Lung nodule classification using artificial crawlers, directional texture and support vector machine
详细信息 查看全文
- 作者：Bruno Rodrigues Froz ; a ; bfroz63@gmail.com" class="auth_mail" title="E-mail the corresponding author ; Antonio Oseas de Carvalho Filhoa ; antoniooseas@gmail.com" class="auth_mail" title="E-mail the corresponding author ; Aristó ; fanes Corrê ; a Silvaa ; ari@dee.ufma.br" class="auth_mail" title="E-mail the corresponding author ; Anselmo Cardoso de Paivaa ; paiva@deinf.ufma.br" class="auth_mail" title="E-mail the corresponding author ; Rodolfo Acatauassú ; Nunesb ; rodolfoacatauassu@yahoo.com.br" class="auth_mail" title="E-mail the corresponding author ; Marcelo Gattassc ; mgattass@tecgraf.puc-rio.br" class="auth_mail" title="E-mail the corresponding author
- 关键词：Artificial life ; Artificial crawlers ; Rose diagram ; Lung nodule classification
- 刊名：Expert Systems with Applications
- 出版年：2017
- 出版时间：1 March 2017
- 年：2017
- 卷：69
- 期：Complete
- 页码：176-188
- 全文大小：1872 K
文摘
- class="listitem" id="list_celist0004">
- class="label">•
We develop a methodology to classify lung nodule and non-nodules.
- class="label">•
833 lung scans were extracted from LIDC-IRDI database.
- class="label">•
Three methods used was artificial crawlers, rose diagram and hybrid model.
- class="label">•
The artificial crawlers was adapted to work on 3D images.
- class="label">•
The best result was achieved from hybrid model, with accuracy of 94,3%.