Aortic stenosis is the narrowing of the aortic valve due to structural damage. Common cause is the formation of calcification in the aortic valve, which occurs with age. An expert cardiologist generates an approximate score for the volume and severity of calcification by focusing on the aortic regions of interest in echocardiography images for the diagnosis of aortic stenosis. In this study, a novel dataset consisting of echocardiography images of the aortic valve was created. The regions of interest in the aortic valve were manually annotated in the training and validation datasets, and the images in the training dataset were used to train the four sub-versions of the YOLOv5 model. The validation dataset was used to evaluate the performance of the network during the training process, and the test dataset were used to evaluate the performance of the trained models. The highest mAP value of 99.5% was achieved with the YOLOv5-x model at an IoU threshold of 0.9. Additionally, the precision value was 99.9% and the recall value was 97.5%. The models demonstrated the ability to detect aortic valves very close to the expert cardiologist's accurately labeled ground-truth, even in aortic valves of different scales and orientations.