Scoliosis is a congenital disorder that causes the spine to be deformed from its normal shape, it greatly influences the patient’s health and quality of life, therefore, optimal solutions for scoliosis patient’s follow-up are needed. The most common method for diagnosing scoliosis in humans is two-dimensional X-ray imaging. However, better exploration and analysis of the scoliosis spine may be possible with 3-D visualization. Moreover, it is much better to be able to manipulate or interact with the scoliosis spine when it is represented in three dimensions. Applications of Augmented Reality (AR) technology have recently showed promising results in supporting medical decisions, diagnosis, and therapy. In this paper, we propose a conceptual augmented reality framework for spinal disorders 3-D representation based on 2-D X-ray images. Accurate scoliosis measurement in spinal radiographs is critical for clinical assessment, diagnosis, and treatment planning. Conventional manual methods are time-consuming and subject to inter-observer variability. Here, we propose an automatic scoliosis measurement method, tested on X-ray images from Bounaama Djilali Hospital (CHU Douera), Algeria. Results show that 3-D reconstruction from 2-D landmarks is feasible and that Cobb angle estimation validates the effectiveness of the approach.