• Dataset used: MRBrainS18
• 148 is used as validation

 
• Registered and Bias Field Correction was already done in the dataset
• Skull stripping was done only for T1 weighted MRI using DeepBrain library which creates a mask for skull removal.
• Furthermore, contrast of T1 weighted MRI was improved using Histogram Equalization technique

Regularized Biased Field Corrected MRI	Removed Skull	Histogram equalization

Cortical gray matter, White matter, Cerebrospinal fluid in the extracerebral space can be easily reduced by appling thresholding to T1- weighted MRI further a small U-Net was used to denoise the threshold image.

For rest of the classes training was done on a custom model inspired by Unet Architecture. The model has 3 encoders stacked together in bottleneck layer and then a single decoder. There are skip connections from encoder to decoder to enhance segmentation.

Dice coefficient is used as Loss function in final training though Jaccard distance and crossentropy were also tried.

• Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations (https://arxiv.org/abs/1707.03237)
• U-Net: Convolutional Networks for Biomedical Image Segmentation (https://arxiv.org/pdf/1505.04597.pdf)
• MRBrainS18 (https://mrbrains18.isi.uu.nl/)

This project was made as part of the Smart India hackathon 2018 - Software Edition, a 36 hour hackathon organised by Government of India. The problem statement was given by Department of Atomic Energy, India