The objective of this project is to develop and automatic Image segmentation algorithm to detect the low grade brain tumors. Recent studies have discovered that low grade gliomas are associated with the genomic subtypes that have a particular feature shape and by analyzing the feature shape we can predict which genomic subtype that was responsible for the development of the brain tumor. The conventional way to find out the genomic subtype is by making a biopsy which involves making and incision into the skull and collecting the tissue from the tumor site which involves a huge amount of risk factors. For more details we encourage you to read the referenced paper below. we have implemented Three models "U-net", "MultiRes Unet" and a model proposed by us that we have named it as "Proposed Unet Model".
Contributions:125 commits in 3 days
image-segmentationsubtype
Contributions:1 push in 3 years