Postural anomalies are a significant concern in sports medicine and physiotherapy, as they can contribute to musculoskeletal disorders and reduced athletic performance. In this paper introduces an Artificial Intelligence (AI)-based framework for automated detection of postural deviations using computer vision and Machine Learning (ML) techniques. The proposed approach integrates feature extraction from depth images with classification algorithms to evaluate spinal alignment and joint positioning. Comparative experiments were conducted using both conventional classifiers and Deep Learning (DL) models on a dataset of annotated postural images. The results demonstrate that the proposed model achieves superior accuracy and robustness compared to baseline methods, highlighting its potential for practical application in clinical and sports environments. Importantly, the system provides an objective, non-invasive, and rapid assessment tool that supports physiotherapists and sports practitioners in early anomaly detection and personalized intervention planning. This research contributes to the growing field of AI-driven healthcare by bridging the gap between clinical expertise and computational methods. Future work will focus on expanding the dataset with more diverse postural conditions and integrating the system into real-time monitoring applications.