One of the most important challenges in diagnosis of Adolescent Idiopathic Scoliosis (AIS) is to find a correlation between internal spinal deformity and external shape of the back and to create a model to predict internal curve from 3D scans of back surfaces. In this way, X-ray imaging and analysis could be potentially avoided, which lead to reducing cumulative dose to the patient and decreasing the risk for malignancy development later in life. We propose a regressive model that considers correlation between 3D coordinates of internal spinal alignment and 3D coordinates of external curve, and between internal parameters (e.g., Cobb angles, vertebrae rotations) and external back surface, and predicts from external curve coordinates of the internal spinal line. With a limited number of samples used to create it, this model is able to provide a good approximation of the shape of the spine, with a mean 3D displacement between predicted and real curve of around 10 mm. Quality of prediction could be anyway significantly improved by simultaneously acquiring 3D scans of the back surface and X-ray scans of the patient, and adding to the model external features like scapulae or rib cage.