skip to main content


Title: An improved training scheme for deep neural network ultrasound beamforming
Deep neural networks have been shown to be effective adaptive beamformers for ultrasound imaging. However, when training with traditional L p norm loss functions, model selection is difficult because lower loss values are not always associated with higher image quality. This ultimately limits the maximum achievable image quality with this approach and raises concerns about the optimization objective. In an effort to align the optimization objective with the image quality metrics of interest, we implemented a novel ultrasound-specific loss function based on the spatial lag-one coherence and signal-to-noise ratio of the delayed channel data in the short-time Fourier domain. We employed the R-Adam optimizer with look ahead and cyclical learning rate to make the training more robust to initialization and local minima, leading to better model performance and more reliable convergence. With our custom loss function and optimization scheme, we achieved higher contrast-to-noise-ratio, higher speckle signal-to-noise-ratio, and more accurate contrast ratio reconstruction than with previous deep learning and delay-and-sum beamforming approaches.  more » « less
Award ID(s):
1750994
PAR ID:
10138381
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
2019 IEEE International Ultrasonics Symposium (IUS)
Page Range / eLocation ID:
568 to 570
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. The generalized contrast-to-noise ratio (gCNR) is a relatively new image quality metric designed to assess the probability of lesion detectability in ultrasound images. Although gCNR was initially demonstrated with ultrasound images, the metric is theoretically applicable to multiple types of medical images. In this paper, the applicability of gCNR to photoacoustic images is investigated. The gCNR was computed for both simulated and experimental photoacoustic images generated by amplitude-based (i.e., delay-and-sum) and coherence-based (i.e., short-lag spatial coherence) beamformers. These gCNR measurements were compared to three more traditional image quality metrics (i.e., contrast, contrast-to-noise ratio, and signal-to-noise ratio) applied to the same datasets. An increase in qualitative target visibility generally corresponded with increased gCNR. In addition, gCNR magnitude was more directly related to the separability of photoacoustic signals from their background, which degraded with the presence of limited bandwidth artifacts and increased levels of channel noise. At high gCNR values (i.e., 0.95-1), contrast, contrast-to-noise ratio, and signal-to-noise ratio varied by up to 23.7-56.2 dB, 2.0-3.4, and 26.5-7.6×1020, respectively, for simulated, experimental phantom, andin vivodata. Therefore, these traditional metrics can experience large variations when a target is fully detectable, and additional increases in these values would have no impact on photoacoustic target detectability. In addition, gCNR is robust to changes in traditional metrics introduced by applying a minimum threshold to image amplitudes. In tandem with other photoacoustic image quality metrics and with a defined range of 0 to 1, gCNR has promising potential to provide additional insight, particularly when designing new beamformers and image formation techniques and when reporting quantitative performance without an opportunity to qualitatively assess corresponding images (e.g., in text-only abstracts).

     
    more » « less
  2. ABSTRACT

    We introduce a new class of iterative image reconstruction algorithms for radio interferometry, at the interface of convex optimization and deep learning, inspired by plug-and-play methods. The approach consists in learning a prior image model by training a deep neural network (DNN) as a denoiser, and substituting it for the handcrafted proximal regularization operator of an optimization algorithm. The proposed AIRI (‘AI for Regularization in radio-interferometric Imaging’) framework, for imaging complex intensity structure with diffuse and faint emission from visibility data, inherits the robustness and interpretability of optimization, and the learning power and speed of networks. Our approach relies on three steps. First, we design a low dynamic range training data base from optical intensity images. Secondly, we train a DNN denoiser at a noise level inferred from the signal-to-noise ratio of the data. We use training losses enhanced with a non-expansiveness term ensuring algorithm convergence, and including on-the-fly data base dynamic range enhancement via exponentiation. Thirdly, we plug the learned denoiser into the forward–backward optimization algorithm, resulting in a simple iterative structure alternating a denoising step with a gradient-descent data-fidelity step. We have validated AIRI against clean, optimization algorithms of the SARA family, and a DNN trained to reconstruct the image directly from visibility data. Simulation results show that AIRI is competitive in imaging quality with SARA and its unconstrained forward–backward-based version uSARA, while providing significant acceleration. clean remains faster but offers lower quality. The end-to-end DNN offers further acceleration, but with far lower quality than AIRI.

     
    more » « less
  3. Abstract

    Fluorescence imaging in centimeter-deep tissues with high resolution is highly desirable for many biomedical applications. Recently, we have developed a new imaging modality, ultrasound-switchable fluorescence (USF) imaging, for achieving this goal. In our previous work, we successfully achieved USF imaging with several types of USF contrast agents and imaging systems. In this study, we introduced a new USF imaging system: an intensified charge-coupled device (ICCD) camera-based, time-domain USF imaging system. We demonstrated the principle of time-domain USF imaging by using two USF contrast agents. With a series of USF imaging experiments, we demonstrated the tradeoffs among different experimental parameters (i.e., data acquisition time, including CCD camera recording time and intensifier gate delay; focused ultrasound (FU) power; and imaging depth) and the image qualities (i.e., signal-to-noise ratio, spatial resolution, and temporal resolution). In this study, we also discussed several imaging strategies for achieving a high-quality USF image via this time-domain system.

     
    more » « less
  4. We present a method to automatically synthesize efficient, high-quality demosaicking algorithms, across a range of computational budgets, given a loss function and training data. It performs a multi-objective, discrete-continuous optimization which simultaneously solves for the program structure and parameters that best tradeoff computational cost and image quality. We design the method to exploit domain-specific structure for search efficiency. We apply it to several tasks, including demosaicking both Bayer and Fuji X-Trans color filter patterns, as well as joint demosaicking and super-resolution. In a few days on 8 GPUs, it produces a family of algorithms that significantly improves image quality relative to the prior state-of-the-art across a range of computational budgets from 10 s to 1000 s of operations per pixel (1 dB–3 dB higher quality at the same cost, or 8.5–200× higher throughput at same or better quality). The resulting programs combine features of both classical and deep learning-based demosaicking algorithms into more efficient hybrid combinations, which are bandwidth-efficient and vectorizable by construction. Finally, our method automatically schedules and compiles all generated programs into optimized SIMD code for modern processors. 
    more » « less
  5. Most research on deep learning algorithms for image denoising has focused on signal-independent additive noise. Focused ion beam (FIB) microscopy with direct secondary electron detection has an unusual Neyman Type A (compound Poisson) measurement model, and sample damage poses fundamental challenges in obtaining training data. Model-based estimation is difficult and ineffective because of the nonconvexity of the negative log likelihood. In this paper, we develop deep learning-based denoising methods for FIB micrographs using synthetic training data generated from natural images. To the best of our knowledge, this is the first attempt in the literature to solve this problem with deep learning. Our results show that the proposed methods slightly outperform a total variation-regularized model-based method that requires time-resolved measurements that are not conventionally available. Improvements over methods using conventional measurements and less accurate noise modeling are dramatic - around 10 dB in peak signal-to-noise ratio. 
    more » « less