MRIQC- Automated analysis of weekly MRI Quality Control Images for ACR Accreditation

The aim of this project is to automatically detect the key features of MRI phantom images and use those features to determine the state of the clinical MRI equipment being used at the site . The challenge here is to develop a model that is robust and gives accurate results even in low contrast and noisy images. The model should first be able to localize the specific regions of interest in the series of images and then extract specific features which would be used determine the state of MRI equipment being used. The next part would be use relevant pre-processing , that should reduce the noise and enhance the specific features of the image . The final part would be use a segmentation algorithm that accurately extracts the key features of the image and which can then be used to determine the state of MRI equipment being used.