Brain Tumour Diagnostics Support Based on Medical Image Segmentation

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60461373%3A22340%2F09%3A00022069" target="_blank" >RIV/60461373:22340/09:00022069 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Brain Tumour Diagnostics Support Based on Medical Image Segmentation
Original language description
High-grade gliomas represent rapidly growing malignant brain tumours. Early diagnostics of this decease and immediately applied treatment entails better life prognosis for the patient. The goal of this work is to initialize the development of an automated tumour recognition method based on computed tomography images processing. The resulting method is aimed at early glioma diagnostics support by distinguishing between the healthy tissue and the tumour tissue. The proposed technique involves, subsequently, image preprocessing, feature extraction, and classification of the extracted features using an artificial neural network. To obtain features, we compute the skewness of the discrete wavelet transform coefficients from selected rectangular image regions using three neighboring image slices to increase the number of samples while preserving good segmentation resolution within the image. The segmentation results are evaluated in cooperation with the neurologist.
Czech name
—
Czech description
—

Publication year
2009
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings 17th Annual Conference Technical Computing Prague 2009
ISBN
978-80-7080-733-0
ISSN
—
e-ISSN
—
Number of pages
7
Pages from-to
—
Publisher name
Humusoft
Place of publication
Praha
Event location
Praha
Event date
Nov 19, 2009
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
—

Similar results(10)