Machine learning techniques for prostate ultrasound image diagnosis

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F10%3A86080854" target="_blank" >RIV/61989100:27240/10:86080854 - isvavai.cz</a>
Result on the web
<a href="http://dx.doi.org/10.1007/978-3-642-05177-7_19" target="_blank" >http://dx.doi.org/10.1007/978-3-642-05177-7_19</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-642-05177-7_19" target="_blank" >10.1007/978-3-642-05177-7_19</a>

Result language
angličtina
Original language name
Machine learning techniques for prostate ultrasound image diagnosis
Original language description
Estimation of prostate location and volume is essential in determining a dose plan for ultrasound-guided brachytherapy, a common prostate cancer treatment. However, manual segmentation is difficult, time consuming and prone to variability. In this chapter, we present a machine learning scheme, employing a combination of fuzzy sets, wavelets and rough sets, for analyzing prostrate ultrasound images in order diagnose prostate cancer. To address the image noise problem we first utilize an algorithm based on type-II fuzzy sets to enhance the contrast of the ultrasound image. This is followed by performing a modified fuzzy c-mean clustering algorithm in order to detect the boundary of the prostate pattern. Then, a wavelet features are extracted and normalized, followed by application of a rough set analysis for discrimination of different regions of interest to determine whether they represent cancer or not. The experimental results obtained, show that the overall classification accuracy of
Czech name
—
Czech description
—

Type
J<sub>x</sub> - Unclassified - Peer-reviewed scientific article (Jimp, Jsc and Jost)
CEP classification
IN - Informatics
OECD FORD branch
—

Publication year
2010
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)