GE HealthCare has scored FDA clearance for its deep learning-based image processing software for PET/CT.

Precision DL improves contrast-to-noise ratio, contrast recovery, and quantitative accuracy, and is available on Omni Legend, GE HealthCare’s first all-digital PET/CT scanner, which has two times the sensitivity of previous digital scanners and is designed to enhance theranostic imaging.

Unveiled in October at the European Association of Nuclear Medicine annual meeting in Spain, the scanner can image high-count tracers for cardiac and neuro imaging, producing faster total scan times, with a greater ability to identify small lesions.
Precision DL, on average, improves contrast recovery by 11%; contrast-to-noise ratio by 23%; feature quantification accuracy by 14%; and small, low contrast lesion detectability by 42%.

Its addition enables Omni Legend to decode coincidence events at fine resolutions for diagnostic and treatment planning.

“Precision DL enhances image quality, enabling us to spot small lesions, including on images obtained with very low-dose injections and short bedtimes, to potentially start treatment and monitoring early, which might result in improved patient outcomes,” said Dr. Flavio Forrer, chairman of nuclear medicine in the division of radiology and nuclear medicine at Kantonsspital St. Gallen in Switzerland, in a statement.

It was developed using a deep neural network trained on thousands of images created with multiple reconstruction methods, including ToF reconstruction.

According to a study published in the European Journal of Nuclear Medicine and Molecular Imaging, deep learning models trained for ToF image enhancement improved feature quantitation, overall image sharpness, and overall diagnostic value, especially for lesion detection and diagnostic confidence, in PET/CT images reconstructed without ToF.

Other Recon DL applications include TrueFidelity for CT, Helix DL for X-ray, and AIR Recon DL for MR, which uses raw data to reconstruct images and was recently adopted by Jefferson Radiology in Connecticut, to speed up exams.

Health IT Homepage