Liver segmentation is an important task in medical imaging because it helps to identify the location and size of the liver in CT scans and MRI scans, which is essential information for the diagnosis and treatment of various liver diseases. It plays a vital role in liver cancer diagnosis and treatment planning, as it allows physicians to define the exact location and extent of tumors, that are crucial factors in determining the best treatment options and assessment of treatment response.
This is an end-to-end tutorial, proving a guideline on steps of liver segmentation using a U-Net architecture. Training and testing are performed on a public database including 20 pre-annotated 3D CT series. Read Full Article
The preliminary results based on light training of the UNet model:
