3D Complex-valued Convolutional Network for Fast Ultrasound Diverging Wave Imaging - Reconnaissance des Formes, Image, Apprentissage et Perception 2024 Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

3D Complex-valued Convolutional Network for Fast Ultrasound Diverging Wave Imaging

Résumé

Diverging wave acquisition has become the standard technique for fast ultrasound imaging due to its high temporal resolution. However, this approach is limited by the number of transmitted waves. Recently, deep learning has taken place in a variety of studies to improve the quality of ultrasound imaging. Most of these approaches have been performed on Bmode images, radio frequency signals, and in-phase/quadrature posterior to the delay-and-sum beamformer. In this work, we employ a 3D complex convolutional neural network to reconstruct an enhanced ultrasound image from fewer diverging waves prior to the sum in delay-and-sum beamformer. The network was trained using simulated data and evaluated using in-vitro phantom data. We provide experimental evidence that our approach produces a high contrast and quality image from 3 steered acquisitions that is comparable to those obtained from diverging wave compounding.
Fichier principal
Vignette du fichier
3D Complex-valued Convolutional Network for Fast Ultrasound Diverging Wave Imaging.pdf (2.33 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04617986 , version 1 (20-06-2024)

Identifiants

  • HAL Id : hal-04617986 , version 1

Citer

Ahmed Bentaleb, Christophe Sintes, Chafiaa Hamitouche. 3D Complex-valued Convolutional Network for Fast Ultrasound Diverging Wave Imaging. RFIAP 2024, Jul 2024, Lille, France. ⟨hal-04617986⟩
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More