Merck is a premier research-intensive biopharmaceutical company. We use the power of leading-edge science to save and improve lives around the world.
The dicom header of CT images contains information such as image resolution, patient orientation, and identifiers of the series and individual images. This information is essential to successful reconstruction of the 2D dicom images into 3D volumes. In addition, the reconstructed 3D volumes require quality check to ensure proper alignment of anatomical directions. Specifically, the Discovery team will register the reconstructed 3D volumes to an atlas and calculate similarity metrics between the registered 3D volumes and the atlas. The similarity metrics should be better in the reconstructed volumes that are correctly aligned than the mis-aligned volumes. The team will be able to work on real-word CT data and create a registration-based algorithm to check the quality of 3D volume reconstruction.
