Thyroid Nodule Recognition Based on Feature Selection and Pixel Classification Methods Journal of Digital Imaging, 05/07/2012

Bibicu D et al. – It is found that the proposed approach is capable of identifying thyroid nodules with a correct classification rate of 83 % when whole image is analyzed and with a percent of 91 % when the region of interests (ROIs) are analyzed.

Methods

Twenty ultrasound images of normal thyroid and 20 that present thyroid nodules were used.
The analysis involves both the whole thyroid ultrasound images and the region of interests (ROIs).

Results

The proposed system and the classification results are validated using the receiver operating characteristics which give a better overall view of the classification performance of methods.

Read this article on Journal of Digital Imaging

Print Article | Back to Articles

Stay current on the latest literature, research and clinical news
Get special communications and offers from MDLinx and our sponsors
Receive invitations to paid market research

View Samples and Register

Stay current - Media Tool

Newsletter

RSS

Facebook

Receive free subspecialty
"5-minute updates" via email

Send the E-mail Newsletter to a Colleague

♦ Subscribe to our free RSS feeds:
Get the latest news in your specialty automatically added to your newsreader or your personal My Yahoo!, Google, My MSN or My AOL page. Learn More ♦

♦ Follow Us on Twitter
Twitter is a rich source of instantly updated information. Join today and follow @MDLinx to start receiving tweets. Learn More ♦

♦ Follow Us on Facebook

Search the latest
Medical Student Jobs
@MDLinx Career Center
View the full listing of physician jobs

Enter your email address and your password will be sent to you.
Email:

E-mail:
Password:
	Remember me Forgot your password?