Data-Driven Speed-of-Sound Reconstruction for Medical Ultrasound: Impacts of Training Data Format and Imperfections on Convergence

Jush, Farnaz Khun; Dueppenbecker, Peter Michael; Maier, Andreas

doi:10.1007/978-3-030-80432-9_11

Cited by 8 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Results on the human neck and calf muscles provided promising indications of SoS variations. Similarly, Jush et al explored a deep neural network for SoS reconstruction for US breast imaging using single plane wave US acquisition [31] and further extended to in-phase and quadrature data as the input [38]. Inspired by the feasibility of retrieving SoS distributions from US channel data with deep neural networks, a learningbased SoS correction method for PA imaging was proposed based on a dual-modal PA/US imaging system.…”

Section: Discussionmentioning

confidence: 99%

Spatiotemporal SVD for denoising in LED-based photoacoustic imaging

Shi

Vercauteren

Xia

2023

Photons Plus Ultrasound: Imaging and Sensing 2023

View full text Add to dashboard Cite

Photoacoustic (PA) image reconstruction involves acoustic inversion that necessitates the specification of the speed of sound (SoS) within the medium of propagation. Due to the lack of information on the spatial distribution of the SoS within heterogeneous soft tissue, a homogeneous SoS distribution (such as 1540 m/s) is typically assumed in PA image reconstruction, similar to that of ultrasound (US) imaging. Failure to compensate the SoS variations leads to aberration artefacts, deteriorating the image quality. In this work, we developed a deep learning framework for SoS reconstruction and subsequent aberration correction in a dual-modal PA/US imaging system sharing a clinical US probe. As the PA and US data were inherently coregistered, the reconstructed SoS distribution from US channel data using deep neural networks was utilised for accurate PA image reconstruction. On a numerical and a tissue-mimicking phantom, this framework was able to significantly suppress US aberration artefacts, with the structural similarity index measure (SSIM) of up to 0.8109 and 0.8128 as compared to the conventional approach (0.6096 and 0.5985, respectively). The networks, trained only on simulated US data, also demonstrated a good generalisation ability on data from ex vivo tissues and the wrist and fingers of healthy human volunteers, and thus could be valuable in various in vivo applications to enhance PA image reconstruction.

show abstract

Section: Discussionmentioning

confidence: 99%

Spatiotemporal SVD for denoising in LED-based photoacoustic imaging

Shi

Vercauteren

Xia

2023

Photons Plus Ultrasound: Imaging and Sensing 2023

View full text Add to dashboard Cite

show abstract

“…Phase Noise Studies showed that phase distortions can a↵ect SoS predictions (Stähli et al, 2020a;Khun Jush et al, 2021). Therefore, we applied uncorrelated phase distortions by shifting the phase of each channel by a random value in the range [ 0.7, 0.7] radian.…”

Section: Number Of Reflective Specklesmentioning

confidence: 99%

Deep Learning for Ultrasound Speed-of-Sound Reconstruction: Impacts of Training Data Diversity on Stability and Robustness

Jush

Biele

Dueppenbecker

et al. 2023

Melba

View full text Add to dashboard Cite

Ultrasound b-mode imaging is a qualitative approach and diagnostic quality strongly depends on operators’ training and experience. Quantitative approaches can provide information about tissue properties; therefore, can be used for identifying various tissue types, e.g., speed-of-sound in the tissue can be used as a biomarker for tissue malignancy, especially in breast imaging. Recent studies showed the possibility of speed-of-sound reconstruction using deep neural networks that are fully trained on simulated data. However, because of the ever-present domain shift between simulated and measured data, the stability and performance of these models in real setups are still under debate. In prior works, for training data generation, tissue structures were modeled as simplified geometrical structures which does not reflect the complexity of the real tissues. In this study, we proposed a new simulation setup for training data generation based on Tomosynthesis images. We combined our approach with the simplified geometrical model and investigated the impacts of training data diversity on the stability and robustness of an existing network architecture. We studied the sensitivity of the trained network to different simulation parameters, e.g., echogenicity, number of scatterers, noise, and geometry. We showed that the network trained with the joint set of data is more stable on out-of-domain simulated data as well as measured phantom data.

show abstract

“…For SoS reconstruction, acquiring sufficient measured data alongside its corresponding GT for training a deep neural network is challenging because there is no known gold-standard method capable of creating exact GT for reflection data (pulse-echo ultrasound) and there are only a few phantoms available with known heterogeneous SoS. Therefore, for deep learning-based approaches using simulated data for training is a common practice [7,10,11,12,13,14,15,16]. K-Wave toolbox [26] (Version 1.3) is used for the simulation of training data.…”

Section: Datasetmentioning

confidence: 99%

“…Similar techniques are employed for SoS reconstruction from ultrasound echo data. [10,11,12,13,14,15,16] investigated encoder-decoder networks with multiple or single steering angles for SoS reconstruction. In these studies, an encoder-decoder network takes sensor domain data as input and directly reconstructs the SoS map in the output.…”

Section: Introductionmentioning

confidence: 99%

AutoSpeed: A Linked Autoencoder Approach for Pulse-Echo Speed-of-Sound Imaging for Medical Ultrasound

Jush¹,

Biele²,

Dueppenbecker³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Quantitative ultrasound, e.g., speed-of-sound (SoS) in tissues, provides information about tissue properties that have diagnostic value. Recent studies showed the possibility of extracting SoS information from pulse-echo ultrasound raw data (a.k.a. RF data) using deep neural networks that are fully trained on simulated data. These methods take sensor domain data, i.e., RF data, as input and train a network in an end-to-end fashion to learn the implicit mapping between the RF data domain and SoS domain. However, such networks are prone to overfitting to simulated data which results in poor performance and instability when tested on measured data. We propose a novel method for SoS mapping employing learned representations from two linked autoencoders. We test our approach on simulated and measured data acquired from human breast mimicking phantoms. We show that SoS mapping is possible using linked autoencoders. The proposed method has a Mean Absolute Percentage Error (MAPE) of 2.39% on the simulated data. On the measured data, the predictions of the proposed method are close to the expected values with MAPE of 1.1 %. Compared to an end-to-end trained network, the proposed method shows higher stability and reproducibility.

show abstract

Data-Driven Speed-of-Sound Reconstruction for Medical Ultrasound: Impacts of Training Data Format and Imperfections on Convergence

Cited by 8 publications

References 19 publications

Spatiotemporal SVD for denoising in LED-based photoacoustic imaging

Spatiotemporal SVD for denoising in LED-based photoacoustic imaging

Deep Learning for Ultrasound Speed-of-Sound Reconstruction: Impacts of Training Data Diversity on Stability and Robustness

AutoSpeed: A Linked Autoencoder Approach for Pulse-Echo Speed-of-Sound Imaging for Medical Ultrasound

Contact Info

Product

Resources

About