We apply artificial intelligence to medical image segmentation process to generate 3D models that faithfully represent the patient’s anatomy.
How does our artificial intelligence develop?
For the creation of 3D models, advanced medical image segmentation systems are required. Combining algorithms, radiomics and artificial intelligence, we can have technologies that are revolutionizing the interpretation of medical images, obtaining a high degree of accuracy in anatomical reconstruction. These high-tech processes are supervised by radiologists.
Deep learning is the branch of AI that explores the use of artificial neural networks, a form of algorithm inspired by the structure and function of the human brain. At Cella, we develop artificial neural networks made up of different layers of interconnected neurons that learn to perform tasks related to segmentation autonomously.
Convolutional Neural Networks
Convolutional Neural Networks (CNN) are used to solve different tasks within the Computer Vision domain, enabling a semantic segmentation of anatomical elements, which represents a spectacular advance in radiology. The objective of this task is to label each pixel of an image with the corresponding class of what it is representing.
What are the benefits of using artificial intelligence in segmentation?
- It enables a semantic segmentation of the anatomical elements.
- Get to relate each pixel of an image with the class it is representing.