Atherosclerosis is a cardiovascular disease very widespread into population. It is characterized by a thickening of the arterial walls, which affects blood flow. The intima-media thickness (IMT) has emerged as a reliable early indicator of this pathology. Nowadays, the IMT is still measured by the physician using images acquired with a B-scan ultrasound. This can lead to several problems such as the variability between observers, or the obtainment of a minimum few points along the entire length of the arterial walls. Image segmentation can detect the IMT throughout the artery length as well as statistics such as the maximum, the minimum or the average IMT. The segmentation method proposed in this work is based on active contours, consisting of two parametric curves that are adjusted adaptively to the existing edges in the image. It allows getting measurements similar to those of a specialist, but with minimal user interaction.