Breast Ultrasound Automated ROI Segmentation with Region Growing

Lee, Lay-Khoon and Liew, Siau-Chuin (2015) Breast Ultrasound Automated ROI Segmentation with Region Growing. In: IEEE 4th International Conference on Software Engineering and Computer Systems, 19-21 August 2015 , Kuantan, Pahang, Malaysia. pp. 177-182.. ISBN 978-1-4673-6722-6

Breast ultrasound automated ROI segmentation with.pdf

Download (658kB) | Preview


Image segmentation is an important technology used in different areas ranging from image processing to image analysis. One of the simplest methods for image segmentation that is widely implemented in medical images is the region growing method. Current researches mostly focus on using the region growing method to automatically detect the presence of tumor in MRI (Magnetic Resonance) images instead of ultrasound images. In this paper, we present an algorithm to automatically detect tumors in ultrasound images. Inspired by SergeBeucher and Balasubramanian’s road segmentation algorithm, this paper will implement the road segmentation algorithm into medical image segmentation. Results show that, the road segmentation algorithm actually works on the segmentation of medical image. The dice coefficient was used to evaluate the accuracy of the algorithm, eventually getting a value of 0.988 ± 0.00147 as the mean and standard deviation. This value is significant, because the higher the DC value, the more accurate is the segmentation. Besides that, the DC value can use for future reference and comparison.

Item Type: Conference or Workshop Item (Lecture)
Uncontrolled Keywords: Breast ultrasound; Combination; Dice coefficient; Medical image; Segmentation; Tumor detection
Subjects: Q Science > Q Science (General)
Q Science > QA Mathematics > QA76 Computer software
Faculty/Division: Faculty of Computer System And Software Engineering
Depositing User: Dr. Liew Siau Chuin
Date Deposited: 21 Oct 2015 03:37
Last Modified: 21 Jul 2016 04:33
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item