Bölgesel evrişimsel sinir ağları tabanlı MR görüntülerinde tümör tespiti
Yükleniyor...
Dosyalar
Tarih
2019
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Öz: Beyin tümörlerinden kaynaklı insan ölümleri günümüzde artmaktadır. Beyin tümörü çok hızlı büyüyerek, normal boyutunun iki katına çıkabilir. Bu yüzden uzmanlar, Manyetik Rezonans (MR) görüntülerini inceleme sürecini dikkatli ve hızlı bir şekilde yapmalıdır. Erken teşhis, kanser tanısında, tedavi planlamasında ve tedavi sonucunun değerlendirilmesinde hayati öneme sahiptir. Eğer beyin tümörü olan bir hasta doğru ve erken tedavi görmemişse, hastanın hayatta kalma şansı düşebilir ve ölümle sonuçlanabilir. Bu makalede, beyin MR görüntülerinden tümörü kolayca tespit eden ve tümörün yerini belirten, uzmanlara yardımcı olabilecek bilgisayar destekli otomatik tümör tespit sistemi geliştirilmiştir. Geliştirilen sistem derin öğrenme mimarilerinden olan Bölgesel tabanlı Evrişimsel Sinir Ağları (BESA) tabanlıdır. BESA, Evrişimsel Sinir Ağları (ESA) mimarisini kullanan bir yapı olmakla birlikte giriş görüntüsüne ek olarak ilgilenilen bölgenin de giriş olarak verildiği bir yapı olarak düşünülebilir. Farklı BESA mimarileri tasarlanarak Benchmark, Rembredant ve Harvard veri setleri üzerinde test edilmiştir. Elde edilen en yüksek doğruluk değeri %99,10 ile BESA4 mimarisi ve Benchmark veri setinden elde edilmiştir. En yüksek ortalama doğruluk ise yine BESA4 mimarisi ile %98,66 olarak hesaplanmıştır. Ayrıca, önerilen yöntemin başarımı, literatürdeki bazı yöntemler ile kıyaslanmıştır. Kıyaslamalar da önerilen yöntemin daha başarılı olduğu görülmüştür.
Öz: Human deaths caused by brain tumors are increasing nowadays. The brain tumor can grow very fast and can get twice of it is usual size. Therefore, physician have to analysis the Magnetic Resonance (MR) images quickly. This step is vital for the diagnosis of cancer, for treatment planning and for evaluation of the treatment outcome. If the patient who has the tumor in his brain is not treated correctly and quickly, the patient's chance of survival may decrease and resulted in death. In this article, we were developed a computer-assisted automated tumor detection system that can assist the physician in detecting and locating the tumor easily from brain MR images. The developed system is based on Regional based Convolutional Neural Networks (RCNN), which is one of the deep learning architectures. Besides RCNN is a structure that uses the architecture of the Convolutional Neural Networks (CNN), it can be considered as a structure in which the interested region is given as input in addition to the input images. In the proposed method, different RCNN architectures were designed and tested on Benchmark, Rembredant and Harvard datasets. The highest accuracy was obtained from the RCNN4 architecture on Benchmark data set is 99.10%. The highest average accuracy was calculated as 98.66% with RCNN4 architecture. Also, the success of the proposed method was compared with some of the methods exist in the literature. These comparisons showed that the proposed method is more successful and effective.
Öz: Human deaths caused by brain tumors are increasing nowadays. The brain tumor can grow very fast and can get twice of it is usual size. Therefore, physician have to analysis the Magnetic Resonance (MR) images quickly. This step is vital for the diagnosis of cancer, for treatment planning and for evaluation of the treatment outcome. If the patient who has the tumor in his brain is not treated correctly and quickly, the patient's chance of survival may decrease and resulted in death. In this article, we were developed a computer-assisted automated tumor detection system that can assist the physician in detecting and locating the tumor easily from brain MR images. The developed system is based on Regional based Convolutional Neural Networks (RCNN), which is one of the deep learning architectures. Besides RCNN is a structure that uses the architecture of the Convolutional Neural Networks (CNN), it can be considered as a structure in which the interested region is given as input in addition to the input images. In the proposed method, different RCNN architectures were designed and tested on Benchmark, Rembredant and Harvard datasets. The highest accuracy was obtained from the RCNN4 architecture on Benchmark data set is 99.10%. The highest average accuracy was calculated as 98.66% with RCNN4 architecture. Also, the success of the proposed method was compared with some of the methods exist in the literature. These comparisons showed that the proposed method is more successful and effective.
Açıklama
Anahtar Kelimeler
Kaynak
Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi
WoS Q Değeri
Q4
Scopus Q Değeri
Q2
Cilt
34
Sayı
3
Künye
ARI A,HANBAY D (2019). Bölgesel evrişimsel sinir ağları tabanlı MR görüntülerinde tümör tespiti. Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, 34(3), 1395 - 1408. Doi: 10.17341/gazimmfd.460535
