Computerized intracranial tumor detection using morphological operations on MRI


Dundar M. S., Gumus K., İçer S., Ciraci S.

European Biotechnology Conference, Letonya, 5 - 07 Mayıs 2016, cilt.231, (Özet Bildiri)

  • Yayın Türü: Bildiri / Özet Bildiri
  • Cilt numarası: 231
  • Doi Numarası: 10.1016/j.jbiotec.2016.05.300
  • Basıldığı Ülke: Letonya
  • Erciyes Üniversitesi Adresli: Evet

Özet

Magnetic Resonance (MR) is an important imaging instrument for intracranial tumor treatment because it provides ability to visualize brain tumors. It ensures to detect the location of a tumor and quantify its volume, thus aids the planning of a treatment. Tumor detection is often performed by a radiologist and tumor volume calculation is often done manually and may not be precise enough for surgical operations. It should be noted that because of varying signal intensities on different type of MR images, tumor detection by naked eye might be challenging.

This study presents an automatic method for brain tumor segmentation on MR images using morphological operations on MATLAB platform. We developed a script that is capable of automatically detecting and segmenting tumor masses and calculating their areas using morphological operations and enables visualization of tumor volume by coloring it.

In this study, T2-weighted MR images of two patients diagnosed with intracranial lesions were examined. The developed script was able to successfully detect the tumors and estimate their areas. The process and results will be discussed in poster presentation.