Compressive sensing in medical imaging

FusionOpt-Net: A Transformer-Based Compressive Sensing Reconstruction Algorithm

Compressive sensing (CS) is a notable technique in signal processing, especially in multimedia, as it allows for simultaneous signal acquisition and dimensionality reduction. Recent advancements in deep learning (DL) have led to the creation of deep unfolding architectures, which overcome the inefficiency and subpar quality of traditional CS reconstruction methods. In this paper, we introduce a novel CS image reconstruction algorithm that leverages the strengths of the fast iterative shrinkage-thresholding algorithm (FISTA) and modern Transformer networks. To enhance computational efficiency, we employ a block-based sampling approach in the sampling module. By mapping FISTA's iterative process onto neural networks in the reconstruction module, we address the hyperparameter challenges of traditional algorithms, thereby improving reconstruction efficiency. Moreover, the robust feature extraction capabilities of Transformer networks significantly enhance image reconstruction quality. Experimental results show that the FusionOpt-Net model surpasses other advanced methods on various public benchmark datasets.

Compressed-sensing-inspired reconstruction algorithms in low-dose computed tomography: A review

BACKGROUND

Optimization of the dose the patient receives during scanning is an important problem in modern medical X-ray computed tomography (CT). One of the basic ways to its solution is to reduce the number of views. Compressed sensing theory helped promote the development of a new class of effective reconstruction algorithms for limited data CT. These compressed-sensing-inspired (CSI) algorithms optimize the Lp (0 ≤ p ≤ 1) norm of images and can accurately reconstruct CT tomograms from a very few views. The paper presents a review of the CSI algorithms and discusses prospects for their further use in commercial low-dose CT.

METHODS

Many literature references with the CSI algorithms have been were searched. To structure the material collected the author gives a classification framework within which he describes Lp regularization methods, the basic CSI algorithms that are used most often in few-view CT, and some of their derivatives. Lots of examples are provided to illustrate the use of the CSI algorithms in few-view and low-dose CT.

RESULTS

A list of the CSI algorithms is compiled from the literature search. For better demonstrativeness they are summarized in a table. The inference is done that already today some of the algorithms are capable of reconstruction from 20 to 30 views with acceptable quality and dose reduction by a factor of 10.

DISCUSSION

In conclusion the author discusses how soon the CSI reconstruction algorithms can be introduced in the practice of medical diagnosis and used in commercial CT scanners.

Task-based assessment for neural networks: evaluating undersampled MRI reconstructions based on human observer signal detection

Purpose

Recent research explores using neural networks to reconstruct undersampled magnetic resonance imaging. Because of the complexity of the artifacts in the reconstructed images, there is a need to develop task-based approaches to image quality. We compared conventional global quantitative metrics to evaluate image quality in undersampled images generated by a neural network with human observer performance in a detection task. The purpose is to study which acceleration (2×, 3×, 4×, 5×) would be chosen with the conventional metrics and compare it to the acceleration chosen by human observer performance.

Approach

We used common global metrics for evaluating image quality: the normalized root mean squared error (NRMSE) and structural similarity (SSIM). These metrics are compared with a measure of image quality that incorporates a subtle signal for a specific task to allow for image quality assessment that locally evaluates the effect of undersampling on a signal. We used a U-Net to reconstruct under-sampled images with 2×, 3×, 4×, and 5× one-dimensional undersampling rates. Cross-validation was performed for a 500- and a 4000-image training set with both SSIM and MSE losses. A two-alternative forced choice (2-AFC) observer study was carried out for detecting a subtle signal (small blurred disk) from images with the 4000-image training set.

Results

We found that for both loss functions, the human observer performance on the 2-AFC studies led to a choice of a 2× undersampling, but the SSIM and NRMSE led to a choice of a 3× undersampling.

Conclusions

For this detection task using a subtle small signal at the edge of detectability, SSIM and NRMSE led to an overestimate of the achievable undersampling using a U-Net before a steep loss of image quality between 2×, 3×, 4×, 5× undersampling rates when compared to the performance of human observers in the detection task.

AliasNet: Alias artefact suppression network for accelerated phase-encode MRI

Sparse reconstruction is an important aspect of MRI, helping to reduce acquisition time and improve spatial-temporal resolution. Popular methods are based mostly on compressed sensing (CS), which relies on the random sampling of k-space to produce incoherent (noise-like) artefacts. Due to hardware constraints, 1D Cartesian phase-encode under-sampling schemes are popular for 2D CS-MRI. However, 1D under-sampling limits 2D incoherence between measurements, yielding structured aliasing artefacts (ghosts) that may be difficult to remove assuming a 2D sparsity model. Reconstruction algorithms typically deploy direction-insensitive 2D regularisation for these direction-associated artefacts. Recognising that phase-encode artefacts can be separated into contiguous 1D signals, we develop two decoupling techniques that enable explicit 1D regularisation and leverage the excellent 1D incoherence characteristics. We also derive a combined 1D + 2D reconstruction technique that takes advantage of spatial relationships within the image. Experiments conducted on retrospectively under-sampled brain and knee data demonstrate that combination of the proposed 1D AliasNet modules with existing 2D deep learned (DL) recovery techniques leads to an improvement in image quality. We also find AliasNet enables a superior scaling of performance compared to increasing the size of the original 2D network layers. AliasNet therefore improves the regularisation of aliasing artefacts arising from phase-encode under-sampling, by tailoring the network architecture to account for their expected appearance. The proposed 1D + 2D approach is compatible with any existing 2D DL recovery technique deployed for this application.

Modeling human observer detection in undersampled magnetic resonance imaging reconstruction with total variation and wavelet sparsity regularization

Purpose

Task-based assessment of image quality in undersampled magnetic resonance imaging provides a way of evaluating the impact of regularization on task performance. In this work, we evaluated the effect of total variation (TV) and wavelet regularization on human detection of signals with a varying background and validated a model observer in predicting human performance.

Approach

Human observer studies used two-alternative forced choice (2-AFC) trials with a small signal known exactly task but with varying backgrounds for fluid-attenuated inversion recovery images reconstructed from undersampled multi-coil data. We used a 3.48 undersampling factor with TV and a wavelet sparsity constraints. The sparse difference-of-Gaussians (S-DOG) observer with internal noise was used to model human observer detection. The internal noise for the S-DOG was chosen to match the average percent correct (PC) in 2-AFC studies for four observers using no regularization. That S-DOG model was used to predict the PC of human observers for a range of regularization parameters.

Results

We observed a trend that the human observer detection performance remained fairly constant for a broad range of values in the regularization parameter before decreasing at large values. A similar result was found for the normalized ensemble root mean squared error. Without changing the internal noise, the model observer tracked the performance of the human observers as the regularization was increased but overestimated the PC for large amounts of regularization for TV and wavelet sparsity, as well as the combination of both parameters.

Conclusions

For the task we studied, the S-DOG observer was able to reasonably predict human performance with both TV and wavelet sparsity regularizers over a broad range of regularization parameters. We observed a trend that task performance remained fairly constant for a range of regularization parameters before decreasing for large amounts of regularization.

RISING: A new framework for model-based few-view CT image reconstruction with deep learning

Medical image reconstruction from low-dose tomographic data is an active research field, recently revolutionized by the advent of deep learning. In fact, deep learning typically yields superior results than classical optimization approaches, but unstable results have been reported both numerically and theoretically in the literature. This paper proposes RISING, a new framework for sparse-view tomographic image reconstruction combining an early-stopped Rapid Iterative Solver with a subsequent Iteration Network-based Gaining step. In our two-step approach, the first phase executes very few iterations of a regularized model-based algorithm, whereas the second step completes the missing iterations by means of a convolutional neural network. The proposed method is ground-truth free; it exploits the computational speed and flexibility of a data-driven approach, but it also imposes sparsity constraints to the solution as in the model-based setting. Experiments performed both on a digitally created and on a real abdomen data set confirm the numerical and visual accuracy of the reconstructed RISING images in short computational times. These features make the framework promising to be used on real systems for clinical exams.

Adaptive High-Resolution Imaging Method Based on Compressive Sensing

Compressive sensing (CS) is a signal sampling theory that originated about 16 years ago. It replaces expensive and complex receiving devices with well-designed signal recovery algorithms, thus simplifying the imaging system. Based on the application of CS theory, a single-pixel camera with an array-detection imaging system is established for high-pixel detection. Each detector of the detector array is coupled with a bundle of fibers formed by fusion of four bundles of fibers of different lengths, so that the target area corresponding to one detector is split into four groups of target information arriving at different times. By comparing the total amount of information received by the detector with the threshold set in advance, it can be determined whether the four groups of information are calculated separately. The simulation results show that this new system can not only reduce the number of measurements required to reconstruct high quality images but can also handle situations wherever the target may appear in the field of view without necessitating an increase in the number of detectors.

3D medical images security via light-field imaging

This Letter proposes a selective encryption scheme for three-dimensional (3D) medical images using light-field imaging and two-dimensional (2D) Moore cellular automata (MCA). We first utilize convolutional neural networks (CNNs) to obtain the saliency of each elemental image (EI) originating from a 3D medical image with different viewpoints, and successfully extract the region of interest (ROI) in each EI. In addition, we use 2D MCA with balanced rule to encrypt the ROI of each EI. Finally, the decrypted elemental image array (EIA) can be reconstructed into a full-color and full-parallax 3D image using the display device, which can be visually displayed to doctors so that they can observe from different angles to design accurate treatment plans and improve the level of medical treatment. Our work also requires no preprocessing of 3D images, which is more efficient than the method of using slices for encryption.

Block compressive sensing chaotic embedded encryption for MCF-OFDM transmission system

In this paper, we propose a block compressive sensing (BCS) based chaotic embedded encryption scheme for multi-core fiber orthogonal frequency division multiplexing (MCF-OFDM) system. BCS technology is used to recover the entire desired information from the small amounts of data. Meanwhile, a four-dimensional discrete chaotic encryption model generates four masking factors, which are respectively used for coefficient random permutation (CRP), measurement matrix, diffusion and singular value decomposition (SVD) embedding to achieve ultra-high security encryption of four different dimensions. In terms of compressive sensing, CRP can make the discrete cosine transform (DCT) coefficient distribute randomly to improve the sampling efficiency of BCS. Compared with the data without compressive sensing, the data volume is reduced by 75%. In chaotic encryption, SVD technology embeds secret images of noise-like after initial encryption into carrier images to generate encrypted images with visual security. The key space reaches 10¹²⁰ and it realizes the dual protection of source image data and external representation. The proposed scheme using a 2km 7-core optical fiber achieves a 78.75 Gb/s transmission of encrypted OFDM signals. The received optical power is greater than -14 dBm, and the bit error rate (BER) of core1-core7 is lower than 10^-3. When the compression ratio sets to 0.25 and the attack range of encrypted data is up to 30%, the image can still recover the outline and general information. The experimental results show that this scheme can improve the security performance and reduce the complexity of information transmission system. Furthermore, the scheme combines The BCS chaotic embedded encryption technology with MCF-OFDM system, which has a good application prospect in the future optical networks.

De-aliased depth-range-extended optical coherence tomography based on dual under-sampling

We demonstrate a dual under-sampling (DUS) method to achieve de-aliased and depth-range-extended optical coherence tomography (OCT) imaging. The spectral under-sampling can significantly reduce the data size but causes well-known aliasing artifacts. A change in the sampling frequency used to acquire the interference spectrum alters the aliasing period within the output window except for the true image; this feature is utilized to distinguish the true image from the aliasing artifacts. We demonstrate that with DUS, the data size is reduced to 37% at an extended depth range of 24 mm, over which the true depth can be precisely measured without ambiguity. This reduction in data size and precise measuring capability would be beneficial for reducing the acquisition time for OCT imaging in various biomedical and industrial applications.

Secondary Complementary Balancing Compressive Imaging with a Free-Space Balanced Amplified Photodetector

Single-pixel imaging (SPI) has attracted widespread attention because it generally uses a non-pixelated photodetector and a digital micromirror device (DMD) to acquire the object image. Since the modulated patterns seen from two reflection directions of the DMD are naturally complementary, one can apply complementary balanced measurements to greatly improve the measurement signal-to-noise ratio and reconstruction quality. However, the balance between two reflection arms significantly determines the quality of differential measurements. In this work, we propose and demonstrate a simple secondary complementary balancing mechanism to minimize the impact of the imbalance on the imaging system. In our SPI setup, we used a silicon free-space balanced amplified photodetector with 5 mm active diameter which could directly output the difference between two optical input signals in two reflection arms. Both simulation and experimental results have demonstrated that the use of secondary complementary balancing can result in a better cancellation of direct current components of measurements, and can acquire an image quality slightly better than that of single-arm single-pixel complementary measurement scheme (which is free from the trouble of optical imbalance) and over 20 times better than that of double-arm dual-pixel complementary measurement scheme under optical imbalance conditions.

A Dual-Mode 303-Megaframes-per-Second Charge-Domain Time-Compressive Computational CMOS Image Sensor

An ultra-high-speed computational CMOS image sensor with a burst frame rate of 303 megaframes per second, which is the fastest among the solid-state image sensors, to our knowledge, is demonstrated. This image sensor is compatible with ordinary single-aperture lenses and can operate in dual modes, such as single-event filming mode or multi-exposure imaging mode, by reconfiguring the number of exposure cycles. To realize this frame rate, the charge modulator drivers were adequately designed to suppress the peak driving current taking advantage of the operational constraint of the multi-tap charge modulator. The pixel array is composed of macropixels with 2 × 2 4-tap subpixels. Because temporal compressive sensing is performed in the charge domain without any analog circuit, ultrafast frame rates, small pixel size, low noise, and low power consumption are achieved. In the experiments, single-event imaging of plasma emission in laser processing and multi-exposure transient imaging of light reflections to extend the depth range and to decompose multiple reflections for time-of-flight (TOF) depth imaging with a compression ratio of 8× were demonstrated. Time-resolved images similar to those obtained by the direct-type TOF were reproduced in a single shot, while the charge modulator for the indirect TOF was utilized.

Computing a projection operator onto the null space of a linear imaging operator: tutorial

Many imaging systems can be approximately described by a linear operator that maps an object property to a collection of discrete measurements. However, even in the absence of measurement noise, such operators are generally "blind" to certain components of the object, and hence information is lost in the imaging process. Mathematically, this is explained by the fact that the imaging operator can possess a null space. All objects in the null space, by definition, are mapped to a collection of identically zero measurements and are hence invisible to the imaging system. As such, characterizing the null space of an imaging operator is of fundamental importance when comparing and/or designing imaging systems. A characterization of the null space can also facilitate the design of regularization strategies for image reconstruction methods. Characterizing the null space via an associated projection operator is, in general, a computationally demanding task. In this tutorial, computational procedures for establishing projection operators that map an object to the null space of a discrete-to-discrete imaging operator are surveyed. A new machine-learning-based approach that employs a linear autoencoder is also presented. The procedures are demonstrated by use of biomedical imaging examples, and their computational complexities and memory requirements are compared.

Fractional Fourier single-pixel imaging

Single-pixel imaging technology has a number of advantages over conventional imaging approaches, such as wide operation wavelength region, compressive sampling, low light radiation dose and insensitivity to distortion. Here, we report on a novel single-pixel imaging based on fractional Fourier transform (FRFT), which captures images by acquiring the fractional-domain information of targets. With the use of structured illumination of two-dimensional FRFT base patterns, FRFT coefficients of the object could be measured by single-pixel detection. Then, the object image is achieved by performing inverse FRFT on the measurements. Furthermore, the proposed method can reconstruct the object image from sub-Nyquist measurements because of the sparsity of image data in fractional domain. In comparison with traditional single-pixel imaging, it provides a new degree of freedom, namely fractional order, and therefore has more flexibility and new features for practical applications. In experiments, the proposed method has been applied for edge detection of object, with an adjustable parameter as a new degree of freedom.

A Green Prospective for Learned Post-Processing in Sparse-View Tomographic Reconstruction

Deep Learning is developing interesting tools that are of great interest for inverse imaging applications. In this work, we consider a medical imaging reconstruction task from subsampled measurements, which is an active research field where Convolutional Neural Networks have already revealed their great potential. However, the commonly used architectures are very deep and, hence, prone to overfitting and unfeasible for clinical usages. Inspired by the ideas of the green AI literature, we propose a shallow neural network to perform efficient Learned Post-Processing on images roughly reconstructed by the filtered backprojection algorithm. The results show that the proposed inexpensive network computes images of comparable (or even higher) quality in about one-fourth of time and is more robust than the widely used and very deep ResUNet for tomographic reconstructions from sparse-view protocols.

LightSpeed: A Compact, High-Speed Optical-Link-Based 3D Optoacoustic Imager

Wide-scale adoption of optoacoustic imaging in biology and medicine critically depends on availability of affordable scanners combining ease of operation with optimal imaging performance. Here we introduce LightSpeed: a low-cost real-time volumetric handheld optoacoustic imager based on a new compact software-defined ultrasound digital acquisition platform and a pulsed laser diode. It supports the simultaneous signal acquisition from up to 192 ultrasound channels and provides a hig-bandwidth direct optical link (2x 100G Ethernet) to the host-PC for ultra-high frame rate image acquisitions. We demonstrate use of the system for ultrafast (500Hz) 3D human angiography with a rapidly moving handheld probe. LightSpeed attained image quality comparable with a conventional optoacoustic imaging systems employing bulky acquisition electronics and a Q-switched pulsed laser. Our results thus pave the way towards a new generation of compact, affordable and high-performance optoacoustic scanners.

Compressed Sensing Real-Time Cine Reduces CMR Arrhythmia-Related Artifacts

Background and objective: Cardiac magnetic resonance (CMR) is a key tool for cardiac work-up. However, arrhythmia can be responsible for arrhythmia-related artifacts (ARA) and increased scan time using segmented sequences. The aim of this study is to evaluate the effect of cardiac arrhythmia on image quality in a comparison of a compressed sensing real-time (CS_rt) cine sequence with the reference prospectively gated segmented balanced steady-state free precession (Cine_ref) technique regarding ARA. Methods: A total of 71 consecutive adult patients (41 males; mean age = 59.5 ± 20.1 years (95% CI: 54.7–64.2 years)) referred for CMR examination with concomitant irregular heart rate (defined by an RR interval coefficient of variation >10%) during scanning were prospectively enrolled. For each patient, two cine sequences were systematically acquired: first, the reference prospectively triggered multi-breath-hold Cine_ref sequence including a short-axis stack, one four-chamber slice, and a couple of two-chamber slices; second, an additional single breath-hold CS_rt sequence providing the same slices as the reference technique. Two radiologists independently assessed ARA and image quality (overall, acquisition, and edge sharpness) for both techniques. Results: The mean heart rate was 71.8 ± 19.0 (SD) beat per minute (bpm) (95% CI: 67.4–76.3 bpm) and its coefficient of variation was 25.0 ± 9.4 (SD) % (95% CI: 22.8–27.2%). Acquisition was significantly faster with CS_rt than with Cine_ref (Cine_ref: 556.7 ± 145.4 (SD) s (95% CI: 496.7–616.7 s); CS_rt: 23.9 ± 7.9 (SD) s (95% CI: 20.6–27.1 s); p < 0.0001). A total of 599 pairs of cine slices were evaluated (median: 8 (range: 6–14) slices per patient). The mean proportion of ARA-impaired slices per patient was 85.9 ± 22.7 (SD) % using Cine_ref, but this was figure was zero using CS_rt (p < 0.0001). The European CMR registry artifact score was lower with CS_rt (median: 1 (range: 0–5)) than with Cine_ref (median: 3 (range: 0–3); p < 0.0001). Subjective image quality was higher in CS_rt than in Cine_ref (median: 3 (range: 1–3) versus 2 (range: 1–4), respectively; p < 0.0001). In line, edge sharpness was higher on CS_rt cine than on Cine_ref images (0.054 ± 0.016 pixel⁻¹ (95% CI: 0.050–0.057 pixel⁻¹) versus 0.042 ± 0.022 pixel⁻¹ (95% CI: 0.037–0.047 pixel⁻¹), respectively; p = 0.0001). Conclusion: Compressed sensing real-time cine drastically reduces arrhythmia-related artifacts and thus improves cine image quality in patients with arrhythmia.

Optimizing constrained reconstruction in magnetic resonance imaging for signal detection

Constrained reconstruction in magnetic resonance imaging (MRI) allows the use of prior information through constraints to improve reconstructed images. These constraints often take the form of regularization terms in the objective function used for reconstruction. Constrained reconstruction leads to images which appear to have fewer artifacts than reconstructions without constraints but because the methods are typically nonlinear, the reconstructed images have artifacts whose structure is hard to predict. In this work, we compared different methods of optimizing the regularization parameter using a total variation (TV) constraint in the spatial domain and sparsity in the wavelet domain for one-dimensional (2.56×) undersampling using variable density undersampling. We compared the mean squared error (MSE), structural similarity (SSIM), L-curve and the area under the receiver operating characteristic (AUC) using a linear discriminant for detecting a small and a large signal. We used a signal-known-exactly task with varying backgrounds in a simulation where the anatomical variation was the major source of clutter for the detection task. Our results show that the AUC dependence on regularization parameters varies with the imaging task (i.e. the signal being detected). The choice of regularization parameters for MSE, SSIM, L-curve and AUC were similar. We also found that a model-based reconstruction including TV and wavelet sparsity did slightly better in terms of AUC than just enforcing data consistency but using these constraints resulted in much better MSE and SSIM. These results suggest that the increased performance in MSE and SSIM over-estimate the improvement in detection performance for the tasks in this paper. The MSE and SSIM metrics show a big difference in performance where the difference in AUC is small. To our knowledge, this is the first time that signal detection with varying backgrounds has been used to optimize constrained reconstruction in MRI.

A Model-Based Optimization Framework for Iterative Digital Breast Tomosynthesis Image Reconstruction

Digital Breast Tomosynthesis is an X-ray imaging technique that allows a volumetric reconstruction of the breast, from a small number of low-dose two-dimensional projections. Although it is already used in the clinical setting, enhancing the quality of the recovered images is still a subject of research. The aim of this paper was to propose and compare, in a general optimization framework, three slightly different models and corresponding accurate iterative algorithms for Digital Breast Tomosynthesis image reconstruction, characterized by a convergent behavior. The suggested model-based implementations are specifically aligned to Digital Breast Tomosynthesis clinical requirements and take advantage of a Total Variation regularizer. We also tune a fully-automatic strategy to set a proper regularization parameter. We assess our proposals on real data, acquired from a breast accreditation phantom and a clinical case. The results confirm the effectiveness of the presented framework in reconstructing breast volumes, with particular focus on the masses and microcalcifications, in few iterations and in enhancing the image quality in a prolonged execution.

High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning

SIGNIFICANCE

Reducing the bit depth is an effective approach to lower the cost of an optical coherence tomography (OCT) imaging device and increase the transmission efficiency in data acquisition and telemedicine. However, a low bit depth will lead to the degradation of the detection sensitivity, thus reducing the signal-to-noise ratio (SNR) of OCT images.

AIM

We propose using deep learning to reconstruct high SNR OCT images from low bit-depth acquisition.

APPROACH

The feasibility of our approach is evaluated by applying this approach to the quantized 3- to 8-bit data from native 12-bit interference fringes. We employ a pixel-to-pixel generative adversarial network (pix2pixGAN) architecture in the low-to-high bit-depth OCT image transition.

RESULTS

Extensively, qualitative and quantitative results show our method could significantly improve the SNR of the low bit-depth OCT images. The adopted pix2pixGAN is superior to other possible deep learning and compressed sensing solutions.

CONCLUSIONS

Our work demonstrates that the proper integration of OCT and deep learning could benefit the development of healthcare in low-resource settings.

Compressed Optoacoustic Sensing of Volumetric Cardiac Motion

The recently developed optoacoustic tomography systems have attained volumetric frame rates exceeding 100 Hz, thus opening up new venues for studying previously invisible biological dynamics. Further gains in temporal resolution can potentially be achieved via partial data acquisition, though a priori knowledge on the acquired data is essential for rendering accurate reconstructions using compressed sensing approaches. In this work, we suggest a machine learning method based on principal component analysis for high-frame-rate volumetric cardiac imaging using only a few tomographic optoacoustic projections. The method is particularly effective for discerning periodic motion, as demonstrated herein by non-invasive imaging of a beating mouse heart. A training phase enables efficiently compressing the heart motion information, which is subsequently used as prior information for image reconstruction from sparse sampling at a higher frame rate. It is shown that image quality is preserved with a 64-fold reduction in the data flow. We demonstrate that, under certain conditions, the volumetric motion could effectively be captured by relying on time-resolved data from a single optoacoustic detector. Feasibility of capturing transient (non-periodic) events not registered in the training phase is further demonstrated by visualizing perfusion of a contrast agent in vivo. The suggested approach can be used to significantly boost the temporal resolution of optoacoustic imaging and facilitate development of more affordable and data efficient systems.

GPU acceleration of a model-based iterative method for Digital Breast Tomosynthesis

Digital Breast Tomosynthesis (DBT) is a modern 3D Computed Tomography X-ray technique for the early detection of breast tumors, which is receiving growing interest in the medical and scientific community. Since DBT performs incomplete sampling of data, the image reconstruction approaches based on iterative methods are preferable to the classical analytic techniques, such as the Filtered Back Projection algorithm, providing fewer artifacts. In this work, we consider a Model-Based Iterative Reconstruction (MBIR) method well suited to describe the DBT data acquisition process and to include prior information on the reconstructed image. We propose a gradient-based solver named Scaled Gradient Projection (SGP) for the solution of the constrained optimization problem arising in the considered MBIR method. Even if the SGP algorithm exhibits fast convergence, the time required on a serial computer for the reconstruction of a real DBT data set is too long for the clinical needs. In this paper we propose a parallel SGP version designed to perform the most expensive computations of each iteration on Graphics Processing Unit (GPU). We apply the proposed parallel approach on three different GPU boards, with computational performance comparable with that of the boards usually installed in commercial DBT systems. The numerical results show that the proposed GPU-based MBIR method provides accurate reconstructions in a time suitable for clinical trials.

Artificial intelligence in cardiovascular imaging: state of the art and implications for the imaging cardiologist

Healthcare, conceivably more than any other area of human endeavour, has the greatest potential to be affected by artificial intelligence (AI). This potential has been shown by several reports that demonstrate equal or superhuman performance in medical tasks that aim to improve efficiency, diagnosis and prognosis. This review focuses on the state of the art of AI applications in cardiovascular imaging. It provides an overview of the current applications and studies performed, including the potential value, implications, limitations and future directions of AI in cardiovascular imaging.It is envisioned that AI will dramatically change the way doctors practise medicine. In the short term, it will assist physicians with easy tasks, such as automating measurements, making predictions based on big data, and putting clinical findings into an evidence-based context. In the long term, AI will not only assist doctors, it has the potential to significantly improve access to health and well-being data for patients and their caretakers. This empowers patients. From a physician's perspective, reliable AI assistance will be available to support clinical decision-making. Although cardiovascular studies implementing AI are increasing in number, the applications have only just started to penetrate contemporary clinical care.

Accelerating cardiac diffusion tensor imaging combining local low-rank and 3D TV constraint

OBJECTIVE

Diffusion tensor magnetic resonance imaging (DT-MRI, or DTI) is a promising technique for invasively probing biological tissue structures. However, DTI is known to suffer from much longer acquisition time with respect to conventional MRI and the problem is worsened when dealing with in vivo acquisitions. Therefore, faster DTI for both ex vivo and in vivo scans is highly desired.

MATERIALS AND METHODS

This paper proposes a new compressed sensing (CS) reconstruction method that employs local low-rank (LLR) model and three-dimensional (3D) total variation (TV) constraint to reconstruct cardiac diffusion-weighted (DW) images from highly undersampled k-space data. The LLR model takes the set of DW images corresponding to different diffusion gradient directions as a 3D image volume and decomposes the latter into overlapping 3D blocks. Then, the 3D blocks are stacked as two-dimensional (2D) matrix. Finally, low-rank property is applied to each block matrix and the 3D TV constraint to the 3D image volume. The underlying constrained optimization problem is finally solved using the first-order fast method. The proposed method is evaluated on real ex vivo cardiac DTI data as a prerequisite to in vivo cardiac DTI applications.

RESULTS

The results on real human ex vivo cardiac DTI images demonstrate that the proposed method exhibits lower reconstruction errors for DTI indices, including fractional anisotropy (FA), mean diffusivities (MD), transverse angle (TA), and helix angle (HA), compared to existing CS-based DTI image reconstruction techniques.

CONCLUSION

The proposed method provides better reconstruction quality and more accurate DTI indices in comparison with the state-of-the-art CS-based DW image reconstruction methods.

Sparse Wavenumber Recovery and Prediction of Anisotropic Guided Waves in Composites: A Comparative Study

Guided wave methodologies are among the established approaches for structural health monitoring (SHM). For guided wave data, being able to accurately estimate wave properties in the absence of ample measurements can greatly facilitate the often time-consuming and potentially expensive data acquisition procedure. Nevertheless, inherent complexities of the guided waves, including their multimodal and frequency dispersive nature, hinder processing, analysis, and behavior prediction. The severity of these complexities is even higher in anisotropic media, such as composites. Several methods, including sparse wavenumber analysis (SWA), have been proposed in the literature to characterize guided wave propagation by extracting wave characteristics in a particular medium from the information contained in a few measurements, and subsequently using this information for full wavefield prediction. In this paper, we investigate the efficacy of guided wave reconstruction techniques, based on SWA, for predicting the behavior of guided waves in composite materials. We implement these techniques on several experimental and simulation data sets. We study their performance in estimating the frequency-dependent (dispersive) and anisotropic velocities of guided waves and in reconstructing full wavefields from limited available information.

Advancing Regulatory Science With Computational Modeling for Medical Devices at the FDA's Office of Science and Engineering Laboratories

Protecting and promoting public health is the mission of the U.S. Food and Drug Administration (FDA). FDA's Center for Devices and Radiological Health (CDRH), which regulates medical devices marketed in the U.S., envisions itself as the world's leader in medical device innovation and regulatory science-the development of new methods, standards, and approaches to assess the safety, efficacy, quality, and performance of medical devices. Traditionally, bench testing, animal studies, and clinical trials have been the main sources of evidence for getting medical devices on the market in the U.S. In recent years, however, computational modeling has become an increasingly powerful tool for evaluating medical devices, complementing bench, animal and clinical methods. Moreover, computational modeling methods are increasingly being used within software platforms, serving as clinical decision support tools, and are being embedded in medical devices. Because of its reach and huge potential, computational modeling has been identified as a priority by CDRH, and indeed by FDA's leadership. Therefore, the Office of Science and Engineering Laboratories (OSEL)-the research arm of CDRH-has committed significant resources to transforming computational modeling from a valuable scientific tool to a valuable regulatory tool, and developing mechanisms to rely more on digital evidence in place of other evidence. This article introduces the role of computational modeling for medical devices, describes OSEL's ongoing research, and overviews how evidence from computational modeling (i.e., digital evidence) has been used in regulatory submissions by industry to CDRH in recent years. It concludes by discussing the potential future role for computational modeling and digital evidence in medical devices.

A wavelet gradient sparsity based algorithm for reconstruction of reduced-view tomography datasets obtained with a monochromatic synchrotron-based X-ray source

High-resolution synchrotron computed tomography (CT) is very helpful in the diagnosis and monitor of chronic diseases including osteoporosis. Osteoporosis is characterized by low bone mass and cortical bone porosity best imaged with CT. Synchrotron CT requires a large number of angular projections to reconstruct images with high resolution for detailed and accurate diagnosis. However, this poses great risks and challenges for serial in-vivo human and animal imaging due to a large amount of X-ray radiation dose required that can damage living specimens. Also, longer scan times are associated with increased risk of specimen movement and motion artifact in the reconstructed images. We developed a wavelet-gradient sparsity based algorithm to be utilized as a synchrotron tomography reconstruction technique allowing accurate reconstruction of cortical bone porosity assessed for in-vivo preclinical study which significantly reduces the radiation dose and scan time required while maintaining satisfactory image resolution for diagnosis. The results of our study on a rat forelimb sample imaged in the Biomedical Imaging and Therapy Bending Magnet (BMIT-BM) beamline at the Canadian Light Source show that the proposed algorithm can produce satisfactory image quality with more than 50 percent X-ray dose reduction as indicated by both visual and quantitative-based performance.

Enhancing sparse-view photoacoustic tomography with combined virtually parallel projecting and spatially adaptive filtering

To fully realize the potential of photoacoustic tomography (PAT) in preclinical and clinical applications, rapid measurements and robust reconstructions are needed. Sparse-view measurements have been adopted effectively to accelerate the data acquisition. However, since the reconstruction from the sparse-view sampling data is challenging, both the effective measurement and the appropriate reconstruction should be taken into account. In this study, we present an iterative sparse-view PAT reconstruction scheme, where a concept of virtual parallel-projection matching the measurement condition is introduced to aid the "compressive sensing" in the reconstruction procedure, and meanwhile, the non-local spatially adaptive filtering exploring the a priori information of the mutual similarities in natural images is adopted to recover the unknowns in the transformed sparse domain. Consequently, the reconstructed images with the proposed sparse-view scheme can be evidently improved in comparison to those with the universal back-projection method, for the cases of same sparse views. The proposed approach has been validated by the simulations and ex vivo experiments, which exhibits desirable performances in image fidelity even from a small number of measuring positions.

Thermal Light Longitudinal Correlated Imaging with Random Orthogonal Matching Pursuit Algorithm

A thermal light correlated longitudinal imaging experiment is proposed. The quasi-thermal light beam is split into two beams, a test beam and a reference beam, respectively. The light in the test beam is scattered by two amplitude objects with a specific longitudinal distance between them, while the light of the reference beam travels uninterrupted. At the end of the test and reference beams, two charge-coupled detectors (CCDs) are used to measure the intensity of the optical field. Through intensity correlation measurement the images of the two detected objects can be achieved simultaneously, only if the distance between the objects is less than the longitudinal coherent length. The theoretical analysis shows that the longitudinal coherent length is determined by both the transverse size of the incoherent thermal light source and the length of the optical path. The quality of the correlated images of the two objects is improved greatly by making use of the orthogonal matching pursuit (OMP) and the proposed variant random orthogonal matching pursuit (Random-OMP) algorithms. The experimental results show that the Random-OMP algorithm is more effective than the OMP algorithm for increasing both the visibility and continuity of the images. The experimental scenario can mimic an optical tomography imaging system, and the two objects with longitudinal distance can be taken as the two transverse layers of a three-dimensional object. The proposed Random-OMP algorithm is effective for improving the quality of the tomography image and has potential value in optical tomography imaging technology using incoherent light sources.

Feasibility of high spatiotemporal resolution for an abbreviated 3D radial breast MRI protocol

PURPOSE

To develop a volumetric imaging technique with 0.8-mm isotropic resolution and 10-s/volume rate to detect and analyze breast lesions in a bilateral, dynamic, contrast-enhanced MRI exam.

METHODS

A local low-rank temporal reconstruction approach that also uses parallel imaging and spatial compressed sensing was designed to create rapid volumetric frame rates during a contrast-enhanced breast exam (vastly undersampled isotropic projection [VIPR] spatial compressed sensing with temporal local low-rank [STELLR]). The dynamic-enhanced data are subtracted in k-space from static mask data to increase sparsity for the local low-rank approach to maximize temporal resolution. A T₁ -weighted 3D radial trajectory (VIPR iterative decomposition with echo asymmetry and least squares estimation [IDEAL]) was modified to meet the data acquisition requirements of the STELLR approach. Additionally, the unsubtracted enhanced data are reconstructed using compressed sensing and IDEAL to provide high-resolution fat/water separation. The feasibility of the approach and the dual reconstruction methodology is demonstrated using a 16-channel breast coil and a 3T MR scanner in 6 patients.

RESULTS

The STELLR temporal performance of subtracted data matched the expected temporal perfusion enhancement pattern in small and large vascular structures. Differential enhancement within heterogeneous lesions is demonstrated with corroboration from a basic reconstruction using a strict 10-second temporal footprint. Rapid acquisition, reliable fat suppression, and high spatiotemporal resolution are presented, despite significant data undersampling.

CONCLUSION

The STELLR reconstruction approach of 3D radial sampling with mask subtraction provides a high-performance imaging technique for characterizing enhancing structures within the breast. It is capable of maintaining temporal fidelity, while visualizing breast lesions with high detail over a large FOV to include both breasts.

Optoacoustic imaging at kilohertz volumetric frame rates

State-of-the-art optoacoustic tomographic imaging systems have been shown to attain three-dimensional (3D) frame rates of the order of 100 Hz. While such a high volumetric imaging speed is beyond reach for other bio-imaging modalities, it may still be insufficient to accurately monitor some faster events occurring on a millisecond scale. Increasing the 3D imaging rate is usually hampered by the limited throughput capacity of the data acquisition electronics and memory used to capture vast amounts of the generated optoacoustic (OA) data in real time. Herein, we developed a sparse signal acquisition scheme and a total-variation-based reconstruction approach in a combined space-time domain in order to achieve 3D OA imaging at kilohertz rates. By continuous monitoring of freely swimming zebrafish larvae in a 3D region, we demonstrate that the new approach enables significantly increasing the volumetric imaging rate by using a fraction of the tomographic projections without compromising the reconstructed image quality. The suggested method may benefit studies looking at ultrafast biological phenomena in 3D, such as large-scale neuronal activity, cardiac motion, or freely behaving organisms.

A comparison of linear interpolation models for iterative CT reconstruction

PURPOSE

Recent reports indicate that model-based iterative reconstruction methods may improve image quality in computed tomography (CT). One difficulty with these methods is the number of options available to implement them, including the selection of the forward projection model and the penalty term. Currently, the literature is fairly scarce in terms of guidance regarding this selection step, whereas these options impact image quality. Here, the authors investigate the merits of three forward projection models that rely on linear interpolation: the distance-driven method, Joseph's method, and the bilinear method. The authors' selection is motivated by three factors: (1) in CT, linear interpolation is often seen as a suitable trade-off between discretization errors and computational cost, (2) the first two methods are popular with manufacturers, and (3) the third method enables assessing the importance of a key assumption in the other methods.

METHODS

One approach to evaluate forward projection models is to inspect their effect on discretized images, as well as the effect of their transpose on data sets, but significance of such studies is unclear since the matrix and its transpose are always jointly used in iterative reconstruction. Another approach is to investigate the models in the context they are used, i.e., together with statistical weights and a penalty term. Unfortunately, this approach requires the selection of a preferred objective function and does not provide clear information on features that are intrinsic to the model. The authors adopted the following two-stage methodology. First, the authors analyze images that progressively include components of the singular value decomposition of the model in a reconstructed image without statistical weights and penalty term. Next, the authors examine the impact of weights and penalty on observed differences.

RESULTS

Image quality metrics were investigated for 16 different fan-beam imaging scenarios that enabled probing various aspects of all models. The metrics include a surrogate for computational cost, as well as bias, noise, and an estimation task, all at matched resolution. The analysis revealed fundamental differences in terms of both bias and noise. Task-based assessment appears to be required to appreciate the differences in noise; the estimation task the authors selected showed that these differences balance out to yield similar performance. Some scenarios highlighted merits for the distance-driven method in terms of bias but with an increase in computational cost. Three combinations of statistical weights and penalty term showed that the observed differences remain the same, but strong edge-preserving penalty can dramatically reduce the magnitude of these differences.

CONCLUSIONS

In many scenarios, Joseph's method seems to offer an interesting compromise between cost and computational effort. The distance-driven method offers the possibility to reduce bias but with an increase in computational cost. The bilinear method indicated that a key assumption in the other two methods is highly robust. Last, strong edge-preserving penalty can act as a compensator for insufficiencies in the forward projection model, bringing all models to similar levels in the most challenging imaging scenarios. Also, the authors find that their evaluation methodology helps appreciating how model, statistical weights, and penalty term interplay together.

A sparsity-based iterative algorithm for reconstruction of micro-CT images from highly undersampled projection datasets obtained with a synchrotron X-ray source

Synchrotron X-ray Micro Computed Tomography (Micro-CT) is an imaging technique which is increasingly used for non-invasive in vivo preclinical imaging. However, it often requires a large number of projections from many different angles to reconstruct high-quality images leading to significantly high radiation doses and long scan times. To utilize this imaging technique further for in vivo imaging, we need to design reconstruction algorithms that reduce the radiation dose and scan time without reduction of reconstructed image quality. This research is focused on using a combination of gradient-based Douglas-Rachford splitting and discrete wavelet packet shrinkage image denoising methods to design an algorithm for reconstruction of large-scale reduced-view synchrotron Micro-CT images with acceptable quality metrics. These quality metrics are computed by comparing the reconstructed images with a high-dose reference image reconstructed from 1800 equally spaced projections spanning 180°. Visual and quantitative-based performance assessment of a synthetic head phantom and a femoral cortical bone sample imaged in the biomedical imaging and therapy bending magnet beamline at the Canadian Light Source demonstrates that the proposed algorithm is superior to the existing reconstruction algorithms. Using the proposed reconstruction algorithm to reduce the number of projections in synchrotron Micro-CT is an effective way to reduce the overall radiation dose and scan time which improves in vivo imaging protocols.

Single pixel imaging with tunable terahertz parametric oscillator

A method of active terahertz imaging based on compressive sampling is demonstrated. A metal mask structure is designed with all modulation matrices engraved on. The imaging approach based on the mask eliminates the need for imaging object movement in point-wise scanning and shows high sensitivity. A terahertz parametric oscillator with tunability from 0.5 to 2.7 THz was used as the light source. Holes with circular, rectangular, and letter "H" shapes were imaged at 1.75 THz at 20% sampling rate. The influence of sampling rates and averaging times on the image was analyzed. Imaging of the letter "H" at different frequencies from 1.0 to 2.2 THz was tested and evaluated, and recognizable results were obtained in the range of 1.4-2.0 THz.

Morphology supporting function: attenuation correction for SPECT/CT, PET/CT, and PET/MR imaging

Both SPECT, and in particular PET, are unique in medical imaging for their high sensitivity and direct link to a physical quantity, i.e. radiotracer concentration. This gives PET and SPECT imaging unique capabilities for accurately monitoring disease activity for the purposes of clinical management or therapy development. However, to achieve a direct quantitative connection between the underlying radiotracer concentration and the reconstructed image values several confounding physical effects have to be estimated, notably photon attenuation and scatter. With the advent of dual-modality SPECT/CT, PET/CT, and PET/MR scanners, the complementary CT or MR image data can enable these corrections, although there are unique challenges for each combination. This review covers the basic physics underlying photon attenuation and scatter and summarizes technical considerations for multimodal imaging with regard to PET and SPECT quantification and methods to address the challenges for each multimodal combination.