Brain tumors, if detected early, can significantly increase the chances of successful treatment. However, manual interpretation of MRI images is time-consuming and prone to variability among radiologists. By applying Image Processing algorithms, this work aims to offer an additional tool to doctors for the tumor detection process. This project also provides a solution for the Brain Tumor Segmentation (BraTS) Challenge hosted by the medical decathlon. The methodology followed in this work is based on the procedure described in this work.📚 Main concepts applied in this work
- Noise reduction using median filtering
- Otsu Thresholding method
- Morphological operations, dilation, to refine the segmented region
Example
In these images it's possible to appreciate the result of Otsu thresholding and dilation application on original image, and compare the obtained result with the ground-truth.Frontal view
Sagittal view
Horizontal view
📄 License
This project is distributed under the Apache 2.0 License For more details, visit the Github repository. 
Google Scholar
