Résumé : Study Design. A prospective study was performed on the assessment of both thoracic and lumbar spine sagittal profi les (from C7 to S1). Objective. To propose a new noninvasive method for measuring the spine curvatures in standing and lying prone positions and to analyze their relationship with various biometric characteristics. Summary of Background Data. Modifi cations of spine curvatures ( i.e. lordosis or kyphosis) are of importance in the development of spinal disorders. Studies have emphasized the development of new devices to measure the spine sagittal profi les using a noninvasive and low-cost method. To date, it has not been applied for analyzing both lumbar and thoracic alterations for various positioning. Methods. Seventy-fi ve healthy subjects (mean 22.6 ± 4.3 yr) were recruited to participate in this study. Thoracic and lumbar sagittal profi les were assessed in standing and lying prone positions using a 3D digitizer. In addition, several biometric data were collected including maximal trunk isometric strength for fl exion and extension movement. Statistical analysis consisted in data comparisons of spine profi les and a multivariate analysis including biometric features, to classify individuals considering low within- and high between-variability. Result. Kyphosis and lordosis angles decreased signifi cantly from standing to lying prone position by an average of 13.4 ° and 16.6 ° , respectively. Multivariate analysis showed a sample clustering of 3 homogenous subgroups. The fi rst group displayed larger lordosis and fl exibility, and had low data values for height, weight, and strength. The second group had lower values than the overall trend of the whole sample, whereas the third group had larger score values for the torques, height, weight, waist, body mass index, and kyphosis angle but a reduced fl exibility. Conclusion. The present results demonstrate a signifi cant effect of the positioning on both thoracic and lumbar spine sagittal profi les and highlight the use of cluster analysis to categorize subgroups after biometric characteristics including curvature measurement.