AI-generated Pseudoplanars for Ventilation/Perfusion Scans

The objective of this project is to develop artificial intelligence (AI)-based solutions aimed at denoising ventilation-perfusion (V/Q) planar images for the detection of pulmonary embolism (PE). The denoising algorithm has the potential to solve two major limitations of V/Q scans: long acquisition times and currently insufficient techniques for generating pseudo-planar images from SPECT acquisitions. Our solution aims to reduce acquisition times up to a factor of 10 for planar acquisitions or to produce realistic and diagnostically equivalent pseudo-planar images from SPECT acquisitions. The academic partner, The Ottawa Hospital, is a world-renowned center in PE management with a state-of-the-art nuclear medicine department. The industry partner, Jubilant Radiopharma, develops, manufactures, and markets innovative diagnostic imaging radiopharmaceutical solutions, including those for V/Q scans. The AI solutions developed in this project will increase the intrinsic value of V/Q scans and hopefully stimulate their use over alternative unnecessary imaging studies with higher risks.

Faculty Supervisor:

Ran Klein

Student:

Partner:

Jubilant Radiopharma

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Ottawa

Program: