Deep Learning for Photoacoustic Tomography from Sparse Data

Antholzer, Stephan; Haltmeier, Markus; Schwab, Johannes

doi:10.48550/arxiv.1704.04587

Cited by 9 publications

(16 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…U-Net: In this case the reconstruction network (5) takes the form Φ Φ U-Net AE B, where Φ U-Net is the U-Net. The U-Net was initially proposed for image segmentation in [9], and lately has been used successfully for reconstruction tasks like low dose CT [5,6] and PAT [2]. The U-Net is a deep CNN, where each convolution is followed by the same nonlinearity, namely the rectified linear unit (ReLU) which is defined by Ê ÄÍ´Üµ Ñ Ü Ü ¼ .…”

Section: Proposed Reconstruction Networkmentioning

confidence: 99%

“…In particular, such structures allow to investigate the performance of the proposed deep learning methods on phantom classes without sharp boundaries. Results for piecewise constants phantoms can be found in [2]. To demonstrate stability of our deep learning approach with respect to measurement error, we added ½¼± Gaussian white noise to the simulated PAT measurements.…”

Section: Training and Evaluation Datamentioning

confidence: 99%

“…It is state of the art in many different domains and outperforms most comparable algorithms [1]. However, only recently they have been used for image reconstruction, see for example [2,3,4,5,6,7,8]. In image reconstruction with deep learning, a convolutional neural network (CNN) is designed to map the measured data to the desired reconstructed image.…”

Section: Introductionmentioning

confidence: 99%

“…Using deep learning and in particular deep CNNs for image reconstruction in PAT has first been proposed in our previous work [2]. In particular, in that paper we proposed the combination of the the FBP with a trained network for which the U-Net has been used.…”

Section: Introductionmentioning

confidence: 99%

“…Other learning approaches to PAT can be found in [10,11,12,13]. Opposed to [2], in this paper we present numerical results using the S-Net and compare it to the U-Net. Moreover, we use a different class of test phantoms (for training and testing) that mimic a nonuniform, more realistic, light distribution within the samples, and different measurement setups including limited view.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Photoacoustic image reconstruction via deep learning

Antholzer,

Schwab,

Nuster

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Applying standard algorithms to sparse data problems in photoacoustic tomography (PAT) yields low-quality images containing severe under-sampling artifacts. To some extent, these artifacts can be reduced by iterative image reconstruction algorithms which allow to include prior knowledge such as smoothness, total variation (TV) or sparsity constraints. These algorithms tend to be time consuming as the forward and adjoint problems have to be solved repeatedly. Further, iterative algorithms have additional drawbacks. For example, the reconstruction quality strongly depends on a-priori model assumptions about the objects to be recovered, which are often not strictly satisfied in practical applications. To overcome these issues, in this paper, we develop direct and efficient reconstruction algorithms based on deep learning. As opposed to iterative algorithms, we apply a convolutional neural network, whose parameters are trained before the reconstruction process based on a set of training data. For actual image reconstruction, a single evaluation of the trained network yields the desired result. Our presented numerical results (using two different network architectures) demonstrate that the proposed deep learning approach reconstructs images with a quality comparable to state of the art iterative reconstruction methods.

show abstract

Section: Proposed Reconstruction Networkmentioning

confidence: 99%

Section: Training and Evaluation Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Photoacoustic image reconstruction via deep learning

Antholzer,

Schwab,

Nuster

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Evaluation of Adjoint Methods in Photoacoustic Tomography with Under-Sampled Sensors

Lin

Azuma

Ünlü

et al. 2018

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018

View full text Add to dashboard Cite

Photo-Acoustic Tomography (PAT) can reconstruct a distribution of optical absorbers acting as instantaneous sound sources in subcutaneous microvasculature of a human breast. Adjoint methods for PAT, typically Time-Reversal (TR) and Back-Projection (BP), are ways to refocus time-reversed acoustic signals on sources by wave propagation from the position of sensors. TR and BP have different treatments for received signals, but they are equivalent under continuously sampling on a closed circular sensor array in two dimensions. Here, we analyze image quality with discrete under-sampled sensors in the sense of the Shannon sampling theorem. We investigate resolution and contrast of TR and BP, respectively in one source-sensor pair configuration and the frequency domain. With Hankel's asymptotic expansion to the integrands of imaging functions, our main contribution is to demonstrate that TR and BP have better performance on contrast and resolution, respectively. We also show that the integrand of TR includes additional side lobes which degrade axial resolution whereas that of BP conversely has relatively small amplitudes. Moreover, omnidirectional resolution is improved if more sensors are employed to collect the received signals. Nevertheless, for the under-sampled sensors, we propose the Truncated Back-Projection (TBP) method to enhance the contrast of BP using removing higher frequency components in the received signals. We conduct numerical experiments on the two-dimensional projected phantom model extracted from OA-Breast Database. The experiments verify our theories and show that the proposed TBP possesses better omnidirectional resolution as well as contrast compared with TR and BP with under-sampled sensors.

show abstract

Deep Learning Enhanced Mobile-Phone Microscopy

et al. 2018

View full text Add to dashboard Cite

Mobile-phones have facilitated the creation of field-portable, cost-effective imaging and sensing technologies that approach laboratory-grade instrument performance. However, the optical imaging interfaces of mobile-phones are not designed for microscopy and produce spatial and spectral distortions in imaging microscopic specimens. Here, we report on the use of deep learning to correct such distortions introduced by mobile-phone-based microscopes, facilitating the production of high-resolution, denoised and colour-corrected images, matching the performance of benchtop microscopes with high-end objective lenses, also extending their limited depth-of-field. After training a convolutional neural network, we successfully imaged various samples, including blood smears, histopathology tissue sections, and parasites, where the recorded images were highly compressed to ease storage and transmission for telemedicine applications. This method is applicable to other low-cost, aberrated imaging systems, and could offer alternatives for costly and bulky microscopes, while also providing a framework for standardization of optical images for clinical and biomedical applications. enhancement and aberration correction were performed computationally using a deep convolutional neural network (see Fig. 1, Supplementary Fig. 1, and the Methods section). Deep learning 12 is a powerful machine learning technique that can perform complex operations using a multi-layered artificial neural network and has shown great success in various tasks for which data are abundant [13][14][15][16] . The use of deep learning has also been demonstrated in numerous biomedical applications, such as diagnosis 17,18 , image classification 19 , among others 20-24 . In our method, a supervised learning approach is first applied by feeding the designed deep network with input (smartphone microscope images) and labels (gold standard benchtop microscope images obtained for the same samples) and optimizing a cost function that guides the network to learn the statistical transformation between the input and label. Following the deep network training phase, the network remains fixed and a smartphone microscope image input into the deep network is rapidly enhanced in terms of spatial resolution, signal-to-noise ratio, and colour response, attempting to match the overall image quality and the field of view (FOV) that would result from using a 20× objective lens on a high-end benchtop microscope. In addition, we demonstrate that the image output by the network will have a larger depth of field (DOF) than the corresponding image acquired using a high-NA objective lens on a benchtop microscope. Each enhanced image of the mobile microscope is inferred by the deep network in a non-iterative, feed-forward manner. For example, the deep network generates an enhanced output image with a FOV of ~0.57 mm 2 (the same as that of a 20× objective lens), from a smartphone microscope image within ~0.42 s, using a standard personal computer equipped with a dual graphics-processing un...

show abstract

Deep Learning for Photoacoustic Tomography from Sparse Data

Cited by 9 publications

References 0 publications

Photoacoustic image reconstruction via deep learning

Photoacoustic image reconstruction via deep learning

Evaluation of Adjoint Methods in Photoacoustic Tomography with Under-Sampled Sensors

Deep Learning Enhanced Mobile-Phone Microscopy

Contact Info

Product

Resources

About