Our objective is to create an interactive image segmentation system of the abdominal area allowing quick volume segmentation requiring minimal intervention of the human operator. Our contribution to tackle this problem is to enhance an Active Learning image segmentation system with Expert Knowledge, allowing quick and accurate volume segmentation requiring minimal intervention of the human operator. As a first step, image segmentation is produced by a Random Forest (RF) classifier applied on a set of standard image features. The human operator is presented with the most uncertain unlabeled voxels to select some of them for inclusion in the training set, retraining the RF classifier. The approach is applied to the segmentation of the thrombus in CTA data of Abdominal Aortic Aneurysm (AAA) patients. The expert knowledge on the expected shape of the target structures is used to filter out undesired detections. We have performed computational experiments over 8 datasets between 216 and 560 slices each that consists in real human contrast-enhanced datasets of the abdominal area. The performance measure of the experiments is the true positive rate (TPR). 3-fold cross validation is applied. We average the TPR obtained on each slice at each iteration of the process, with a corresponding variance value. Surface rendering is computed to show a 3D visualization of the segmented thrombus. Accurate segmentation is obtained after a few iterations in areas where it is difficult to distinguish the anatomical structures from surrounding tissues due to a variety of noise conditions and similar the gray levels (i.e. thrombus).