Carotid wall segmentation in longitudinal ultrasound images using structured random forest

摘要: Abstract Edge detection is a primary image processing technique used for object detection, data extraction, and segmentation. Recently, edge-based segmentation using structured classifiers has been receiving increasing attention. The intima media thickness (IMT) of the common carotid artery mainly as primitive indicator development cardiovascular disease. For efficient measurement IMT, we propose fast edge-detection based on random forest classifier. accuracy IMT degraded owing to speckle noise found in ultrasound images. To address this issue, use state-of-the-art denoising method reduce noise, followed by an enhancement increase contrast. Furthermore, present novel approach automatic region interest extraction which pre-trained classifier algorithm applied quantifying IMT. proposed exhibits IMTmean ±  standard deviation 0.66mm ± 0.14, closer ground truth value 0.67mm ± 0.15 compared techniques.