In this use case, we show how we typically perform an initial project research for an Innovation service project for the use case of AI-based denoising for applications in cone beam CTs.
Using the example of the automatic segmentation of hip and femur from computed tomography, this blog post will show how the path to creating an AI model proceeds and what requirements must be met.
We have developed an algorithm that intelligently reduces the number of parameters of a neural network, while trying to maintain the highest possible accuracy.
Showcase for our Image Registration to compensate respiratory motion in a PET-CT image pair of a melanoma patient. The images were acquired at different times with individual scanners.
In this use case we realized a robust and accurate segmentation of the liver, the spleen, the kidneys and the heart in various Magnetic Resonance Imaging (MRI) sequences.
