Deep Learning Based Ultrasound Image Reconstruction Method: A Time Coherence Study

Perdios, Dimitris; Vonlanthen, Manuel; Martinez, Florian; Arditi, Marcel; Thiran, Jean‐Philippe

doi:10.1109/ultsym.2019.8925595

Cited by 7 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It should be noted that such modes heavily rely on the timecoherence between consecutive frames, and that current static image metrics used throughout this work cannot guarantee such a coherence. Yet, we already carried a preliminary study on the latter aspect with positive outcomes [61], and are currently evaluating it in greater detail, including all major image reconstruction improvements described in the present work. From the visual assessment of the in vivo carotid sequences, provided in video format as supplementary material, the proposed approach also seems to preserve the time-coherence of moving structures (smooth movement of speckle patterns).…”

Section: Application Perspectivesmentioning

confidence: 99%

CNN-Based Image Reconstruction Method for Ultrafast Ultrasound Imaging

Perdios,

Vonlanthen,

Martinez

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Ultrafast ultrasound (US) revolutionized biomedical imaging with its capability of acquiring full-view frames at over 1 kHz, unlocking breakthrough modalities such as shear-wave elastography and functional US neuroimaging. Yet, it suffers from strong diffraction artifacts, mainly caused by grating lobes, side lobes, or edge waves. Multiple acquisitions are typically required to obtain a sufficient image quality, at the cost of a reduced frame rate. To answer the increasing demand for high-quality imaging from single-shot acquisitions, we propose a two-step convolutional neural network (CNN)-based image reconstruction method, compatible with real-time imaging. A low-quality estimate is obtained by means of a backprojection-based operation, akin to conventional delay-and-sum beamforming, from which a highquality image is restored using a residual CNN with multi-scale and multi-channel filtering properties, trained specifically to remove the diffraction artifacts inherent to ultrafast US imaging. To account for both the high dynamic range and the radio frequency property of US images, we introduce the mean signed logarithmic absolute error (MSLAE) as training loss function. Experiments were conducted with a linear transducer array, in single plane wave (PW) imaging. Trainings were performed on a simulated dataset, crafted to contain a wide diversity of structures and echogenicities. Extensive numerical evaluations demonstrate that the proposed approach can reconstruct images from single PWs with a quality similar to that of gold-standard synthetic aperture imaging, on a dynamic range in excess of 60 dB. In vitro and in vivo experiments show that trainings performed on simulated data translate well to experimental settings.

show abstract

Section: Application Perspectivesmentioning

confidence: 99%

CNN-Based Image Reconstruction Method for Ultrafast Ultrasound Imaging

Perdios,

Vonlanthen,

Martinez

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, deep learning has entered the ultrasound imaging field, achieving great results in many applications [6]. Neural networks are being used to estimate images [7], using ultrasound data as input and trained in order to beamform an image as their prediction [8] [9]. Further work has been performed to obtain an image and a corresponding material segmentation simultaneously from single channel raw data [10].…”

Section: Introductionmentioning

confidence: 99%

Single Plane-Wave Imaging using Physics-Based Deep Learning

Pilikos,

de Korte,

van Leeuwen

et al. 2021

Preprint

View full text Add to dashboard Cite

In plane-wave imaging, multiple unfocused ultrasound waves are transmitted into a medium of interest from different angles and an image is formed from the recorded reflections. The number of plane waves used leads to a tradeoff between frame-rate and image quality, with single-planewave (SPW) imaging being the fastest possible modality with the worst image quality. Recently, deep learning methods have been proposed to improve ultrasound imaging. One approach is to use image-to-image networks that work on the formed image and another is to directly learn a mapping from data to an image. Both approaches utilize purely data-driven models and require deep, expressive network architectures, combined with large numbers of training samples to obtain good results.Here, we propose a data-to-image architecture that incorporates a wave-physics-based image formation algorithm in-between deep convolutional neural networks. To achieve this, we implement the Fourier (FK) migration method as network layers and train the whole network end-to-end. We compare our proposed datato-image network with an image-to-image network in simulated data experiments, mimicking a medical ultrasound application. Experiments show that it is possible to obtain high-quality SPW images, almost similar to an image formed using 75 plane waves over an angular range of ±16 • . This illustrates the great potential of combining deep neural networks with physics-based image formation algorithms for SPW imaging.

show abstract

Fast ultrasonic imaging using end-to-end deep learning

Pilikos

Hörchens

Batenburg

et al. 2020

2020 IEEE International Ultrasonics Symposium (IUS)

View full text Add to dashboard Cite

Ultrasonic imaging algorithms used in many clinical and industrial applications consist of three steps: A data preprocessing, an image formation and an image post-processing step. For efficiency, image formation often relies on an approximation of the underlying wave physics. A prominent example is the Delay-And-Sum (DAS) algorithm used in reflectivity-based ultrasonic imaging. Recently, deep neural networks (DNNs) are being used for the data pre-processing and the image postprocessing steps separately. In this work, we propose a novel deep learning architecture that integrates all three steps to enable endto-end training. We examine turning the DAS image formation method into a network layer that connects data pre-processing layers with image post-processing layers that perform segmentation. We demonstrate that this integrated approach clearly outperforms sequential approaches that are trained separately. While network training and evaluation is performed only on simulated data, we also showcase the potential of our approach on real data from a non-destructive testing scenario.

show abstract

Deep Learning Based Ultrasound Image Reconstruction Method: A Time Coherence Study

Cited by 7 publications

References 8 publications

CNN-Based Image Reconstruction Method for Ultrafast Ultrasound Imaging

CNN-Based Image Reconstruction Method for Ultrafast Ultrasound Imaging

Single Plane-Wave Imaging using Physics-Based Deep Learning

Fast ultrasonic imaging using end-to-end deep learning

Contact Info

Product

Resources

About