Segmentation and Registration of Kidneys from Contrast Enhanced Abdominal MR image

Akyar H., SELVER M. A., KALAYCI DEMİR G.

IEEE 16th Signal Processing and Communications Applications Conference, Aydın, Türkiye, 20 - 22 Nisan 2008, ss.569-572 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2008.4632678
  • Basıldığı Şehir: Aydın
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.569-572
  • Dokuz Eylül Üniversitesi Adresli: Evet

Özet

Subtraction of contrast enhanced magnetic resonance images acquired before and after the injection of a contrast agent is a common method used in clinical applications for identifying the tumor region. Because of the long duration of the acquisition process, patient movement and respiratory motion are induced and registration is required to obtain accurate subtraction. In this paper, a method for more accurate subtraction of kidney images from MR is presented Our starting point is the segmentation of kidneys from abdominal images to apply registration to only related region of interest. Segmentation of kidneys is automatically done with morphological operators based on the extraction of spine location. Then, Demon's deformable method for registration is applied to segmented kidneys by using an open source toolkit, called matitk, provided by insight toolkit.