Search

The initial assessment and postoperative monitoring of patients with various abnormalities of the foot in clinical routine practice is primarily based on the analysis of radiographs taken in the weightbearing position. Conventional x-ray imaging, however, only provides a 2-dimensional projection of 3-dimensional (3D) bony structures, and the clinical parameters assessed from these images can be affected by projection biases. In the present work, we addressed this issue by proposing an accurate 3D reconstruction method of the foot in the weightbearing position from low-dose biplanar radiographs with clinical index measurement assessment for clinical routine practice. The accuracy of the proposed reconstruction method was evaluated for both shape and clinical indexes by comparing 3D reconstructions of 6 cadaveric adult feet from computed tomographic images and from biplanar radiographs. For the reproducibility study, 3D reconstructions from the biplanar radiographs of the foot of 6 able-bodied subjects were considered, with 2 observers repeating each measurement of anatomic landmarks 3 times. Baseline assessment of important 3D clinical parameters was performed on 17 subjects (34 feet; mean age 27.7, range 20 to 52 years). The average point to surface distance between the 3D stereoradiographic reconstruction and the computed tomographic scan-based reconstruction was 1 mm (range 0mm to 6mm). The selected radiographic landmarks were highly reproducible (95% confidence interval

Resume Introduction La prise en charge de la scoliose idiopathique de l’adolescent (SIA) peut necessiter des traitements chirurgicaux aux consequences lourdes. L’existence d’un desequilibre frontal, authentifie par l’examen clinique a l’aide d’un fil a plomb (DFFP) est un element important qui pousse a un traitement chirurgical. La technique de reference pour l’appreciation de l’equilibre postural du tronc reste neanmoins la materialisation de la ligne de gravite (LG) et sa projection dans un referentiel determine et connu. De recentes etudes ont montre la possibilite de materialiser la LG d’un patient a partir de stereoradiographies apres reconstruction de l’enveloppe corporelle (technique de barycentremetrie). L’hypothese que nous nous proposons de verifier est qu’il existe une concordance entre le DFFP et la LG. Materiels et methodes Dans un premier temps la determination de la LG par la methode de barycentremetrie a ete validee chez 14 SIA en la comparant aux donnees recueillies sur une plate-forme de force. Les valeurs normales de DFFP et LG ont ensuite ete determinees chez 27 adolescents asymptomatiques au moyen d’une technique stereoradiographique. Une comparaison de ces memes valeurs a ete ensuite realisee sur une population de 53 patients porteurs d’une SIA. Resultats La fiabilite des mesures de la LG dans le plan frontal avec la technique de barycentremetrie etait de 2,4 mm (intervalle de confiance a 95 %). Il existait une concordance statistiquement forte entre la distance de la LG par rapport au milieu de la ligne joignant les centres des 2 tetes femorales (LG/HA) et le DFFP calcule comme la distance entre la verticale au milieu de T1 et le centre du plateau sacre (T1 V/S) : r = 0,71 p Conclusion Cette etude valide la determination de la LG a partir de stereoradiographies dans une population de SIA. Le DFFP etait statistiquement correle a la LG, mais moins precis pour decider d’une indication chirurgicale. Niveau de preuve IV, Etude retrospective.

RESUME Les modeles biomecaniques sont maintenant utilises dans les crashs tests virtuels en amont de la construction de prototypes de vehicules. Cependant ces modeles ne prennent que partiellement en compte l’extreme variabilite du corps humain. Cette variabilite souleve des defis lies a la personnalisation des modeles, qu’il s’agisse de la geometrie ou des proprietes mecaniques des tissus. Les progres en imagerie medicale et en particulier la radiographie biplane et les reconstructions 3D associees, utilisees en routine clinique, ont permis des avancees drastiques en modelisation geometrique et donnent aujourd’hui acces a de larges bases de donnees permettant d’analyser et de modeliser les variations interindividuelles. Les connaissances quant a la caracterisation des proprietes mecaniques osseuses ont egalement connu une progression rapide. La caracterisation in vivo des proprietes mecaniques des tissus mous du sujet humain beneficie du developpement recent des moyens d’elastographie ultrasonore. Au-dela du domaine de la securite routiere, les recherches en modelisation personnalisee sont egalement actives en orthopedie et traumatologie pour l’identification de sujets a risques ou pour la planification des traitements conservateurs ou chirurgicaux. De veritables « outils metiers » devraient emerger dans un futur proche pour l’aide a la reflexion, au diagnostic et a la planification d’une strategie de traitement tenant compte des specificites de chaque patient, ouvrant ainsi une voie vers la medecine individualisee.

Purpose The aim of this work is to propose a classification algorithm to automatically detect treatment for scoliosis (brace, implant or no treatment) in postero-anterior radiographs. Such automatic labelling of radiographs could represent a step towards global automatic radiological analysis. Methods Seven hundred and ninety-six frontal radiographies of adolescents were collected (84 patients wearing a brace, 325 with a spinal implant and 387 reference images with no treatment). The dataset was augmented to a total of 2096 images. A classification model was built, composed by a forward convolutional neural network (CNN) followed by a discriminant analysis; the output was a probability for a given image to contain a brace, a spinal implant or none. The model was validated with a stratified tenfold cross-validation procedure. Performance was estimated by calculating the average accuracy. Results 98.3% of the radiographs were correctly classified as either reference, brace or implant, excluding 2.0% unclassified images. 99.7% of brace radiographs were correctly detected, while most of the errors occurred in the reference group (i.e. 2.1% of reference images were wrongly classified). Conclusion The proposed classification model, the originality of which is the coupling of a CNN with discriminant analysis, can be used to automatically label radiographs for the presence of scoliosis treatment. This information is usually missing from DICOM metadata, so such method could facilitate the use of large databases. Furthermore, the same model architecture could potentially be applied for other radiograph classifications, such as sex and presence of scoliotic deformity. Acknowledgements The authors are grateful to the ParisTech BiomecAM chair program on subject-specific musculoskeletal modelling (with the support of ParisTech and Yves Cotrel Foundations, Société Générale, Proteor and Covea).

Introduction: Maintaining balance during gait allows subjects to minimize energy expenditure and avoid falls. Gait balance can be measured by assessing the relationship between the center of mass (COM) and center of pressure (COP) during gait. Demographics, skeletal and postural parameters are known to influence gait balance. Purpose: What are the determinants of dynamic balance during gait in asymptomatic adults among skeletal and demographic parameters? Methods: 115 adults underwent 3D gait analysis and full-body biplanar X-rays. Angles between the COM-COP line and the vertical were calculated in frontal and sagittal planes during gait: maxima, minima, and ROM were evaluated. Full-body 3D reconstructions were obtained; skeletal and postural parameters of the spine (lumbar lordosis, thoracic kyphosis, sagittal vertical axis SVA), pelvis (pelvic tilt and incidence, acetabular orientation in the 3 planes) and lower limbs (neck shaft angle femoral and tibial torsions) were calculated. A univariate followed by a multivariate analysis were computed between the COM-COP parameters and skeletal and demographic parameters. Results: The univariate analysis showed that in the frontal plane, maximum (4.6°) of the COMCOP angle was significantly correlated with weight (r =0.53), age (r =0.28), height (r = 0.35), SVA (r = 0.23), T1T12 (r = 0.24) and pelvic width (r = 0.25).In the sagittal plane, maximum COM-COP (19.7 ± 2.8°) angle was significantly correlated to acetabular tilt (r = 0.25) and acetabular anteversion (r =0.21). The multivariate analysis showed that, in the frontal plane, an increase in the maximum of the COM-COP angle was determined by a decreasing height (β = −0.28), an increasing weight (β = 0.48), being a male (β = −0.42), and an increasing posterior acetabular coverage (β = 0.22). In the sagittal plane, an increasing maximum COMCOP angle was determined by a decreasing height (β = −0.38) and an increasing SVA (β = 0.19). Conclusion: Frontal imbalance appeared to be mainly correlated to demographic parameters. Sagittal imbalance was found to be correlated with weight, height, acetabular parameters and SVA. These results suggest that in addition to demographic parameters, acetabular parameters and SVA are important determinants of balance during gait. This research was funded by the University of Saint-Joseph (grant FM183). The funding source did not intervene in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

The data collection was approved of by the ethical commit-tee (CPP 6001 Ile de France V), and patients and their parents signed an informed consent. Bracing is the most common treatment to stop the progression of adolescent idiopathic scoliosis. Finite element modeling could help improve brace design, but model validation is still a challenge. In this work, the clinical relevance of a predictive and subject-specific model for bracing was evaluated in forty-six AIS patients. The model reproduces brace action and the patient’s spinopelvic adjustments to keep balance. The model simulated 70% or more patients with geometrical parameters within a preselected tolerance level. Although the model simulation of the sagittal plane could be improved, the approach is promising for a realistic and predictive simulation of brace action. The authors are grateful to the ParisTech BiomecAM chair program on subject-specific musculoskeletal modelling (with the support of ParisTech and Yves Cotrel Foundations, Proteor, Société Génerale and Covea). We are also thankful to David Barrie Colridge for his support.

The scapholunate joint is one of the keystones of the wrist kinematics, and its study is difficult due to the carpal bones size and the richness of surrounding ligaments. We propose a new method of quantitative assessment of scapholunate kinematics through bone motion tracking in order to investigate scapholunate ligament lesion as well as repair techniques. On 6 intact wrists, steel beads were inserted into the bones of interest to track their motions. Experimental set up allowed wrist flexion extension and radio-ulnar deviation motions. Low-dose bi-planar radiographs were performed each 10° of movement for different configurations: 1) intact wrist, 2) scapholunate ligament division, 3) repair by soft anchors at the posterior then 4) anterior part. Beads' 3D coordinates were computed at each position from biplanar X-Rays, allowing accurate registration of each wrist bone. The Monte Carlo sensitivity study showed accuracy between 0.2° and 1.6 ° for the scaphoid and the lunate in motions studied. The maximum flexion-extension range of motion of the scaphoid significantly decreased after anterior repair from 73° in injured wrist to 62.7°. The proposed protocol appears robust, and the tracking allowed to quantify the anchor's influence on the wrist kinematics. ZimmerBiomet, Biomecam Chair program on subject specific musculoskeletal modelling.

Purpose: To evaluate the 3D deformity of the acetabula and lower limbs in subjects with adolescent idiopathic scoliosis (AIS) and their relationship with spino-pelvic alignment. Methods: Two hundred and seventy-four subjects with AIS (frontal Cobb: 33.5° ± 18° [10°-110°]) and 84 controls were enrolled. All subjects underwent full-body biplanar X-rays with subsequent 3D reconstructions. Classic spino-pelvic and lower limb parameters were collected as well as acetabular parameters: acetabular orientation in the 3 planes (tilt, anteversion and abduction), center-edge angle (CEA) and anterior and posterior sector angles. Subjects with AIS were represented by both lower limb sides and classified by elevated (ES) or lowered (LS), depending on the frontal pelvic obliquity. Parameters were then compared between groups. Determinants of acetabular and lower limb alterations were investigated among spino-pelvic parameters. Results: Acetabular abduction was higher on the ES in AIS (59.2° ± 6°) when compared to both LS (55.6° ± 6°) and controls (57.5° ± 3.9°, p < 0.001). CEA and acetabular anteversion were higher on the LS in AIS (32° ± 6.1°, 20.5° ± 5.7°) when compared to both ES (28.7° ± 5.1°, 19.8° ± 5.1°) and controls (29.8° ± 4.8°, 19.1° ± 4°, respectively, p < 0.001). Anterior sector angle was lower on both ES and LS in AIS when compared to controls. CEA, acetabular abduction and acetabular anteversion were found to be mostly determined (adjusted R2: 0.08-0.32) by pelvic tilt and less by frontal pelvic obliquity, frontal Cobb and T1T12. Conclusions: Subjects with AIS had a more abducted acetabulum at the lowered side, more anteverted acetabulum and a lack of anterior coverage of both acetabula. These alterations were strongly related to pelvic tilt. This study was funded by the University of Saint-Joseph (Grant No. FM300).

Purpose This in vitro biomechanical study compares residual lumbar range of motion (ROM) and rod strain after lumbopelvic instrumentation using 2 rods, 4 rods and interbody cages. Methods Seven human cadaveric specimens were instrumented from L1 to sacrum, and pelvic screws were implanted. The pelvis was constrained and moments up to 7.5 Nm were applied to T12. Segmental L1–S1 ROM was analyzed by tracking radiopaque balls implanted in each vertebra using biplanar radiographs. Deformation within principal rods was measured by strain gauges. Four configurations were compared: 2 rods (2R), 4 rods (4R), 4 rods + ALIF at L4–L5 and L5–S1 (4R + ALIF), 2 rods + ALIF (2R + ALIF). Results Intact average global L1–S1 ROM was 42.9° (27.9°–66.0°) in flexion–extension (FE), 35.2° (26.8°–51.8°) in lateral bending (LB), 18.6° (6.7°–47.8°) in axial rotation (AR). In FE, average ROM was 1.9° with both 4-rod configurations versus 2.5° with 2R and 2.8° with 2R + ALIF (p < 0.05). In LB, ROM ranged between 1.2° and 1.5° without significant differences. In AR, ROM was 2.5° with both 4-rod configurations versus 2.9° with 2R (p = 0.07) and 3.1° with 2R ALIF (p = 0.01). In FE, strain decreased by 64% and 65% in principal rods at L3–L4 with 4-rod. When comparing 2-rod configurations, strain decreased by 1% in flexion and increased by 22% in extension at L3–L4 when adding an ALIF at L4–L5 and L5–S1. Conclusions Double rods and interbody cages decrease residual ROM in FE and AT. Double rods seem efficient in limiting strain in principal rods. The use of single rods with cages at the lumbosacral junction increases strain at the first adjacent level without cage. A research grant was obtained from the Société Française de Chirurgie Rachidienne (SFCR). Technical and financial support was provided by Clariance.

International audience; Background: The magnetic resonance imaging (MRI) parameters used to diagnose rotator cuff tears are weakly correlated to shoulder function. Our hypothesis was that adding 3-dimensional morphologic parameters resulting from biplanar radiographs (3DXR parameters) to the MRI parameters would improve this correlation.Methods: We assessed 52 patients with rotator cuff tears with an EOS Imaging radiographic examination, MRI study, and clinical evaluation of the shoulder, as well as the Constant score. The bones of the 52 shoulders were reconstructed 3-dimensionally, and eleven 3DXR parameters were automatically extracted. First, the trueness and reliability of these parameters were evaluated. Then, bivariate correlations between each parameter and the Constant score were made. A linear regression model was subsequently built to correlate the 11 parameters and 5 MRI findings with shoulder function at diagnosis, as assessed by the Constant score.Results: The parameters showed good trueness and reliability of most 3DXR parameters. Supraspinatus tear extension, muscle atrophy, and the distance between the greater and deltoid tuberosities were the only parameters with a statistically significant correlation to a lower Constant score (P < .05) in the bivariate study. These correlations were either weak or negligible. A regression model was successfully built with one MRI parameter and four 3DXR parameters. Correlation to function increased from 16.7% to 43.3% with this model.Conclusion: For patients with rotator cuff tears, the combination of MRI and 3DXR parameters of the shoulder in a linear regression model improves the correlation with the Constant score (shoulder function) at diagnosis.

International audience; Study design Biomechanical human cadaver study.Objective To determine the three-dimensional intervertebral ranges of motion (ROMs) of intact and hook-instrumented tho-racic spine specimens subjected to physiological loads, using an in vitro experimental protocol with EOS biplane radiography.Summary of background data Pedicle screws are commonly used in thoracic instrumentation constructs, and their biome-chanical properties have been widely studied. Promising clinical results have been reported using a T1–T5 thoracic hook–claw construct for proximal rod anchoring. Instrumentation stability is a crucial factor in minimizing mechanical complications rates but had not been assessed for this construct in a biomechanical study.Methods Six fresh-frozen human cadaver C6–T7 thoracic spines were studied. The first thoracic vertebrae were instrumented using two claws of supra-laminar and pedicle hooks, each fixed on two adjacent vertebrae, on either side of a single free vertebra. Quasi-static pure-moment loads up to 5 Nm were applied to each specimen before and after instrumentation, in flexion–extension, right and left bending, and axial rotation. Five steel beads impacted in each vertebra allowed 3D tracking of vertebral movements on EOS biplanar radiographs acquired after each loading step. The relative ranges of motion (ROMs) of each pair of vertebras were computed.Results Mean ROMs with the intact specimens were 17° in flexion–extension, 27.9° in lateral bending, and 29.5° in axial rotation. Corresponding values with the instrumented specimens were 0.9°, 2.6°, and 7.3°, respectively. Instrumentation sig-nificantly (P < 0.05) decreased flexion–extension (by 92–98%), lateral bending (by 87–96%), and axial rotation (by 68–84%).Conclusion This study establishes the biomechanical stability of a double claw–hook construct in the upper thoracic spine, which may well explain the low mechanical complication rate in previous clinical studies.Level of evidence Not applicable, experimental cadaver study.

Objective To quantify muscle characteristics (volumes and fat infiltration) and identify their relationship to sagittal malalignment and compensatory mechanism recruitment. Methods Female adult spinal deformity patients underwent T1-weighted MRI with a 2-point Dixon protocol from the proximal tibia up to the T12 vertebra. 3D reconstructions of 17 muscles, including extensors and flexors of spine, hip and knee, were obtained. Muscle volume standardized by bone volume and percentage of fat infiltration (Pfat) were calculated. Correlations and regressions were performed. Results A total of 22 patients were included. Significant correlations were observed between sagittal alignment and muscle parameters. Fat infiltration of the hip and knee flexors and extensors correlated with larger C7-S1 SVA. Smaller spinal flexor/extensor volumes correlated with greater PI-LL mismatch (r = − 0.45 and − 0.51). Linear regression identified volume of biceps femoris as only predictor for PT (R2 = 0.34, p = 0.005) and Pfat of gluteus minimus as only predictor for SVA (R2 = 0.45, p = 0.001). Sagittally malaligned patients with larger PT (26.8° vs. 17.2°) had significantly smaller volume and larger Pfat of gluteus medius, gluteus minimus and biceps femoris, but similar values for gluteus maximus, the hip extensor. Conclusion This study is the first to quantify the relationship between degeneration of spino-femoral muscles and sagittal malalignment. This pathoanatomical study identifies the close relationship between gluteal, hamstring muscles and PT, SVA, which deepens our understanding of the underlying etiology that contributes to adult spinal deformity. The manuscript submitted does not contain information about medical device(s)/drug(s). This work received funding from Youth Fund of Natural Science Foundation of Jiangsu Province (BK20180122). This work received funding from Key Project supported by Medical Science and Technology Development Foundation, Nanjing Department of Health (YKK18092).

International audience; Introduction: Postural alignment is altered with spine deformities that might occur with age. Alteration of spino-pelvic and postural alignment parameters are known to affect daily life activities such as gait. It is still unknown how spino-pelvic and postural alignment parameters are related to gait kinematics. Research question: To assess the relationships between spino-pelvic/postural alignment parameters and gait kinematics in asymptomatic adults. Methods: 134 asymptomatic subjects (aged 18-59 years) underwent 3D gait analysis, from which kinematics of the pelvis and lower limbs were extracted in the 3 planes. Subjects then underwent full-body biplanar X-rays, from which skeletal 3D reconstructions and spino-pelvic and postural alignment parameters were obtained such as sagittal vertical axis (SVA), center of auditory meatus to hip axis plumbline (CAM-HA), thoracic kyphosis (TK) and radiologic pelvic tilt (rPT). In order to assess the influence of spino-pelvic and postural alignment parameters on gait kinematics a univariate followed by a multivariate analysis were performed. Results: SVA was related to knee flexion during loading response (β = 0.268); CAM-HA to ROM pelvic obliquity (β = -0.19); rPT to mean pelvic tilt (β = -0.185) and ROM pelvic obliquity (β = -0.297); TK to ROM hip flexion/extension in stance (β = -0.17), mean foot progression in stance (β = -0.329), walking speed (β = -0.19), foot off (β = 0.223) and step length (β = -0.181). Significance: This study showed that increasing SVA, CAM-HA, TK and rPT, which is known to occur in adults with spinal deformities, could alter gait kinematics. Increases in these parameters, even in asymptomatic subjects, were related to a retroverted pelvis during gait, a reduced pelvic obliquity and hip flexion/extension mobility, an increased knee flexion during loading response as well as an increase in external foot progression angle. This was associated with a decrease in the walking pace: reduced speed, step length and longer stance phase.

Gait balance,assessed by the angle formed between the line joining the center-of-mass (COM) to the center-of-pressure (COP) and the vertical during gait, has been shown to be related to skeletal-postural and anthro-pometric parameters [1]. Although skeletal-postural parameters are mea-sured on standing radiographs, they are known to vary during gait. There are currently no studies evaluating how the variations of skeletal-postural parameters during gait influence subject’s balance during walking. Research question How does the variation of skeletal-postural parameters during gait influence subject’s balance during walking? Methods 72 asymptomatic subjects (age: 28.6 ± 11 years [18–59], 29F) underwent 3D gait analysis [2] with additional markers on the thighs and shanks. The COM-COP angle with the vertical was calculated in both the frontal and sagittal planes during the gait cycle [3]( Fig. 1).Subjects then underwent low-dose full-body biplanar X-rays with themarkers still in place. 3D reconstructions were obtained for the spine,pelvis and lower limbs. 3D bones were registered at each frame of thegait cycle [4]. A new technique developed for this study, utilizingfiniteelement modelling, was used to reduce soft tissue artefacts. Skeletal-postural parameters were then computed during the gait cycle, using the 3D registered bones, at each time frame (Fig. 2); mean, minimum, maximum and ROM were calculated on the waveforms during the gaitcycle. In order to determine which varying skeletal-postural parameterduring gait determined the variation of the COM-COP angles, univariate analysis (Pearson’s correlation) followed by a multivariate analysis (stepwise-multiple-linear-regression models) were computed; COM-COP parameters were the dependant variables and varying skeletal-postural parameters during gait were the independent variables. Results Minimum (−14.2 ± 3.4°) and average (3.1 ± 1.6°) of the sagittal COM-COP angle were found to be determined by the minimum of the posterior coverage (post_cov) of the femoral head by the acetabulum during gait (β= 0.40; R2 = 0.16; p = 0.003 andβ= 0.32; R2 = 0.1;p = 0.001, respectively). ROM (33.9 ± 5.1°) and maximum(19.7 ± 2.8°) of the sagittal COM-COP angle were found to be related to the ROM (β= 0.29; R2 = 0.09; p = 0.03) and maximum (β= 0.34;R2 = 0.11; p = 0.006) of the acetabular abduction during gait, respectively. ROM of the frontal COM-COP angle (8.8 ± 2.51°) was found to be determined by the average of the post_cov (β= 0.51;R2 = 0.26; p = 0.004) during gait. Discussion This is the first study to evaluate how the variation of skeletal-postural parameters during walking influences the stability during gait(Fig. 3). A less pronounced posterior coverage of the acetabulum during gait predisposes to more instability by decreasing the minimum COM-COP angle; a more pronounced acetabular abduction (decreased lateralcoverage) during gait predisposes to more instability by increasing the ROM and the maximum of the COM-COP angle. Therefore, gait instability in the sagittal plane seems to be influenced by the variation of the posterior and lateral coverage of the femoral head by the acet-abulum during walking

Acetabular cup orientation is crucial for total hip arthroplasty (THA), and its malpositioning could lead to impingement and dislocation [1]. Acetabular cup orientation currently relies on static 3D hip parameters [2] and was shown to be related to changes in pelvic positioning [3]. While pelvic position varies during walking, it is still unknown how dynamic variation of hip orientation during gait could influence hip kinematics.

Patients with adult spinal deformities (ASD) are known to have altered postural alignment affecting their quality of life and activities of daily living, especially gait. The Sagittal Vertical Axis (SVA), a postural parameter calculated as the distance between the posterior corner of the sacrum and the C7-plumbline on full-body sagittal radiographs [1], has been shown to be highly altered in ASD. Even though this parameter is positional and could vary during gait, no studies have investigated its variation during walking even in asymptomatic subjects.

To evaluate the 3D deformity of the acetabula and lower limbs in subjects with adolescent idiopathic scoliosis (AIS) and their relationship with spino-pelvic alignment.Two hundred and seventy-four subjects with AIS (frontal Cobb: 33.5° ± 18° [10°-110°]) and 84 controls were enrolled. All subjects underwent full-body biplanar X-rays with subsequent 3D reconstructions. Classic spino-pelvic and lower limb parameters were collected as well as acetabular parameters: acetabular orientation in the 3 planes (tilt, anteversion and abduction), center-edge angle (CEA) and anterior and posterior sector angles. Subjects with AIS were represented by both lower limb sides and classified by elevated (ES) or lowered (LS), depending on the frontal pelvic obliquity. Parameters were then compared between groups. Determinants of acetabular and lower limb alterations were investigated among spino-pelvic parameters.Acetabular abduction was higher on the ES in AIS (59.2° ± 6°) when compared to both LS (55.6° ± 6°) and controls (57.5° ± 3.9°, p 0.001). CEA and acetabular anteversion were higher on the LS in AIS (32° ± 6.1°, 20.5° ± 5.7°) when compared to both ES (28.7° ± 5.1°, 19.8° ± 5.1°) and controls (29.8° ± 4.8°, 19.1° ± 4°, respectively, p 0.001). Anterior sector angle was lower on both ES and LS in AIS when compared to controls. CEA, acetabular abduction and acetabular anteversion were found to be mostly determined (adjusted RSubjects with AIS had a more abducted acetabulum at the lowered side, more anteverted acetabulum and a lack of anterior coverage of both acetabula. These alterations were strongly related to pelvic tilt.

Obstructive sleep apnea syndrome (OSAS) is associated with postural dysfunction characterized by abnormal spinal curvature and disturbance of balance and walking, whose pathophysiology is poorly understood. We hypothesized that it may be the result of a pathological interaction between postural and ventilatory functions. Twelve patients with OSAS (4 women, age 53 years [51–63] (median [quartiles]), apnea hypopnea index 31/h [24–41]) were compared with 12 healthy matched controls. Low dose biplanar X-rays (EOS® system) were acquired and personalized three-dimensional models of the spine and pelvis were reconstructed. We also estimated posturo-respiratory coupling by measurement of respiratory emergence, obtaining synchronized center of pressure data from a stabilometric platform and ventilation data recorded by an optico-electronic system of movement analysis. Compared with controls, OSAS patients, had cervical hyperextension with anterior projection of the head (angle OD-C7 12° [8; 14] vs. 5° [4; 8]; p = 0.002), and thoracic hyperkyphosis (angle T1–T12 65° [51; 71] vs. 49° [42; 59]; p = 0.039). Along the mediolateral axis: (1) center of pressure displacement was greater in OSAS patients, whose balance was poorer (19.2 mm [14.2; 31.5] vs. 8.5 [1.4; 17.8]; p = 0.008); (2) respiratory emergence was greater in OSAS patients, who showed increased postural disturbance of respiratory origin (19.2% [9.9; 24.0] vs. 8.1% [6.4; 10.4]; p = 0.028). These results are evidence for the centrally-mediated and primarily respiratory origin of the postural dysfunction in OSAS. It is characterized by an hyperextension of the cervical spine with a compensatory hyperkyphosis, and an alteration in posturo-respiratory coupling, apparently secondary to upper airway instability.

International audience; AbstractObjectives To determine lumbar intervertebral disc properties with shear wave elastography in adolescent idiopathic scoliotic(AIS) patients before and after surgery, and compare them with asymptomatic controls.Methods Twenty-five severe AIS patients with an indication for fusion surgery (15 ± 1.5 years old, the Cobb angle rangingbetween 40 and 93°) and fifty-nine asymptomatic adolescents (13 ± 2 years old) were included prospectively. Shear wave speed(SWS) was measured in the annulus fibrosus of L3-L4, L4-L5, and L5-S1 discs of each subject. In AIS patients, measurementstook place before surgery, and 3 months (N = 13) or 1 year after (N = 12).Results No difference was observed between disc levels in any group. When pooling disc levels, SWS was significantly higher inpreop AIS patients (4.0 ± 0.5 m/s) than in asymptomatic subjects (3.1 ± 0.5 m/s, p < 0.0001). SWS decreased 3 months postop(3.5 ± 0.3 m/s), and it decreased further towards normal values 1 year after (3.3 ± 0.4 m/s). SWS in preop AIS patients tended todecrease with the Cobb angle (Spearman’s rho = − 0.4, p = 0.05).Conclusion Shear wave elastography measurements showed that discs in AIS patients were altered relative to asymptomaticones, and this alteration tended to normalize 1 year post fusion surgery. Further studies should aim at determining if bracing ofmild scoliosis has an effect on disc properties.Key Points• Shear wave elastography shows alteration of annulus fibrosus in adolescent idiopathic scoliosis.• Disc elastography tends to normalize 1 year after surgery.

Introduction In children treated for idiopathic equinovarus clubfoot (EVCF), the relation between morphologic defects on clinical examination and standard X-ray on the one hand and functional abnormalities on the other is difficult to objectify. The aim of the present study was to demonstrate the feasibility of combined 3D analysis of the foot and lower limb based on biplanar EOS radiographs and gait analysis. The study hypothesis was that this provides better understanding of abnormalities in form and function. Methods Ten children with unilateral EVCF and “very good” clinical results were included. They underwent gait analysis on the Rizzoli Institute multisegment foot model. Kinematic data were collected for the hip, knee, ankle and foot (hindfoot/midfoot, midfoot/forefoot and hindfoot/forefoot). Biplanar EOS radiographs were taken to determine anatomic landmarks and radiological parameters. Results Complete acquisition time was around 2 hours per patient. No significant differences were found between EVCF and healthy feet except for calcaneal incidence, tibiocalcaneal angle and hindfoot/midfoot and hindfoot/forefoot inversion. Discussion The feasibility of the combined analysis was confirmed. There were no differences in range of motion, moment or power between EVCF and healthy feet in this series of patients with very good results. The functional results are related to radiological results within the normal range. The protocol provided anatomic and kinematic reference data. A larger-scale study could more objectively assess the contribution of EOS radiography using optoelectronic markers. Level of evidence II, low-power prospective study. The authors thank Chaire ParisTech BiomecAM (personalized musculoskeletal modeling) for financial help.

International audience; Introduction While the clinical impact of coronal and sagittal alignment in adult spinal deformity (ASD) patients (pts) is established, there is a paucity of data in terms of axial plane deformity and potential association between muscle degenera- tion and 3D deformity. The purpose of this study was to analyze spinopelvic muscles characteristics in association with the 3D deformity of ASD patients. Methods This is a prospective cohort study; primary lumbar scoliosis patients (Cobb > 20°) were enrolled and sustained a low-dose X-rays with 3D spinal reconstructions and a fat/water separation MRI (from C7 to the knee). Volumetric 3D reconstructions and fat infiltration (FI) of 6 muscles groups were performed. Relationships between muscular data, radio- graphic parameters and health-related quality of life were investigated. Patients were stratified and compared based on the SRS classification, the odontoid-hip axis (ODHA) angle (> or < 6.1°) and occurrence of rotatory subluxation. Results and discussion Twenty-eight patients were enrolled with a mean age of 60 ± 16yo and mean body mass index of 26 ± 4 kg/m2 without differences between groups. There were a moderate sagittal malalignment and a Cobb angle of 45 ± 11° (table). Muscular volume was smaller in patients with more severe deformity (p > 0.05). Pts with ODHA > 6.1° or pelvic incidence minus lumbar lordosis > 10° had significantly higher FI for the 6 muscular groups, patients with pelvic tilt > 20° had significantly higher FI for erector spinae, hip flexors and extensors (p < 0.05). SF36-PCS significantly correlated with the muscular volume; SRS and Oswestry disability index correlated with the erectors spinae volume (p < 0.05). Conclusion This study analyzed for the first time the relationship between 3D radiographic parameters and muscular charac- teristics in ASD. Sagittal malalignment is associated with increased FI and decreased muscle volumes with poor outcomes.

Purpose The origin of the deformity due to adolescent idiopathic scoliosis (AIS) is not known, but mechanical instability of the spine could be involved in its progression. Spine slenderness (the ratio of vertebral height to transversal size) could facilitate this instability, thus playing a role in scoliosis progression. The purpose of this work was to investigate slenderness and wedging of vertebrae and intervertebral discs in AIS patients, relative to their curve topology and to the morphology of control subjects. Methods A total of 321 AIS patients (272 girls, 14 ± 2 years old, median Risser sign 3, Cobb angle 35° ± 18°) and 83 controls were retrospectively included (56 girls, median Risser 2, 14 ± 3 years). Standing biplanar radiography and 3D reconstruction of the spine were performed. Geometrical features were computed: spinal length, vertebral and disc sizes, slenderness ratio, frontal and sagittal wedging angles. Measurement reproducibility was evaluated. Results AIS girls before 11 years of age had slightly longer spines than controls (p = 0.04, Mann–Whitney test). AIS verte- brae were significantly more slender than controls at almost all levels, almost independently of topology. Frontal wedging of apical vertebrae was higher in AIS, as expected, but also lower junctional discs showed higher wedging than controls. Conclusion AIS patients showed more slender spines than the asymptomatic population. Analysis of wedging suggests that lower junctional discs and apex vertebra could be locations of mechanical instability. Numerical simulation and longitudinal clinical follow-up of patients could clarify the impact of wedging, slenderness and growth on the biomechanics of scoliosis progression. The authors are grateful to the BiomecAM chair programme on subject-specific musculoskeletal modelling (with the support of ParisTech and Yves Cotrel Foundations, Société Générale, Covea and Proteor) and to the DHU MAMUTH for funding.

Background Stand-to-sit pelvis kinematics is commonly considered as a rotation around the bicoxofemoral axis. However, abnormal kinematics could occur for patients with musculoskeletal disorders, affecting the hip-spine complex. The aim of this study is to perform a quantitative analysis of the stand-to-sit pelvis kinematics using 3D reconstruction from biplanar x-rays. Methods Thirty volunteers as a control group (C), 30 patients with hip pathology (Hip), and 30 patients with spine pathology (Spine) were evaluated. All subjects underwent standing and sitting full-body biplanar x-rays. Three-dimensional reconstruction was performed in each configuration and then translated such as the middle of the line joining the center of each acetabulum corresponds to the origin. Rigid registration quantified the finite helical axis (FHA) describing the transition between standing and sitting with two specific parameters. The orientation angle (OA) is the signed 3D angle between FHA and bicoxofemoral axis, and the rotation angle (RA) represents the signed angle around FHA. Results The mean OA was −1.8° for the C group, 0.3° for Hip group, and −2.4° for Spine group. There was no significant difference in mean OA between groups. However, variability was higher for the Spine group with a standard deviation (SD) of 15.9° compared with 10.8° in the C group and 12.3° in the Hip group. The mean RA in the C group was 18.1° (SD, 9.0°). There was significant difference in RA between the Hip and Spine groups (21.1° [SD, 8.0°] and 16.4° [SD, 10.8°], respectively) (P = .04). Conclusion Hip and spine pathologies affect stand-to-sit pelvic kinematics.

International audience; Purpose: To determine the short-term effect of bracing of adolescent idiopathic scoliotic (AIS) patients on the relationships between spinopelvic parameters related to balance, by comparing their in and out-of-brace geometry and versus healthy subjects. Methods: Forty-two AIS patients (Cobb angle 29° ± 12°, ranging from 16° to 61°) with a prescription of orthotic treatment were included retrospectively and prospectively. They all underwent biplanar radiography and 3D reconstruction of the spine and pelvis before bracing as well as less than 9 months after bracing. Eighty-three age-matched healthy adolescents were also included as control group and underwent biplanar radiography and 3D reconstruction. Results: Sacral slope was higher in AIS than healthy patients (p = 0.005). Bracing induced large changes of pelvic tilt (between - 9° and 9°), although patients' sagittal spinopelvic alignment tended to remain within the normality corridors defined by the healthy patients. Patients had flatter backs compared to healthy subjects and bracing further reduced their spinal curves. The head tended to remain above the pelvis in-brace. Conclusion: Analysis of sagittal alignment from head to pelvis showed that bracing further flattened the patients' backs and induced large compensating reorientations of the pelvis. Sagittal balance should be included in the planning and evaluation of brace treatment, since it could play a role in its outcome. These slides can be retrieved under Electronic Supplementary Material.

The effect of dominance on upper limb (UL) kinematics has only been studied on scapular movements. Moreover, when an anatomical UL movement is performed in a specific plane, secondary movements in the remaining planes involuntarily occur. These secondary movements have not been previously evaluated. The aim of this study was to compare the kinematics of primary and secondary angles of dominant and non-dominant UL during anatomical movements in asymptomatic adults. 25 asymptomatic adults performed 6 anatomical movements bilaterally: shoulder flexion-extension, abduction-adduction, horizontal abduction-adduction, internal-external rotation, elbow flexion-extension and wrist pronation-supination. Kinematics of the dominant and non-dominant UL were compared by their ranges of motion (ROM) and their angular waveforms (Coefficient of Multiple Correlations, CMC). The comparison between dominant and non-dominant UL kinematics showed different strategies of movement, most notably during elbow flexion-extension (CMC = 0.29): the dominant UL exhibited more pronation at maximal elbow flexion. Significant secondary angles were found on most of the UL anatomical movements; e.g. a secondary ROM of shoulder (humero-thoracic) external-internal rotation (69° ± 16°) was found when the subject intended to perform maximal shoulder abduction-adduction (119° ± 21°). Bias of dominance should be considered when comparing pathological limb to the controlateral one. Normative values of primary and secondary angles during anatomical movements could be used as a reference for future studies on UL of subjects with neurological or orthopedic pathologies. This study was sponsored by the research council of the University of Saint-Joseph under the Grant number IPHY5

Accurate calibration of the medio-lateral axis of the femur is crucial for clinical decision making based on gait analysis. This study proposes a protocol utilizing biplanar radiographs to provide a reference medio-lateral axis based on the anatomy of the femur. The biplanar radiographs allowed 3D modelling of the bones of the lower limbs and the markers used for motion capture, in the standing posture. A comprehensive analysis was performed and results from biplanar radiographs were reliable for 3D marker localization (±0.35 mm) and for 3D localization of the anatomical landmarks (±1 mm), leading to a precision of 1° for the orientation of the condylar axis of the femur and a 95% confidence interval of ±3° after registration with motion capture data. The anatomical condylar axis was compared to a conventional, marker-based, axis and three functional calibration techniques (axis transformation, geometric axis fit and DynaKAD). Results for the conventional method show an average difference with the condylar axis of 15° (SD: 6°). Results indicate DynaKAD functional axis was the closest to the anatomical condylar axis, mean: 1° (SD: 5°) when applied to passive knee flexion movement. However, the range of the results exceeded 15° for all methods. Hence, the use of biplanar radiographs, or an alternative imaging technique, may be required to locate the medio-lateral axis of the femur reliably prior to clinical decision making for femur derotational osteotomies.

Background Existing imaging techniques and single-parameter analyses, in nonfunctional positions, fail to detect the differences between patients with good vs poor results after total hip arthroplasty. Methods The present study developed an analysis method using the EOS full-body, low-dose, biplanar, weightbearing imaging system to compare good vs poor patients after total hip arthroplasty and to report on our preliminary experiences (17 good, 18 poor). Results All revision cases were found to have at least 4 high or low implant or anatomic parameters relative to the good group. These included acetabular cup orientation, sagittal pelvic tilt, sacral slope, femoral offset, and neck–shaft angle. Acetabular cup orientation differed significantly between groups. Conclusion With the EOS system, a large cohort can be studied relatively quickly and at low dose, which could lead to patient-specific guidelines.

Radiographs are often performed to assess pelvic and hip parameters, but results depend upon correct pelvis positioning. Three-dimensional (3D) reconstruction from biplanar-radiographs should provide parameters that are less sensitive to pelvic orientation, but this remained to be evaluated.Computerized-tomographic scans of six patients were used both as a reference and for generating simulated frontal and lateral radiographs. These simulated radiographs were generated while introducing axial rotations of the pelvis ranging from 0° to 20°. Simulated biplanar-radiographs were utilized by four operators, three times each, to perform pelvic 3D-reconstructions. These reconstructions were used to assess the trueness, precision and global uncertainty of radiological pelvic and hip parameters for each position.In the neutral position, global uncertainty ranged between ± 2° for pelvic tilt and ± 9° for acetabular posterior sector angle and was mainly related to precision errors (ranging from 1.5° to 7°). With increasing axial rotation, global uncertainty increased and ranged between ± 5° for pelvic tilt and ± 11° for pelvic incidence, sacral slope and acetabular anterior sector angle, mainly due to precision errors.Radiological parameters obtained from 3D-reconstructions, based on biplanar-radiographs, are less sensitive to axial rotation compared to plain radiographs. However, the axial rotation should nonetheless not exceed 10°.• Pelvic radiological parameters could be affected by patient malpositioning. • Biplanar radiograph-based 3D reconstructions were performed at increments of axial rotation. • Trueness, precision and global uncertainty were evaluated for pelvic and hip radiological parameters. • Hip parameters were less affected by rotation compared to pelvic parameters. • Maintaining the pelvis close to the neutral position is recommended to ensure the highest possible accuracy.

BACKGROUND CONTEXT Adults with spinal deformity (ASD) are known to have postural malalignment affecting their quality of life. Classical evaluation and follow-up are usually based on full-body static radiographs and health related quality of life questionnaires. Despite being an essential daily life activity, a formal gait assessment is lacking in clinical practice. Moreover, gait alterations and their relationships with static deformities in ASD still remain to be elucidated PURPOSE To investigate gait alterations in subjects with ASD and their static radiographic determinants. STUDY DESIGN/SETTING Prospective Single Center. PATIENT SAMPLE A total of 52 Adult Spinal Deformity patients, and 63 control subjects. OUTCOME MEASURES Full body image, 3D gait analysis. METHODS A total of 52 Adult Spinal Deformity (ASD) patients age and weight-matched to 63 controls underwent full-body 3D gait analysis with extraction of 3D lower limb and trunk kinematics. All subjects underwent full-body biplanar X-rays with subsequent calculation of classical 3D postural parameters. Kinematics and postural parameters were compared between ASD and controls. Determinants of gait alterations among postural radiographic parameters were explored in ASD. RESULTS ASD subjects had significantly increased SVA (33.6±59 vs -4.6±20mm), CAM-HA (12±62 vs -21±26mm), PT (19±13 vs 11±6°), frontal Cobb (25±21 vs 4±6°) compared to controls (all p CONCLUSIONS Subjects with ASD are known to have spinal malalignment causing postural alterations such as forward bending of the trunk; they also recruit compensatory mechanisms in the pelvis and lower limbs by increasing radiographic pelvic tilt and knee flexion. This study showed that static compensations in the lower limbs persist during gait where ASD patient exhibit a flexed attitude at the trunk, hips and knees, a reduced hip mobility and loss of lordosis. ASD walked at a slower pace with an elevation of their single support time that might contribute to their stability while walking. These dynamic discrepancies were shown to be strongly related to skeletal radiological alterations, specifically the increased forward shift of the head and trunk, the pelvic retroversion and the lack of lumbar lordosis. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.

Various spinal and rib cage parameters measured from complex examinations were found to be correlated with preoperative pulmonary function tests (PFT). The aim was to investigate the relationship between preoperative rib cage parameters and PFT using biplanar stereoradiography in patients with severe adolescent idiopathic scoliosis. Fifty-four patients, 45 girls and nine boys, aged 13.8 ± 1.2 years, with Lenke 1 or 2 thoracic scoliosis (> 50°) requiring surgical correction were prospectively included. All patients underwent preoperative PFT and low-dose biplanar X-rays. The following data were collected: forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), FEV1/FVC ratio, residual volume, slow vital capacity (SVC), total lung capacity (TLC), rib cage volume (RCV), maximum rib hump, maximum width, mean thoracic index, spinal penetration index, apical vertebral rotation, main curve Cobb angle (MCCA), T4–T12 kyphosis. The primary outcome was the relationship between rib cage parameters and PFT. The secondary outcome was the relationship between rib cage parameters and spine parameters. Data were analyzed using Spearman’s rank test. A multivariable regression analysis was performed to compare PFTs and structural parameters. Significance was set at α = 0.05. The mean MCCA was 68.7° ± 16.7°. RCV was highly correlated with all pulmonary capacities: TLC (r = 0.76, p

International audience; Adolescent idiopathic scoliosis (AIS) is a 3D deformity of the spine detectable by trunk asymmetry. As a decision-aid tool, a body scanner can help assessing non-invasively the external shape of the trunk. However if AIS is diagnosed, additional personalized 3D reconstructions of the spine, rib cage and pelvis are useful to have a complete understanding of the pathology and plan a treatment. For instance, both internal and external geometry are required for computer-aided brace design. Since a body scanner cannot currently guarantee reliable bones reconstruction, low-dose biplanar X-rays (BXR) is a relevant alternative. Indeed, it enables to compute reliable reconstructions of bony structures using validated methods daily used in clinical routine. More recently, 3D body shape reconstruction of asymptomatic subjects from BXR was also proved to be feasible with good accuracy. In order to prevent young scoliotic subjects from multiple examinations, we investigated the feasibility of trunk shape reconstruction from BXR. The proposed method relies on the 3D reconstructions of the spine, rib cage and pelvis priory performed and 10 radio-opaque markers placed on the subject on which we fit a statistical shape model (SSM). This model was built on a training set of 50 asymptomatic and 15 scoliotic female subjects for whom spine, rib cage, pelvis and trunk shape were reconstructed. The trunk shape reconstructions were assessed using the above mentioned method of body shape reconstruction. For the training on scoliotic subjects, this solution was corrected by 100 radio-opaque markers placed on the subjects and detected on the X-rays. During testing, after regression using the SSM, the trunk shape is registered automatically on apparent radiographic contours. Finally, few manual adjustments can be performed. This method has been evaluated on the 15 scoliotic subjects using a leave-one-out procedure. Signed marker-to-surface errors were computed on several trunk regions. The bias was everywhere lower than 0.7mm in absolute value and the standard deviation lower than 6mm. These results are promising and could be improved with a larger database. Thus, this study is a first step toward computer-aided brace design with a single examination.

The authors regret that one of the co-author's name was misspelt as "Brice Illharreborde, MD, PhD'" in the proof. However, the name should be spelled as "Brice Ilharreborde, MD, PhD'".

STUDY DESIGN: Retrospective observational study. OBJECTIVES: To determine the mass distribution along the scoliotic trunk using barycentremetry and its relationship with vertebral axial rotation and torque. SUMMARY OF THE BACKGROUND DATA: Deformity progression in adolescent idiopathic scoliosis (AIS) is not yet fully understood, but gravity load on the spine could play a role. Barycentremetry allows to characterize body mass distribution in standing position, which could provide a better understanding the mechanisms of progression. METHODS: 81 subjects (27 healthy adolescents and 53 AIS patients) underwent biplanar radiography and 3D reconstruction of the spine and body envelope. Position of the gravity line was estimated, as well as trunk segmental centers of mass COMs at each vertebral level and resulting axial torques to each vertebra. RESULTS: The COM of all trunk segments was less than 1 cm from the gravity line in the frontal plane for healthy subjects, and less than 1.5 cm for AIS patients. Vertebral axial torque was 0.7 ± 0.5 Nm in healthy subjects, 2.9 ± 2.1 Nm at the junctional vertebrae of AIS patients and 0.5 ± 0.5 at the apex. A strong association was found between high torque and high intervertebral rotation at junctions, with low torque and low intervertebral axial rotation at the apex. CONCLUSION: Results suggest that AIS patients can maintain the COM of each body segment close to their gravity line, irrespective of the severity and asymmetry of their deformity. Moreover, torque analysis shed some light on the importance of junctional vertebrae in the spinal deformity and, potentially, in the vicious cycle determining scoliosis progression. LEVEL OF EVIDENCE: Level III. The authors are grateful to the ParisTech BiomecAM chair program on subject-specific musculoskeletal model- ling (with the support of ParisTech and Yves Cotrel Foundations, Société Géenérale, Proteor, and Covea)

Manually annotating medical images with few landmarks to initialize 3D shape models is a common practice. For instance, when reconstructing the 3D spine from biplanar X-rays, the spinal midline, passing through vertebrae body centers (VBCs) and endplate midpoints, is required. This paper presents an automated spinal midline delineation method on frontal and sagittal views by using Mask R-CNN. The network detects all vertebrae from C7 to L5, followed by vertebrae segmentation and classification at the same time. After postprocessing to discard outliers, the vertebrae mask centers were regarded as VBCs to get the spine midline by polynomial fitting. Evaluation of the spinal midline on 136 images used root mean square error (RMSE) with respect to manual ground-truth. The RMSE ± standard error values of predicted spinal midlines (C7-L5) were 1.11 mm ± 0.67 mm on frontal views and 1.92mm ± 1.38 mm on sagittal views. The proposed method is capable of delineating spinal midlines on patients with different spine deformity degrees. BiomecAM chair program

The main function of the intervertebral disc is biomechanical function, since it must resist repetitive high loadings, while giving the spine its flexibility and protecting the spinal cord from over-straining. It partially owes its mechanical characteristics to the lamellar architecture of its outer layer, the annulus fibrosus. Today, no non-invasive means exist to characterize annulus lamellar structure in vivo. The aim of this work was to test the feasibility of imaging annulus fibrosus microstructure in vivo with ultrasonography. Twenty-nine healthy adolescents were included. Ultrasonographies of L3–L4 disc were acquired with a frontal approach. Annulus fibrosus was segmented in the images to measure the thickness of the lamellae. To validate lamellar appearance in ultrasonographies, multimodality images of two cow tail discs were compared: ultrasonography, magnetic resonance and optical microscopy. In vivo average lamellar thickness was 229.7 ± 91.5 μm, and it correlated with patient body mass index and age. Lamellar appearance in the three imaging modalities in vitro was consistent. Lamellar measurement uncertainty was 7%, with good agreement between two operators. Feasibility of ultrasonography for the analysis of lumbar annulus fibrosus structure was confirmed. Further work should aim at validating measurement reliability, and to assess the relevance of the method to characterize annulus alterations, for instance in disc degeneration or scoliosis. This study was funded by the ParisTech Biome- cAM chair programme on subject-specific musculoskeletal modelling (with the support of ParisTech and Yves Cotrel Foundations, Société Générale, and Covea), DHU MAMUTH, the Société Française de Chi- rurgie Orthopédique et Traumatologie (SOFCOT), the Devon Arthritis and Allied Reasearch Fund and the Henry Smith Charity. The authors are also grateful to Dr. Abdelmalek Benattayallah for his advice and technical support on MR imaging.

Purpose Measurement of vertebral axial rotation (VAR) is relevant for the assessment of scoliosis. Stokes method allows estimating VAR in frontal X-rays from the relative position of the pedicles and the vertebral body. This method requires identifying these landmarks for each vertebral level, which is time-consuming. In this work, a quasi-automated method for pedicle detection and VAR estimation was proposed. Method A total of 149 healthy and adolescent idiopathic scoliotic (AIS) subjects were included in this retrospective study. Their frontal X-rays were collected from multiple sites and manually annotated to identify the spinal midline and pedicle positions. Then, an automated pedicle detector was developed based on image analysis, machine learning and fast manual identification of a few landmarks. VARs were calculated using the Stokes method in a validation dataset of 11 healthy (age 6–33 years) and 46 AIS subjects (age 6–16 years, Cobb 10°–46°), both from detected pedicles and those manually annotated to compare them. Sensitivity of pedicle location to the manual inputs was quantified on 20 scoliotic subjects, using 10 perturbed versions of the manual inputs. Results Pedicles centers were localized with a precision of 84% and mean difference of 1.2 ± 1.2 mm, when comparing with manual identification. Comparison of VAR values between automated and manual pedicle localization yielded a signed difference of − 0.2 ± 3.4°. The uncertainty on pedicle location was smaller than 2 mm along each image axis. Conclusion The proposed method allowed calculating VAR values in frontal radiographs with minimal user intervention and robust quasi-automated pedicle localization. The authors are grateful to the ParisTech BiomecAM chair program on subject-specific musculoskeletal modeling for funding (with the support of ParisTech and Yves Cotrel Foundations, Société Générale, Proteor and Covea).

OBJECTIVEThe aim of this study was to determine if the apical vertebra (AV) in patients with adolescent idiopathic scoliosis (AIS) is the most rotated vertebra in the scoliotic segment.METHODSA total of 158 patients with AIS (Cobb angle range 20°–101°) underwent biplanar radiography with 3D reconstructions of the spine and calculation of vertebral axial rotations. The type of major curvature was recorded (thoracic, thoracolumbar, or lumbar), and both major and minor curvatures were included. The difference of levels (DL) between the level of maximal vertebral rotation (LMVR) and the AV was calculated as follows: DL = 0 if LMVR and AV were the same, DL = 1 if LMVR was directly above or below the AV, and DL = 2 if LMVR was separated by 1 vertebra or more from the AV. To investigate which factors explained the divergence of the LMVR from the AV, multinomial models were computed.RESULTSThe distribution of the DL was as follows: for major curvatures, 143 were DL = 0, 11 were DL = 1, and 4 were DL = 2; and for minor curvatures, 53 were DL = 0, 9 were DL = 1, and 31 were DL = 2. The determinants of a DL = 2 (compared with DL = 0) were lumbar curvature (compared with thoracic; adjusted OR 0.094, p = 0.001), major curvature (compared with minor; adjusted OR 0.116, p = 0.001), and curvatures with increasing apical vertebral rotation (adjusted OR 0.788, p < 0.001).CONCLUSIONSThis study showed that the AV is the most rotated vertebra in the majority of major curvatures, while in minor curvatures, the most rotated vertebra appears to be the junctional vertebra between major and minor curvatures in a significant proportion of cases.

Lower limb (LL) muscle morphology and growth are altered in children with cerebral palsy (CP). Muscle alterations differ with age and with severity of motor impairment, classified according to the gross motor classification system (GMFCS). Muscle alterations differ also with orthopedic intervention, frequently performed at the level of the shank muscles since an early age, such as the gastrocnemius. The aim was to investigate the alterations of treatment-naïve pelvis and thigh muscle lengths and volumes in children with GMFCS levels I and II, of varying ages. 17 children with CP (GMFCS I: N = 9, II: N = 8, age: 11.7 ± 4 years), age-matched to 17 typically developing (TD) children, underwent MRI of the LL. Three-dimensional reconstructions of the muscles were performed bilaterally. Muscle volumes and lengths were calculated in 3D and compared between groups. Linear regression between muscle volumes and age were computed. Adductor-brevis and gracilis lengths, as well as rectus-femoris volume, were decreased in GMFCS I compared to TD (p

International audience; Human pose estimation (HPE) methods based on convolutional neural networks (CNN) have demonstrated significant progress and achieved state-of-the-art results on human pose datasets. In this study, we aimed to assess the perfor-mance of CNN-based HPE methods for measuring anthropometric data. A Vicon motion analysis system as the reference system and a stereo vision system recorded ten asymptomatic subjects standing in front of the stereo vision system in a static posture. Eight HPE methods estimated the 2D poses which were transformed to the 3D poses by using the stereo vision system. Percentage of correct keypoints, 3D error, and absolute error of the body segment lengths are the evaluation measures which were used to assess the results. Percentage of correct keypoints – the stand-ard metric for 2D pose estimation – showed that the HPE methods could estimate the 2D body joints with a minimum accuracy of 99%. Meanwhile, the average 3D error and absolute error for the body segment lengths are 5 cm.