Tunica Media Localization in Intravascular Image with Shadow Artifact Constraint using Circular-like Estimating Techniques
Jiraporn Wongwarn, Suwanna Rasmequan · 2019
The advancement of medical technologies allows diagnosis processes more convenient and time saving. Ultrasound technique is being used in a number of medical diagnosis processes. Vascular diseases are one of the group of diseases that used ultrasound techniques to diagnose. These types of diseases may cause serious illnesses such as heart failure, coronary artery disease, and cardiomyopathy. Ultrasound technique often provides a low-quality image due to the safety precaution. Thus, Intravascular images produced by ultrasound techniques often have low contrast, high noises, and few shadows. These features of the low-quality image made it difficult to identify each layer of the Intra Vascular which is needed to segment the actual area of the plaque. In this paper, Circular-like estimating techniques is proposed to locate the Tunica Media on low-quality Intravascular image. The proposed method provides an interesting result with the JI of 87.45% for images with shadow constraint and 92.06% for nonshadow constraint images in dataset A. In addition, the proposed method achieved JI of 63.90% for shadow constraint images with various patterns in dataset B. These results implied that the proposed method can reasonably apply with both shadow and non-shadow artifacts while those earlier works have to attempt them with different methods.