FDA clears Canon’s CNN image reconstruction tech for CT

Canon Medical Systems USA has received FDA approval for its deep convolutional neural network (DCNN) image reconstruction technology for CT scanning.

According to a statement from the company, its advanced intelligent clear IQ engine (AiCE) uses a deep learning algorithm to distinguish signal from noise, thereby suppressing noise while enhancing signal.

Its algorithm learns from the image quality of model based iterative reconstruction (MBIR) to reconstruct CT scans three- to-five times quicker and with improved spatial resolution, according to the statement.

“Our AiCE technology utilizes a next generation approach to CT image reconstruction, further proving Canon Medical’s leadership and commitment to innovation in diagnostic imaging,” said Dominic Smith, senior director, CT, PET/CT, and MR business units, Canon Medical Systems USA, in the statement. “This technology doesn’t just meet our customers’ evolving needs, it exceeds them, opening doors to clearer, more precise images to help optimize patient care.”