and intraobserver reliability and accuracy to evaluate apical vertebral rotation using four methods: an experimental study using a saw bone model . Spine Deform . 2019 ; 7 ( 1 ): 11 – 17 . Crossref PubMed Google Scholar
24. Boyer L , Shen J , Parent S , Kadoury S , Aubin CE . Accuracy and precision of seven radiography-based measurement methods of vertebral axial rotation in adolescent idiopathic scoliosis . Spine Deform . 2018 ; 6 ( 4 ): 351 – 357 . Crossref PubMed Google Scholar
25. Hong JY , Suh SW , Easwar TR , Modi HN , Yang JH , Park JH . Evaluation of the three-dimensional deformities in scoliosis surgery with computed tomography: efficacy and relationship with clinical outcomes . Spine (Phila Pa 1976) . 2011 ; 36 ( 19 ): E1259 – 65 . Crossref PubMed Google Scholar
26. Pietton R , Bouloussa H , Langlais T , et al. Estimating pulmonary function after surgery for adolescent idiopathic scoliosis using biplanar radiographs of the chest with 3D reconstruction . Bone Joint J . 2022 ; 104-B ( 1 ): 112 – 119 . Crossref PubMed Google Scholar
27. Ferrero E , Lafage R , Vira S , et al. Three-dimensional reconstruction using stereoradiography for evaluating adult spinal deformity: a reproducibility study . Eur Spine J . 2017 ; 26 ( 8 ): 2112 – 2120 . Crossref PubMed Google Scholar
28. Ilharreborde B , Steffen JS , Nectoux E , et al. Angle measurement reproducibility using EOS three-dimensional reconstructions in adolescent idiopathic scoliosis treated by posterior instrumentation . Spine (Phila Pa 1976) . 2011 ; 36 ( 20 ): E1306 – 13 . Crossref PubMed Google Scholar
29. Wan S-T , Wong D-L , To S-H , Meng N , Zhang T , Cheung J-Y . Patient and surgical predictors of 3D correction in posterior spinal fusion: a systematic review . Eur Spine J . 2023 ; 32 ( 6 ): 1927 – 1946 . Crossref PubMed Google Scholar
30. Sullivan TB , Reighard FG , Osborn EJ , Parvaresh KC , Newton PO . Thoracic idiopathic scoliosis severity is highly correlated with 3d measures of thoracic kyphosis . J Bone Joint Surg Am . 2017 ; 99-A ( 11 ): e54 . Crossref PubMed Google Scholar
31. Thenard T , Vergari C , Hernandez T , Vialle R , Skalli W . Analysis of center of mass and gravity-induced vertebral axial torque on the scoliotic spine by barycentremetry . Spine Deform . 2019 ; 7 ( 4 ): 525 – 532 . Crossref PubMed Google Scholar
32. Garg B , Mehta N , Bansal T , Malhotra R . EOS imaging: concept and current applications in spinal disorders . J Clin Orthop Trauma . 2020 ; 11 ( 5 ): 786 – 793 . Crossref Google Scholar
33. Vergari C , Gajny L , Ebermeyer E , et al. Early detection of progressive idiopathic scoliosis through the quasi-automatic 3D reconstruction of the spine from biplanar radiography . Eur Spine J . 2019 ; 28 : 2878 – 2879 . Crossref PubMed Google Scholar
34. Lv Z , Lv W , Wang L , Ou J . Development and validation of machine learning-based models for prediction of adolescent idiopathic scoliosis: a retrospective study . Medicine (Baltimore) . 2023 ; 102 ( 14 ): e33441 . Crossref PubMed Google Scholar
35. Meng N , Wong K-Y , Zhao M , Cheung JPY , Zhang T . Radiograph-comparable image synthesis for spine alignment analysis using deep learning with prospective clinical validation . EClinicalMedicine . 2023 ; 61 : 102050 . Crossref PubMed Google Scholar
36. Zhang T , Zhu C , Zhao Y , et al. Deep learning model to classify and monitor idiopathic scoliosis in adolescents using a single smartphone photograph . JAMA Netw Open . 2023 ; 6 ( 8 ): e2330617 . Crossref PubMed Google Scholar
37. Moher D , Liberati A , Tetzlaff J , Altman DG , PRISMA Group . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement . PLoS Med . 2009 ; 6 ( 7 ): e1000097 . Crossref PubMed Google Scholar
38. Hayden JA , van der Windt DA , Cartwright JL , Côté P , Bombardier C . Assessing bias in studies of prognostic factors . Ann Intern Med . 2013 ; 158 ( 4 ): 280 – 286 . Crossref PubMed Google Scholar
39. Downes MJ , Brennan ML , Williams HC , Dean RS . Development of a critical appraisal tool to assess the quality of cross-sectional studies (AXIS) . BMJ Open . 2016 ; 6 ( 12 ): e011458 . Crossref PubMed Google Scholar
40. Guyatt G , Oxman AD , Akl EA , et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables . J Clin Epidemiol . 2011 ; 64 ( 4 ): 383 – 394 . Crossref PubMed Google Scholar
41. Balshem H , Helfand M , Schünemann HJ , et al. GRADE guidelines: 3. Rating the quality of evidence . J Clin Epidemiol . 2011 ; 64 ( 4 ): 401 – 406 . Crossref PubMed Google Scholar
42. Zhang Y , Coello PA , Guyatt GH , et al. GRADE guidelines: 20. Assessing the certainty of evidence in the importance of outcomes or values and preferences-inconsistency, imprecision, and other domains . J Clin Epidemiol . 2019 ; 111 : 83 – 93 . Crossref PubMed Google Scholar
43. Zhang Y , Alonso-Coello P , Guyatt GH , et al. GRADE guidelines: 19. Assessing the certainty of evidence in the importance of outcomes or values and preferences-Risk of bias and indirectness . J Clin Epidemiol . 2019 ; 111 : 94 – 104 . Crossref PubMed Google Scholar
44. Guyatt GH , Oxman AD , Montori V , et al. GRADE guidelines: 5. Rating the quality of evidence--publication bias . J Clin Epidemiol . 2011 ; 64 ( 12 ): 1277 – 1282 . Crossref PubMed Google Scholar
45. Guyatt GH , Oxman AD , Sultan S , et al. GRADE guidelines: 9. Rating up the quality of evidence . J Clin Epidemiol . 2011 ; 64 ( 12 ): 1311 – 1316 . Crossref PubMed Google Scholar
46. Nault M-L , Mac-Thiong J-M , Roy-Beaudry M , et al. Three-dimensional spinal morphology can differentiate between progressive and nonprogressive patients with adolescent idiopathic scoliosis at the initial presentation: a prospective study . Spine (Phila Pa 1976) . 2014 ; 39 ( 10 ): E601 – 6 . Crossref PubMed Google Scholar
47. Nault M-L , Beauséjour M , Roy-Beaudry M , et al. A predictive model of progression for adolescent idiopathic scoliosis based on 3D spine parameters at first visit . Spine (Phila Pa 1976) . 2020 ; 45 ( 9 ): 605 – 611 . Crossref PubMed Google Scholar
48. Almansour H , Pepke W , Bruckner T , Diebo BG , Akbar M . Three-dimensional analysis of initial brace correction in the setting of adolescent idiopathic scoliosis . J Clin Med . 2019 ; 8 ( 11 ): 1804 . Crossref PubMed Google Scholar
49. Skalli W , Vergari C , Ebermeyer E , et al. Early detection of progressive adolescent idiopathic scoliosis: a severity index . Spine (Phila Pa 1976) . 2017 ; 42 ( 11 ): 823 – 830 . Crossref PubMed Google Scholar
50. Begon M , Scherrer S-A , Coillard C , Rivard C-H , Allard P . Three-dimensional vertebral wedging and pelvic asymmetries in the early stages of adolescent idiopathic scoliosis . Spine J . 2015 ; 15 ( 3 ): 477 – 486 . Crossref PubMed Google Scholar
51. Scherrer S-A , Begon M , Leardini A , Coillard C , Rivard C-H , Allard P . Three-dimensional vertebral wedging in mild and moderate adolescent idiopathic scoliosis . PLoS One . 2013 ; 8 ( 8 ): e71504 . Crossref PubMed Google Scholar
52. Vergari C , Karam M , Pietton R , et al. Spine slenderness and wedging in adolescent idiopathic scoliosis and in asymptomatic population: an observational retrospective study . Eur Spine J . 2020 ; 29 ( 4 ): 726 – 736 . Crossref PubMed Google Scholar
53. Karam M , Vergari C , Skalli W , et al. Assessment of the axial plane deformity in subjects with adolescent idiopathic scoliosis and its relationship to the frontal and sagittal planes . Spine Deform . 2022 ; 10 ( 3 ): 509 – 514 . Crossref PubMed Google Scholar
54. Bisson DG , Sheng K , Kocabas S , et al. Axial rotation and pain are associated with facet joint osteoarthritis in adolescent idiopathic scoliosis . Osteoarthritis Cartilage . 2023 ; 31 ( 8 ): 1101 – 1110 . Crossref PubMed Google Scholar
55. Karam M , Bizdikian AJ , Khalil N , et al. Alterations of 3D acetabular and lower limb parameters in adolescent idiopathic scoliosis . Eur Spine J . 2020 ; 29 ( 8 ): 2010 – 2017 . Crossref PubMed Google Scholar
56. Pasha S , Ecker M , Deeney V . Considerations in sagittal evaluation of the scoliotic spine . Eur J Orthop Surg Traumatol . 2018 ; 28 ( 6 ): 1039 – 1045 . Crossref PubMed Google Scholar
57. Pasha S , Capraro A , Cahill PJ , Dormans JP , Flynn JM . Bi-planar spinal stereoradiography of adolescent idiopathic scoliosis: considerations in 3D alignment and functional balance . Eur Spine J . 2016 ; 25 ( 10 ): 3234 – 3241 . Crossref PubMed Google Scholar
58. Villemure I , Aubin CE , Grimard G , Dansereau J , Labelle H . Progression of vertebral and spinal three-dimensional deformities in adolescent idiopathic scoliosis: a longitudinal study . Spine (Phila Pa 1976) . 2001 ; 26 ( 20 ): 2244 – 2250 . Crossref PubMed Google Scholar
59. Fitzgerald R , Upasani VV , Bastrom TP , et al. Three-dimensional radiographic analysis of two distinct Lenke 1A curve patterns . Spine Deform . 2019 ; 7 ( 1 ): 66 – 70 . Crossref PubMed Google Scholar
60. Karam M , Ghanem I , Vergari C , et al. Global malalignment in adolescent idiopathic scoliosis: the axial deformity is the main driver . Eur Spine J . 2022 ; 31 ( 9 ): 2326 – 2338 . Crossref PubMed Google Scholar
61. Steib JP , Dumas R , Mitton D , Skalli W . Surgical correction of scoliosis by in situ contouring: a detorsion analysis . Spine (Phila Pa 1976) . 2004 ; 29 ( 2 ): 193 – 199 . Crossref PubMed Google Scholar
62. Courvoisier A , Drevelle X , Vialle R , Dubousset J , Skalli W . 3D analysis of brace treatment in idiopathic scoliosis . Eur Spine J . 2013 ; 22 ( 11 ): 2449 – 2455 . Crossref PubMed Google Scholar
63. Kadoury S , Shen J , Parent S . Global geometric torsion estimation in adolescent idiopathic scoliosis . Med Biol Eng Comput . 2014 ; 52 ( 4 ): 309 – 319 . Crossref PubMed Google Scholar
64. Vergari C , Ribes G , Aubert B , et al. Evaluation of a patient-specific finite-element model to simulate conservative treatment in adolescent idiopathic scoliosis . Spine Deform . 2015 ; 3 ( 1 ): 4 – 11 . Crossref PubMed Google Scholar
65. Courvoisier A , Drevelle X , Dubousset J , Skalli W . Transverse plane 3D analysis of mild scoliosis . Eur Spine J . 2013 ; 22 ( 11 ): 2427 – 2432 . Crossref PubMed Google Scholar
66. García-Cano E , Arámbula Cosío F , Duong L , et al. Prediction of spinal curve progression in adolescent idiopathic scoliosis using random forest regression . Comput Biol Med . 2018 ; 103 : 34 – 43 . Crossref PubMed Google Scholar
67. Kadoury S , Mandel W , Roy-Beaudry M , Nault ML , Parent S . 3D morphology prediction of progressive spinal deformities from probabilistic modeling of discriminant manifolds . IEEE Trans Med Imaging . 2017 ; 36 ( 5 ): 1194 – 1204 . Crossref PubMed Google Scholar
68. Drevelle X , Lafon Y , Ebermeyer E , Courtois I , Dubousset J , Skalli W . Analysis of idiopathic scoliosis progression by using numerical simulation . Spine (Phila Pa 1976) . 2010 ; 35 ( 10 ): E407 – 12 . Crossref PubMed Google Scholar
69. Marya S , Tambe AD , Millner PA , Tsirikos AI . Adolescent idiopathic scoliosis: a review of aetiological theories of a multifactorial disease . Bone Joint J . 2022 ; 104-B ( 8 ): 915 – 921 . Crossref PubMed Google Scholar
70. Pérez-Machado G , Berenguer-Pascual E , Bovea-Marco M , et al. From genetics to epigenetics to unravel the etiology of adolescent idiopathic scoliosis . Bone . 2020 ; 140 : 115563 . Crossref PubMed Google Scholar
71. Kouwenhoven J-W , Castelein RM . The pathogenesis of adolescent idiopathic scoliosis: review of the literature . Spine (Phila Pa 1976) . 2008 ; 33 ( 26 ): 2898 – 2908 . Crossref PubMed Google Scholar
72. Schlösser TPC , Castelein RM , Grobost P , Shah SA , Abelin-Genevois K . Specific sagittal alignment patterns are already present in mild adolescent idiopathic scoliosis . Eur Spine J . 2021 ; 30 ( 7 ): 1881 – 1887 . Crossref PubMed Google Scholar
73. Maqsood A , Hashmi SZ , Hartwell M , Sarwark JF . Idiopathic scoliosis: a pilot MR study of early vertebral morphological changes and spinal asymmetry . J Orthop . 2020 ; 19 : 174 – 177 . Crossref PubMed Google Scholar
74. Cheung WK , Cheung JPY . Contribution of coronal vertebral and IVD wedging to Cobb angle changes in adolescent idiopathic scoliosis during growth . BMC Musculoskelet Disord . 2022 ; 23 ( 1 ): 904 . Crossref PubMed Google Scholar
75. Nault M-L , Mac-Thiong J-M , Roy-Beaudry M , et al. Three-dimensional spinal morphology can differentiate between progressive and nonprogressive patients with adolescent idiopathic scoliosis at the initial presentation: a prospective study . Spine (Phila Pa 1976) . 2014 ; 39 ( 10 ): E601 – 6 . Crossref PubMed Google Scholar
76. Crijns TJ , Stadhouder A , Smit TH . Restrained differential growth: the initiating event of adolescent idiopathic scoliosis? Spine (Phila Pa 1976) . 2017 ; 42 ( 12 ): E726 – E732 . Crossref PubMed Google Scholar
77. Meiring AR , de Kater EP , Stadhouder A , van Royen BJ , Breedveld P , Smit TH . Current models to understand the onset and progression of scoliotic deformities in adolescent idiopathic scoliosis: a systematic review . Spine Deform . 2023 ; 11 ( 3 ): 545 – 558 . Crossref PubMed Google Scholar
78. Newton PO , Osborn EJ , Bastrom TP , Doan JD , Reighard FG . The 3D sagittal profile of thoracic versus lumbar major curves in adolescent idiopathic scoliosis . Spine Deform . 2019 ; 7 ( 1 ): 60 – 65 . Crossref PubMed Google Scholar
79. Shen K , Clement RC , Yaszay B , Bastrom T , Upasani VV , Newton PO . Three-dimensional analysis of the sagittal profile in surgically treated Lenke 5 curves in adolescent idiopathic scoliosis . Spine Deform . 2020 ; 8 ( 6 ): 1287 – 1294 . Crossref PubMed Google Scholar
80. Vergari C , Courtois I , Ebermeyer E , et al. Head to pelvis alignment of adolescent idiopathic scoliosis patients both in and out of brace . Eur Spine J . 2019 ; 28 ( 6 ): 1286 – 1295 . Crossref PubMed Google Scholar
81. Schlösser TPC , van Stralen M , Chu WCW , et al. Anterior overgrowth in primary curves, compensatory curves and junctional segments in adolescent idiopathic scoliosis . PLoS One . 2016 ; 11 ( 7 ): e0160267 . Crossref PubMed Google Scholar
82. Upasani VV , Tis J , Bastrom T , et al. Analysis of sagittal alignment in thoracic and thoracolumbar curves in adolescent idiopathic scoliosis: how do these two curve types differ? Spine (Phila Pa 1976) . 2007 ; 32 ( 12 ): 1355 – 1359 . Crossref PubMed Google Scholar
83. Schlösser TPC , Shah SA , Reichard SJ , Rogers K , Vincken KL , Castelein RM . Differences in early sagittal plane alignment between thoracic and lumbar adolescent idiopathic scoliosis . Spine J . 2014 ; 14 ( 2 ): 282 – 290 . Crossref PubMed Google Scholar
84. Pasha S , Baldwin K . Preoperative sagittal spinal profile of adolescent idiopathic scoliosis Lenke types and non-scoliotic adolescents: a systematic review and meta-analysis . Spine (Phila Pa 1976) . 2019 ; 44 ( 2 ): 134 – 142 . Crossref PubMed Google Scholar
85. Adam CJ , Askin GN , Pearcy MJ . Gravity-induced torque and intravertebral rotation in idiopathic scoliosis . Spine (Phila Pa 1976) . 2008 ; 33 ( 2 ): E30 – 7 . Crossref PubMed Google Scholar
86. Villemure I , Aubin CE , Grimard G , Dansereau J , Labelle H . Progression of vertebral and spinal three-dimensional deformities in adolescent idiopathic scoliosis: a longitudinal study . Spine (Phila Pa 1976) . 2001 ; 26 ( 20 ): 2244 – 2250 . Crossref PubMed Google Scholar
87. Hayashi K , Upasani VV , Pawelek JB , et al. Three-dimensional analysis of thoracic apical sagittal alignment in adolescent idiopathic scoliosis . Spine (Phila Pa 1976) . 2009 ; 34 ( 8 ): 792 – 797 . Crossref PubMed Google Scholar
88. Sullivan TB , Bastrom TP , Bartley CE , Dolan LA , Weinstein SL , Newton PO . More severe thoracic idiopathic scoliosis is associated with a greater three-dimensional loss of thoracic kyphosis . Spine Deform . 2020 ; 8 ( 6 ): 1205 – 1211 . Crossref PubMed Google Scholar
89. Parvaresh KC , Osborn EJ , Reighard FG , Doan J , Bastrom TP , Newton PO . Predicting 3D thoracic kyphosis using traditional 2D radiographic measurements in adolescent idiopathic scoliosis . Spine Deform . 2017 ; 5 ( 3 ): 159 – 165 . Crossref PubMed Google Scholar
90. Sullivan TB , Bastrom T , Reighard F , Jeffords M , Newton PO . A novel method for estimating three-dimensional apical vertebral rotation using two-dimensional coronal cobb angle and thoracic kyphosis . Spine Deform . 2017 ; 5 ( 4 ): 244 – 249 . Crossref PubMed Google Scholar
91. Logithasan V , Wong J , Reformat M , Lou E . Using machine learning to automatically measure axial vertebral rotation on radiographs in adolescents with idiopathic scoliosis . Med Eng Phys . 2022 ; 107 : 103848 . Crossref PubMed Google Scholar
92. Bolzinger M , Bernardini I , Thevenin Lemoine C , Gallini A , Accadbled F , Sales de Gauzy J . Monitoring adolescent idiopathic scoliosis by measuring ribs prominence using surface topography device . Spine Deform . 2021 ; 9 ( 5 ): 1349 – 1354 . Crossref PubMed Google Scholar
93. Wei JZ , Cheung BKC , Chu SLH , et al. Assessment of reliability and validity of a handheld surface spine scanner for measuring trunk rotation in adolescent idiopathic scoliosis . Spine Deform . 2023 ; 11 ( 6 ): 1347 – 1354 . Crossref PubMed Google Scholar
94. Vendeuvre T , Tabard-Fougère A , Armand S , Dayer R . Test characteristics of rasterstereography for the early diagnosis of adolescent idiopathic scoliosis . Bone Joint J . 2023 ; 105-B ( 4 ): 431 – 438 . Crossref PubMed Google Scholar
Author contributions
H-T. S. Wan: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing - original draft, Writing - review & editing.
D. L. L. Wong: Data curation, Formal analysis, Investigation, Methodology, Writing - original draft, Writing - review & editing.
C-H. S. To: Data curation, Formal analysis, Investigation.
N. Meng: Data curation.
T. Zhang: Writing - review & editing.
J. P. Y. Cheung: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing - review & editing.
Funding statement
The authors received no financial or material support for the research, authorship, and/or publication of this article.
ICMJE COI statement
The authors confirm that they have no disclosures to declare.
Data sharing
All data generated or analyzed during this study are included in the published article and/or in the supplementary material.
Open access funding
The authors report that they received open access funding for this manuscript from the Department of Orthopaedics and Traumatology, The University of Hong Kong, Hong Kong SAR, China.
Supplementary material
Tables showing the search strategy; details of included studies; risk of bias assessed using Quality in Prognostic Studies (QUIPS); coronal, sagittal, and axial parameters; machine-learning methods; and details of the included 3D parameters.
Social media
Follow J. P. Y. Cheung on X @jasonpycheung
© 2024 Cheung et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
