Microvascular blood flow in the thyroid: Preliminary results with a novel imaging technique

• Published in: Proceedings - IEEE Ultrasonics Symposium pp. 2237 - 2240
• Main Authors: Forsberg, Flemming, Machado, Priscilla, Segal, Sharon, Okamura, Yoko, Guenette, Gilles, Rapp, Cynthia, Lyshchik, Andrej
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2014
• Subjects: Cancer; Colored noise; Doppler effect; Flow imaging; Gray-scale; Image color analysis; image processing; Imaging; thyroid; Ultrasonic imaging
• ISSN: 1051-0117
• DOI: 10.1109/ULTSYM.2014.0557

Nodules in the thyroid are present in 13-76 % of ultrasound (US) imaging evaluations; although only 4-15 % are malignant. A better understanding of thyroid nodule vascularity might be clinically helpful and the purpose of this study was thus to determine the flow imaging capabilities of a new prototype US image processing technique (SMI; Toshiba Medical Systems, Tokyo, Japan) for the depiction of microvascular flow in normal thyroid tissue and thyroid nodules compared to standard color and power Doppler imaging (CDI and PDI). SMI is a novel, microvascular flow imaging mode implemented on the Aplio 500 scanner (Toshiba). By analyzing clutter motion and using a new adaptive algorithm to identify and remove tissue motion SMI is designed to improve the visualization of microvascular blood flow. SMI depicts this information as a color overlay or as a monochrome map of flow. Ten healthy volunteers and 22 patients, with 25 thyroid nodules, were studied. Subjects underwent a thyroid US examination consisting of grayscale US, CDI and PDI followed by color and monochrome SMI. In the volunteers, pulsed Doppler guided by the 4 flow modes determined the lowest velocity measurable within the normal thyroid. For the patient data, 2 radiologists independently scored overall flow detection, vessel branching and noise on a visual analog scale of 1 (worst) to 10 (best). In the volunteers color and monochrome SMI captured microvasculature with lower velocities than CDI and PDI (2.2 ± 0.35 and 2.1 ± 0.32 cm/s vs. 2.6 ± 0.44 and 2.8 ± 0.77 cm/s; p <; 0.012). For all 25 nodules both readers found that color and monochrome SMI showed more microvascular flow and provided better depiction of the vessel branching compared to CDI and PDI (p <; 0.0001). Clutter noise was significantly higher in the monochrome SMI mode than in the other 3 modes (p <; 0.001). Consequently, initial results indicate that SMI can depict more detailed peri- and intra-nodular thyroid microvascular flow than CDI and PDI.