Now a day’s liver cancer is one of the most prevalent diseases that may be extremely fatal if not properly diagnosed at early stage. The exponential growth of cells in liver is termed as liver cancer. Since liver cancer has a very complex histology and imaging modalities, it is difficult to diagnose it at early stage. In order to better diagnose and treat personalized patient- specific tumors, it is often necessary to segment and visualize them. The huge number of slices in the input image means a big challenge for physicians. The need to develop a segmentation technique that is automatic and efficient is a major requirement. It is necessary to develop a technique which would enable us to automate this approach which is reliable, afford- able, and accurate. In this project, we aspire to create a totally automatic method for liver tumor segmentation in CT images using a 2D convolutional deep neural network to beat all of the primary issues and we propose a pyramid- based UNET model and short skip connections for fast and precise segmentation of images which will be helpful for the early detection of tumor in a human’s body Conclusively, deep learning and image processing techniques are adopted to develop a liver CT scan segmentation.
LOGIN PAGE
SIGN UP PAGE
HOME PAGE
SAMPLE IMAGE
PREDICTION
INFO PAGE
REPORT PAGE
