Early cancer detection

How Cognition’s computer vision technology helps radiologists detect early-stage breast cancer


One of the top 5 global consulting firms with a significant market share in the hospital sector wanted to help make the work done by radiologists more efficient and minimize human errors. They hoped to do this by showing professionals areas of an x-ray image that require a more detailed analysis, along with the detected anomalies. The client was interested specifically in the area of breast cancer detection and in exploring the possibility of cancer prediction.


Cognition built a solution on the basis of their computer vision platform, Sigma Vision, to analyze x-ray images and enhance them to facilitate identification of important details such as calcifications, aiding cancer detection and prediction. The custom solution, tailored to the specific task of detecting and predicting breast cancer, was developed in collaboration with radiologists in one of the most prestigious hospitals in Spain. These radiologists selected and annotated x-ray images, and Cognition used them to adapt and train the AI solution.

Cognition also developed a tool to optimize the visualization of the x-ray images. The tool enhanced the images to facilitate easier identification of calcifications and other important details requested by radiologists.


The tool reduced the mammography analysis time by 50%, and has additionally proven useful in training new radiologists.