PerceptFlow: Real-Time Ultrafast Doppler Image Enhancement Using Deep Convolutional Neural Network and Perceptual Loss

• Published in: Ultrasound in medicine & biology Vol. 49; no. 1; pp. 225 - 236
• Main Authors: Blons, Matthieu, Deffieux, Thomas, Osmanski, Bruno-Félix, Tanter, Mickaël, Berthon, Béatrice
• Format: Journal Article
• Language: English
• Published: England Elsevier Inc 01.01.2023; Elsevier
• Subjects: Animals; Bioengineering; Computer Science; Convolutional neural network; Deep learning; Doppler imaging; Image enhancement; Image Enhancement - methods; Image Processing, Computer-Assisted - methods; Imaging; Life Sciences; Medical Imaging; Mice; Neural and Evolutionary Computing; Neural Networks, Computer; Normal Distribution; Perceptual loss; Radiology; Signal-To-Noise Ratio; Ultrasonography, Doppler - methods; Deep learning; Doppler imaging; Perceptual loss; Convolutional neural network; Image enhancement

Ultrafast ultrasound is an emerging imaging modality derived from standard medical ultrasound. It allows for a high spatial resolution of 100 μm and a temporal resolution in the millisecond range with techniques such as ultrafast Doppler imaging. Ultrafast Doppler imaging has become a priceless tool...