Event Title

Prognosis of Glioblastoma using MRI data

Session Number

Project ID: MEDH 08

Advisor(s)

Dr. Jane Wu; Northwestern University Feinberg School of Medicine

Ryan Spear; Northwestern University Feinberg School of Medicine

Discipline

Medical and Health Sciences

Start Date

22-4-2020 8:30 AM

End Date

22-4-2020 8:45 AM

Abstract

Glioblastoma is a highly invasive malignant tumor caused by cancerous astrocytes. Due to the irregular, amorphous form, it is often difficult to identify and distinguish between the tumor core and surrounding edema in MRI imaging, making it difficult for doctors to accurately prognose the condition. This project aims to identify survival using machine learning techniques between the size and location of the tumor and the age of the patient to more accurately prognose a patient with glioblastoma. In order to do so, a segmentation program would segment the images of the tumor from the open source database BraTs 2019 provided by the University of Pennsylvania. Using the images to train the model as well as the provided information about the patients survival and age will be used as factors for a support vector machine to be able to prognose patients into short and long term survival. Data is yet to be collected.

Share

COinS
 
Apr 22nd, 8:30 AM Apr 22nd, 8:45 AM

Prognosis of Glioblastoma using MRI data

Glioblastoma is a highly invasive malignant tumor caused by cancerous astrocytes. Due to the irregular, amorphous form, it is often difficult to identify and distinguish between the tumor core and surrounding edema in MRI imaging, making it difficult for doctors to accurately prognose the condition. This project aims to identify survival using machine learning techniques between the size and location of the tumor and the age of the patient to more accurately prognose a patient with glioblastoma. In order to do so, a segmentation program would segment the images of the tumor from the open source database BraTs 2019 provided by the University of Pennsylvania. Using the images to train the model as well as the provided information about the patients survival and age will be used as factors for a support vector machine to be able to prognose patients into short and long term survival. Data is yet to be collected.