Collaboration Opportunities

Diagnostic medical imaging is the largest contributor of artificial radiation exposure in the general population. The health system in North America follows the As Low As Reasonably Achievable (ALARA) principle with regard to radiation exposure; that is, to operate at levels as low as reasonably achievable. Unfortunately, the general tendency is to expose the patients to high level of radiation to obtain good quality CT scan images. Therefore, there is a need for advanced technologies that can enhance CT scan images without exposing the patients to high level of radiation.

Mr. Ed Boraas at Concordia University of Edmonton has developed a machine learning based algorithm that transforms low quality images obtained at low radiation exposure to high quality images. The algorithm has the capability to learn from a patient’s images and thus can potentially reduce the number of scans required over time. This program has been trained using 1500 chest CT scans. Further studies to train the algorithm using a robust data set are planned.