Figure 3
From: Computer-aided assessment of regional vascularity of thyroid nodules for prediction of malignancy
Image analysis of peripheral, central and overall vascular index (VI) of the thyroid nodule as shown in Fig. 2. The primary region of interest (ROI), i.e. the thyroid nodule, was extracted from the outlined area. Using an offset level of 10%, the peripheral (a, left) and central (b, left) regions of the nodule were segmented and the total number of pixel within the segmented area was counted by the computer algorithm. The color pixels coded by the color Doppler ultrasound were extracted by eliminating the grey scale pixels, and the color pixels were counted by the algorithm (a, right and b, right). The VI of peripheral and central regions of the nodule was the percentage of the number of color pixels to the total number of pixels within the segmented area. The overall VI of the nodule was evaluated by counting the total number of pixel within the ROI (c, left) and the number of color pixel in the image with the grey scale pixels eliminated (c, right).
