Scoliosis curve type classification from 3D trunk image

Adolescent idiopathic scoliosis (AIS) is a deformity of the spine manifested by asymmetry and deformities of the external surface of the trunk. Classification of scoliosis deformities according to curve type is used to plan management of scoliosis patients. Currently, scoliosis curve type is determined based on X-ray exam. However, cumulative exposure to X-rays radiation significantly increases the risk for certain cancer. In this paper, we propose a robust system that can classify the scoliosis curve type from non invasive acquisition of 3D trunk surface of the patients. The 3D image of the trunk is divided into patches and local geometric descriptors characterizing the surface of the back are computed from each patch and forming the features. We perform the reduction of the dimensionality by using Principal Component Analysis and 53 components were retained. In this work a multi-class classifier is built with Least-squares support vector machine (LS-SVM) which is a kernel classifier. For this study, a new kernel was designed in order to achieve a robust classifier in comparison with polynomial and Gaussian kernel. The proposed system was validated using data of 103 patients with different scoliosis curve types diagnosed and classified by an orthopedic surgeon from the X-ray images. The average rate of successful classification was 93.3% with a better rate of prediction for the major thoracic and lumbar/thoracolumbar types.

Non invasive classification system of scoliosis curve types using least-squares support vector machines

Objective To determine scoliosis curve types using non invasive surface acquisition, without prior knowledge from X-ray data. Methods Classification of scoliosis deformities according to curve type is used in the clinical management of scoliotic patients. In this work, we propose a robust system that can determine the scoliosis curve type from non invasive acquisition of the 3D back surface of the patients. The 3D image of the surface of the trunk is divided into patches and local geometric descriptors characterizing the back surface are computed from each patch and constitute the features. We reduce the dimensionality by using principal component analysis and retain 53 components using an overlap criterion combined with the total variance in the observed variables. In this work, a multi-class classifier is built with least-squares support vector machines (LS-SVM). The original LS-SVM formulation was modified by weighting the positive and negative samples differently and a new kernel was designed in order to achieve a robust classifier. The proposed system is validated using data from 165 patients with different scoliosis curve types. The results of our non invasive classification were compared with those obtained by an expert using X-ray images. Results The average rate of successful classification was computed using a leave-one-out cross-validation procedure. The overall accuracy of the system was 95%. As for the correct classification rates per class, we obtained 96%, 84% and 97% for the thoracic, double major and lumbar/thoracolumbar curve types, respectively. Conclusion This study shows that it is possible to find a relationship between the internal deformity and the back surface deformity in scoliosis with machine learning methods. The proposed system uses non invasive surface acquisition, which is safe for the patient as it involves no radiation. Also, the design of a specific kernel improved classification performance.

Validity of a Quantitative Clinical Measurement Tool of Trunk Posture in Idiopathic Scoliosis

STUDY DESIGN: Concurrent validity between postural indices obtained from digital photographs (two-dimensional [2D]), surface topography imaging (three-dimensional [3D]), and radiographs. OBJECTIVE: To assess the validity of a quantitative clinical postural assessment tool of the trunk based on photographs (2D) as compared to a surface topography system (3D) as well as indices calculated from radiographs. SUMMARY OF BACKGROUND DATA: To monitor progression of scoliosis or change in posture over time in young persons with idiopathic scoliosis (IS), noninvasive and nonionizing methods are recommended. In a clinical setting, posture can be quite easily assessed by calculating key postural indices from photographs. METHODS: Quantitative postural indices of 70 subjects aged 10 to 20 years old with IS (Cobb angle, 15 degrees -60 degrees) were measured from photographs and from 3D trunk surface images taken in the standing position. Shoulder, scapula, trunk list, pelvis, scoliosis, and waist angles indices were calculated with specially designed software. Frontal and sagittal Cobb angles and trunk list were also calculated on radiographs. The Pearson correlation coefficients (r) was used to estimate concurrent validity of the 2D clinical postural tool of the trunk with indices extracted from the 3D system and with those obtained from radiographs. RESULTS: The correlation between 2D and 3D indices was good to excellent for shoulder, pelvis, trunk list, and thoracic scoliosis (0.81>r<0.97; P<0.01) but fair to moderate for thoracic kyphosis, lumbar lordosis, and thoracolumbar or lumbar scoliosis (0.30>r<0.56; P<0.05). The correlation between 2D and radiograph spinal indices was fair to good (-0.33 to -0.80 with Cobb angles and 0.76 for trunk list; P<0.05). CONCLUSION: This tool will facilitate clinical practice by monitoring trunk posture among persons with IS. Further, it may contribute to a reduction in the use of radiographs to monitor scoliosis progression.

Suivi longitudinal de la densité osseuse et du statut en vitamine D chez des enfants prépubères avec une allergie au lait de vache non résolue.

Problématique : L'allergie au lait de vache (ALV) est reconnue comme une condition transitoire qui disparaît chez la majorité des enfants avant l’âge de 3-5 ans, mais des données récentes révèlent une persistance de l’ALV. Les enfants souffrant d’une ALV sont à risque d’apports insuffisants en calcium et en vitamine D, deux nutriments impliqués dans la santé osseuse. Une première étude transversale portant sur la santé osseuse d’enfants prépubères ALV a observé que la densité osseuse (DMO) lombaire était significativement inférieure à celle d’enfants sans allergie au lait de vache (SALV). Objectifs : Sur la base de ces résultats, nous désirons documenter l’évolution longitudinale de la santé osseuse, du statut en vitamine D, des apports en calcium et en vitamine D et de l’adhérence à la supplémentation des enfants ALV (n=36) et de comparer ces données aux enfants SALV (n=19). Résultats : Le gain annualisé de la DMO lombaire est similaire entre les enfants ALV et SALV. Bien qu’il n’y ait pas de différence significative entre les deux groupes, la DMO lombaire des enfants ALV demeure cependant inférieure à celle des témoins. Qui plus est, le score-Z de la DMO du corps entier tend à être inférieur chez les enfants-cas comparé aux témoins. Au suivi, la concentration de 25OHD et le taux d’insuffisance en vitamine D sont similaires entre les deux groupes tout comme les apports en calcium et en vitamine D. Davantage d’enfants ALV prennent un supplément de calcium au suivi comparativement au temps initial (42% vs. 49%, p<0,05), mais le taux d’adhérence à la supplémentation a diminué à 4 jours/semaine. Conclusion : Une évaluation plus précoce ainsi qu’une prise en charge de la santé osseuse des enfants ALV pourraient être indiquées afin de modifier l’évolution naturelle de leur santé osseuse. Les résultats justifient aussi le suivi étroit des apports en calcium et vitamine D par une nutritionniste et la nécessité d'intégrer la supplémentation dans le plan de traitement de ces enfants et d’assurer une surveillance de l’adhérence à la supplémentation.