Abstract: The present invention relates to a system 100 and a method 1100 for generating anatomical 3D models from ultrasound images. A first ML model of the system 100 determines anatomical features and corresponding dimensions from ultrasound images and generates a 3D model using shape priors based on nomogram data corresponding to gestational age. To enhance training and overcome data limitations of the first ML model, a second ML model generates synthetic ultrasound images using a generator-discriminator architecture. The system 100 automates anatomical assessments, ensuring accurate and consistent measurements. By integrating image segmentation, feature extraction, and synthetic data generation, the invention improves diagnostic precision and standardizes measurements.