Spatiotemporal clutter filtering via robust principal component analysis (rPCA) has been widely used in ultrasound microvascular imaging. However, the performance of the rPCA clutter filtering highly relies on low-rank modeling for tissue signals and sparse modeling for blood flow signals. Moreover, current rPCA clutter filters are typically based on static processing and have to access a batch of beamformed frames for optimization. This prevents these filters from ultrafast realization. This paper adopts the iteratively reweighted least squares (IRLS) rPCA framework to model tissue and blood flow signals for improved clutter filtering. More importantly, the static IRLS-rPCA filter is upgraded to a spatiotemporal-constrained online method to instantaneously extract blood flow signals from the ongoing beamformed frame. Simulations and in-vivo experiments on a contrast-enhanced rat kidney and a contrast-free human liver demonstrated that the IRLS-rPCA clutter filter achieves higher sensitivity, contrast-to-noise ratio (CNR), and signal-to-noise ratio (SNR) than other rPCA methods. Particularly, the static IRLS-rPCA clutter filter obtains more than 2 dB improvements in CNR over the compared methods in the human liver dataset. The proposed online clutter filter achieves comparable image quality to the static version and processing time of $0.028~\pm ~0.004$ seconds per frame. The corresponding acceleration factor of the online clutter filter over all the tested methods is more than 20.

Photoacoustic computed tomography (PACT) is an emerging hybrid imaging modality with potential applications in biomedicine. A major roadblock to the widespread adoption of PACT is the limited number of detectors, which gives rise to spatial aliasing and manifests as streak artifacts in the reconstructed image. A brute-force solution to the problem is to increase the number of detectors, which, however, is often undesirable due to escalated costs. In this study, we present a novel self-supervised learning approach, to overcome this long-standing challenge. We found that small blocks of PACT channel data show similarity at various downsampling rates. Based on this observation, a neural network trained on downsampled data can reliably perform accurate interpolation without requiring densely-sampled ground truth data, which is typically unavailable in real practice. Our method has undergone validation through numerical simulations, controlled phantom experiments, as well as ex vivo and in vivo animal tests, across multiple PACT systems. We have demonstrated that our technique provides an effective and cost-efficient solution to address the under-sampling issue in PACT, thereby enhancing the capabilities of this imaging technology.

Histological image registration is a fundamental task in histological image analysis. It is challenging because of substantial appearance differences due to multiple staining. Keypoint correspondences, i.e., matched keypoint pairs, have been introduced to guide unsupervised deep learning (DL) based registration methods to handle such a registration task. This paper proposes an iterative keypoint correspondence-guided (IKCG) unsupervised network for non-rigid histological image registration. Fixed deep features and learnable deep features are introduced as keypoint descriptors to automatically establish keypoint correspondences, the distance between which is used as a loss function to train the registration network. Fixed deep features extracted from DL networks that are pre-trained on natural image datasets are more discriminative than handcrafted ones, benefiting from the deep and hierarchical nature of DL networks. The intermediate layer outputs of the registration networks trained on histological image datasets are extracted as learnable deep features, which reveal unique information for histological images. An iterative training strategy is adopted to train the registration network and optimize learnable deep features jointly. Benefiting from the excellent matching ability of learnable deep features optimized with the iterative training strategy, the proposed method can solve the local non-rigid large displacement problem, an inevitable problem usually caused by misoperation, such as tears in producing tissue slices. The proposed method is evaluated on the Automatic Non-rigid Histology Image Registration (ANHIR) website and AutomatiC Registration Of Breast cAncer Tissue (ACROBAT) website. It ranked 1st on both websites as of August 6th, 2024.

Volumetric (3D) ultrasound imaging using a 2D matrix array probe is increasingly developed for various clinical procedures. However, 3D ultrasound imaging suffers from motion artifacts due to tissue motions and a relatively low frame rate. Current Doppler-based motion compensation (MoCo) methods only allow 1D compensation in the in-range dimension. In this work, we propose a new 3D-MoCo framework that combines 3D velocity field estimation and a two-step compensation strategy for 3D diverging wave compounding imaging. Specifically, our framework explores two constraints of a round-trip scan sequence of 3D diverging waves, i.e., Doppler and pair-wise optical flow, to formulate the estimation of the 3D velocity fields as a global optimization problem, which is further regularized by the divergence-free and first-order smoothness. The two-step compensation strategy is to first compensate for the 1D displacements in the in-range dimension and then the 2D displacements in the two mutually orthogonal cross-range dimensions. Systematical in-silico experiments were conducted to validate the effectiveness of our proposed 3D-MoCo method. The results demonstrate that our 3D-MoCo method achieves higher image contrast, higher structural similarity, and better speckle patterns than the corresponding 1D-MoCo method. Particularly, the 2D cross-range compensation is effective for fully recovering image quality.

Registration of multiple stained images is a fundamental task in histological image analysis. In supervised methods, obtaining ground-truth data with known correspondences is laborious and time-consuming. Thus, unsupervised methods are expected. Unsupervised methods ease the burden of manual annotation but often at the cost of inferior results. In addition, registration of histological images suffers from appearance variance due to multiple staining, repetitive texture, and section missing during making tissue sections. To deal with these challenges, we propose an unsupervised structural feature guided convolutional neural network (SFG). Structural features are robust to multiple staining. The combination of low-resolution rough structural features and high-resolution fine structural features can overcome repetitive texture and section missing, respectively. SFG consists of two components of structural consistency constraints according to the formations of structural features, i.e., dense structural component and sparse structural component. The dense structural component uses structural feature maps of the whole image as structural consistency constraints, which represent local contextual information. The sparse structural component utilizes the distance of automatically obtained matched key points as structural consistency constraints because the matched key points in an image pair emphasize the matching of significant structures, which imply global information. In addition, a multi-scale strategy is used in both dense and sparse structural components to make full use of the structural information at low resolution and high resolution to overcome repetitive texture and section missing. The proposed method was evaluated on a public histological dataset (ANHIR) and ranked first as of Jan 18th, 2022.

Image reconstruction from radio-frequency (RF) data is crucial for ultrafast plane wave ultrasound (PWUS) imaging. Compared with the traditional delay-and-sum (DAS) method based on relatively imprecise assumptions, sparse regularization (SR) method directly solves the inverse problem of image reconstruction and has presented significant improvement in the image quality when the frame rate remains high. However, the computational complexity of SR is too high for practical implementation, which is inherently associated with its iterative process. In this work, a deep neural network (DNN), which is trained with an incorporated loss function including sparse regularization terms, is proposed to reconstruct PWUS images from RF data with significantly reduced computational time. It is remarkable that, a self-supervised learning scheme, in which the RF data are utilized as both the inputs and the labels during the training process, is employed to overcome the lack of the "ideal" ultrasound images as the labels for DNN. In addition, it has been also verified that the trained network can be used on the RF data obtained with steered plane waves (PWs), and thus the image quality can be further improved with coherent compounding. Using simulation data, the proposed method has significantly shorter reconstruction time (∼10 ms) than the conventional SR method (∼1-5 mins), with comparable spatial resolution and 1.5-dB higher contrast-to-noise ratio (CNR). Besides, the proposed method with single PW can achieve higher CNR than DAS with 75 PWs in reconstruction of in-vivo images of human carotid arteries.

Two-way dynamic focusing in synthetic transmit aperture (STA) beamforming can benefit high-quality ultrasound imaging with higher lateral spatial resolution and contrast resolution. However, STA requires the complete dataset for beamforming in a relatively low frame rate and transmit power. This paper proposes a deep-learning architecture to achieve high frame rate STA imaging with two-way dynamic focusing. The network consists of an encoder and a joint decoder. The encoder trains a set of binary weights as the apodizations of the high-frame-rate plane wave transmissions. In this respect, we term our network ApodNet. The decoder can recover the complete dataset from the acquired channel data to achieve dynamic transmit focusing. We evaluate the proposed method by simulations at different levels of noise and in-vivo experiments on the human biceps brachii and common carotid artery. The experimental results demonstrate that ApodNet provides a promising strategy for high frame rate STA imaging, obtaining comparable lateral resolution and contrast resolution with four-times higher frame rate than conventional STA imaging in the in-vivo experiments. Particularly, ApodNet improves contrast resolution of the hypoechoic targets with much shorter computational time when compared with other high-frame-rate methods in both simulations and in-vivo experiments.

One primary technical challenge in photoacoustic microscopy (PAM) is the necessary compromise between spatial resolution and imaging speed. In this study, we propose a novel application of deep learning principles to reconstruct undersampled PAM images and transcend the trade-off between spatial resolution and imaging speed. We compared various convolutional neural network (CNN) architectures, and selected a Fully Dense U-net (FD U-net) model that produced the best results. To mimic various undersampling conditions in practice, we artificially downsampled fully-sampled PAM images of mouse brain vasculature at different ratios. This allowed us to not only definitively establish the ground truth, but also train and test our deep learning model at various imaging conditions. Our results and numerical analysis have collectively demonstrated the robust performance of our model to reconstruct PAM images with as few as 2% of the original pixels, which can effectively shorten the imaging time without substantially sacrificing the image quality.

By localizing microbubbles (MBs) in the vasculature, ultrasound localization microscopy (ULM) has recently been proposed, which greatly improves the spatial resolution of ultrasound (US) imaging and will be helpful for clinical diagnosis. Nevertheless, several challenges remain in fast ULM imaging. The main problems are that current localization methods used to implement fast ULM imaging, e.g., a previously reported localization method based on sparse recovery (CS-ULM), suffer from long data-processing time and exhaustive parameter tuning (optimization). To address these problems, in this paper, we propose a ULM method based on deep learning, which is achieved by using a modified sub-pixel convolutional neural network (CNN), termed as mSPCN-ULM. Simulations and in vivo experiments are performed to evaluate the performance of mSPCN-ULM. Simulation results show that even if under high-density condition (6.4 MBs/mm2), a high localization precision (~28 μm in the lateral direction and ~24 μm in the axial direction) and a high localization reliability (Jaccard index of 0.66) can be obtained by mSPCN-ULM, compared to CS-ULM. The in vivo experimental results indicate that with plane wave scan at a transmit center frequency of 15.625 MHz, microvessels with diameters of ~17 μm can be detected and adjacent microvessels with a distance of ~42 μm can be separated. Furthermore, when using GPU acceleration, the data-processing time of mSPCN-ULM can be shortened to ~6 sec/frame in the simulations and ~23 sec/frame in the in vivo experiments, which is 3-4 orders of magnitude faster than CS-ULM. Finally, once the network is trained, mSPCN-ULM does not need parameter tuning to implement ULM. As a result, mSPCN-ULM opens the door to implement ULM with fast data-processing speed, high imaging accuracy, short data-acquisition time, and high flexibility (robustness to parameters) characteristics.

Contrast enhanced ultrasound is a radiation-free imaging modality which uses encapsulated gas microbubbles for improved visualization of the vascular bed deep within the tissue. It has recently been used to enable imaging with unprecedented subwavelength spatial resolution by relying on super-resolution techniques. A typical preprocessing step in super-resolution ultrasound is to separate the microbubble signal from the cluttering tissue signal. This step has a crucial impact on the final image quality. Here, we propose a new approach to clutter removal based on robust principle component analysis (PCA) and deep learning. We begin by modeling the acquired contrast enhanced ultrasound signal as a combination of low rank and sparse components. This model is used in robust PCA and was previously suggested in the context of ultrasound Doppler processing and dynamic magnetic resonance imaging. We then illustrate that an iterative algorithm based on this model exhibits improved separation of microbubble signal from the tissue signal over commonly practiced methods. Next, we apply the concept of deep unfolding to suggest a deep network architecture tailored to our clutter filtering problem which exhibits improved convergence speed and accuracy with respect to its iterative counterpart. We compare the performance of the suggested deep network on both simulations and in-vivo rat brain scans, with a commonly practiced deep-network architecture and with the fast iterative shrinkage algorithm. We show that our architecture exhibits better image quality and contrast.

Segmentation of carotid intima-media (IM) borders from ultrasound sequences is challenging because of unknown image noise and varying IM border morphologies and/or dynamics. In this paper, we have developed a state-space framework to sequentially segment the carotid IM borders in each image throughout the cardiac cycle. In this framework, an ${\mathrm{H}}_{\mathrm{\infty }}$ filter is used to solve the state-space equations, and a grayscale-derivative constraint snake is used to provide accurate measurements for the ${\mathrm{H}}_{\mathrm{\infty }}$ filter. We have evaluated the performance of our approach by comparing our segmentation results to the manually traced contours of ultrasound image sequences of three synthetic models and 156 real subjects from four medical centers. The results show that our method has a small segmentation error (lumen intima, LI: 53 $\pm\, 67\;{\mathrm{\mu }}$m; media-adventitia, MA: 57 $\pm\, 63\;{\mathrm{\mu }}$m) for synthetic and real sequences of different image characteristics, and also agrees well with the manual segmentation (LI: bias = 1.44 ${\mathrm{\mu }}$m; MA: bias = $-$3.38 ${\mathrm{\mu }}$m). Our approach can robustly segment the carotid ultrasound sequences with various IM border morphologies, dynamics, and unknown image noise. These results indicate the potential of our framework to segment IM borders for clinical diagnosis.

A novel beamforming technique, named compressed sensing based synthetic transmit aperture (CS-STA) is proposed to speed up the acquisition of ultrasound imaging. This technique consists of three steps. First, the ultrasound transducer transmits randomly apodized plane waves for a number of times and receives the backscattered echoes. Second, the recorded backscattered echoes are used to recover the full channel dataset of synthetic transmit aperture (STA) with a compressed sensing (CS) reconstruction algorithm. Finally, an STA image is beamformed from the recovered full STA dataset. As CS allows recovering a signal from its few linear measurements with high probability, CS-STA is capable of recovering the STA image with fewer firings (i.e., higher frame rate) and retaining the high resolution of STA. In addition, the contrast of the STA image can be improved at the same time owing to the higher energy of plane wave firing in CS-STA. Simulations demonstrate that CS-STA is capable of recovering the STA channel dataset with a smaller number of firings. The performance of CS-STA is evaluated in phantom experiments through comparisons with STA, multi-element STA, conventional focused mode and coherent plane wave imaging. The results demonstrate that, implemented with the same frame rate, CS-STA achieves higher or comparable resolution and contrast. Moreover, comparisons are conducted on the biceps brachii muscle and thyroid of a human subject, and the results demonstrate the feasibility and competitiveness of CS-STA in the in vivo conditions.

X-ray luminescence computed tomography (XLCT), which aims to achieve molecular and functional imaging by X-rays, has recently been proposed as a new imaging modality. Combining the principles of X-ray excitation of luminescence-based probes and optical signal detection, XLCT naturally fuses functional and anatomical images and provides complementary information for a wide range of applications in biomedical research. In order to improve the data acquisition efficiency of previously developed narrow-beam XLCT, a cone beam XLCT (CB-XLCT) mode is adopted here to take advantage of the useful geometric features of cone beam excitation. Practically, a major hurdle in using cone beam X-ray for XLCT is that the inverse problem here is seriously ill-conditioned, hindering us to achieve good image quality. In this paper, we propose a novel Bayesian method to tackle the bottleneck in CB-XLCT reconstruction. The method utilizes a local regularization strategy based on Gaussian Markov random field to mitigate the ill-conditioness of CB-XLCT. An alternating optimization scheme is then used to automatically calculate all the unknown hyperparameters while an iterative coordinate descent algorithm is adopted to reconstruct the image with a voxel-based closed-form solution. Results of numerical simulations and mouse experiments show that the self-adaptive Bayesian method significantly improves the CB-XLCT image quality as compared with conventional methods.

Color tissue Doppler imaging (TDI) is a well-established methodology to assess local myocardial motion/deformation. Typically, a frame rate of ∼ 200 Hz can be achieved by imaging a narrow sector (∼ 30°, covering one cardiac wall) at moderate line density, using a dedicated pulse sequence and multi-line acquisition. However, a wide angle field-of-view is required in some clinical applications to image the whole left ventricle, which implies a drop in temporal resolution. Hereto, the aim of this study was to propose a novel imaging sequence using a multi-line transmit (MLT) beamforming approach to achieve high frame rate color TDI while preserving a wide field-of-view (i.e., 90° sector). To this end, a color MLT-TDI sequence achieving a frame rate of 208 Hz with a 90°-sector was implemented on an ultrasound experimental scanner interleaved with a conventional color TDI sequence achieving the same frame rate but only with a 22.5°-sector. Using this setup, the septal wall of 9 healthy volunteers was imaged and the corresponding velocity was extracted. The M-mode velocity images and the velocity profiles obtained from the MLT-TDI images presented physiologic patterns, very similar to those from conventional TDI. Moreover, for the peak systolic/diastolic velocities, good agreement and strong correlation between MLT-TDI and conventional TDI were found. The results thus demonstrate the feasibility of the novel MLT based TDI methodology to achieve high frame rate color TDI without compromising the field-of-view. This may open the opportunity to simultaneously assess regional myocardial function of the whole left ventricle at high temporal resolution.

Fluorescence imaging has been successfully used in the study of pharmacokinetic analysis, while dynamic fluorescence molecular tomography (FMT) is an attractive imaging technique for three-dimensionally resolving the metabolic process of fluorescent biomarkers in small animals in vivo. Parametric images obtained by combining dynamic FMT with compartmental modeling can provide quantitative physiological information for biological studies and drug development. However, images obtained with conventional indirect methods suffer from poor image quality because of failure in utilizing the temporal correlations of boundary measurements. Besides, FMT suffers from low spatial resolution due to its ill-posed nature, which further reduces the image quality. In this paper, we propose a novel method to directly reconstruct parametric images from boundary measurements based on maximum a posteriori (MAP) estimation with structural priors in a Bayesian framework. The proposed method can utilize structural priors obtained from an X-ray computed tomography system to mitigate the ill-posedness of dynamic FMT inverse problem, and use direct reconstruction strategy to make full use of temporal correlations of boundary measurements. The results of numerical simulations and in vivo mouse experiments demonstrate that the proposed method leads to significant improvements in the reconstruction quality of parametric images as compared with the conventional indirect method and a previously developed direct method.

Fluorescence molecular tomography (FMT) is an emerging tool in the observation of diseases. A fast and accurate surface reconstruction of the experimental object is needed as a boundary constraint for FMT reconstruction. In this paper, an automatic, noncontact, and 3-D surface reconstruction method named 360◦ Fourier transform profilometry (FTP) is proposed to reconstruct 3-D surface profiles for FMT system. This method can reconstruct 360◦ integrated surface profiles utilizing the single-frame FTP at different angles. Results show that the relative mean error of the surface reconstruction of this method is less than 1.4% in phantom experiments, and is no more than 2.9% in mouse experiments in vivo. Compared with the Radon transform method, the proposed method reduces the computation time by more than 90% with a minimal error increase. At last, a combined 360◦ FTP/FMT experiment is conducted on a nude mouse. Not only can the 360◦ FTP system operate with the FMT system simultaneously, but it can also help to monitor the status of animals. Moreover, the 360◦ FTP system is independent of FMT system and can be performed to reconstruct the surface by itself.

Fluorescence imaging can track the expression of fluorescent protein and18F-fluorodeoxyglucose based positron emission tomography ( [18F]FDG-PET) can evaluate the changes of [18F]FDG uptake in tumor cells during the antitumor treatment. In this work, fluorescence imaging and [18F]FDG-PET were both employed to monitor tumor response to Au nanorod-indocyanine green (AuNR-ICG) conjugates mediated therapy in a subcutaneous MDA-MB-231 mouse xenograft model. Serial fluorescence and [18F]FDG-PET images following the antitumor treatment were obtained and quantitative analysis revealed significant decreases in fluorescence intensity and metabolic activity in tumors treated with AuNR-ICG conjugates under near infrared laser irradiation. The results suggest that the combination of fluorescence and [18F]FDG-PET imaging can provide a noninvasive tool to assess the tumor response to antitumor therapy on a molecular scale.

In this paper, fluorescence molecular tomography (FMT) imaging guided by priors from simultaneous positron emission tomography (PET) was performed on a multi-modality imaging system combining PET and FMT. The target prior information from PET images was employed to the FMT reconstruction procedure using the iteratively reweighted least-squares method. Numerical simulations and phantom experiments were performed to validate the proposed method. The results indicate that incorporating the PET prior information into the FMT reconstruction can potentially improve the spatial resolution of FMT.

Dynamic fluorescence diffuse optical tomography (FDOT) is important for the research of drug delivery, medical diagnosis and treatment. Conventionally, dynamic tomographic images are reconstructed frame by frame, independently. This approach fails to account for the temporal correlations in measurement data. Ideally, the entire image sequence should be considered as a whole and a four-dimensional (4-D) reconstruction should be performed. However, the fully 4-D reconstruction is computationally intensive. In this paper, we propose a new 4-D reconstruction approach for dynamic FDOT, which is achieved by applying a temporal Karhunen-Loève (KL) transformation to the imaging equation. By taking advantage of the decorrelation and compression properties of the KL transformation, the complex 4-D optical reconstruction problem is greatly simplified. To evaluate the performance of the method, simulation, phantom, and in vivo experiments (N=7) are performed on a hybrid FDOT/x-ray computed tomography imaging system. The experimental results indicate that the reconstruction images obtained by the KL method provide good reconstruction quality. Additionally, by discarding high-order KL components, the computation time involved with fully 4-D reconstruction can be greatly reduced in contrast to the conventional frame-by-frame reconstruction.

The abdominal aortic aneurysm (AAA) is a common vascular disease. The current clinical criterion for treating AAAs is an increased diameter above a critical value. However, the maximum diameter does not correlate well with aortic rupture, the main cause of death from AAA disease. AAA disease leads to changes in the aortic wall mechanical properties. The pulse-wave velocity (PWV) may indicate such a change. Because of limitations in temporal and spatial resolution, the widely used foot-to-foot method measures the global, instead of regional, PWV between two points at a certain distance in the circulation. However, mechanical properties are nonuniform along the normal and pathological (e.g., the AAA and atherosclerosis) arteries; thus, such changes are typically regional. Pulse-wave imaging (PWI) has been developed by our group to map the pulse-wave propagation along the abdominal aorta in mice in vivo. By using a retrospective electrocardiogram (ECG) gating technique, the radio-frequency (RF) signals over one cardiac cycle were obtained in murine aortas at the extremely high frame rate of 8 kHz and with a field-of-view (FOV) of 12 x 12 mm(2). The velocities of the aortic wall were estimated using an RF-based speckle tracking method. An Angiotensin II (AngII) infusion-based AAA model was used to simulate the human AAA case. Sequences of wall velocity images can noninvasively and quantitatively map the propagation of the pulse wave along the aortic wall. In the normal and sham aortas, the propagation of the pulse wave was relatively uniform along the wall, while in the AngII-treated aortas, the propagation was shown to be nonuniform. There was no significant difference ( p > 0.05) in the PWV between sham (4.67 +/- 1.15 m/s, n=5) and AngII-treated (4.34 +/- 1.48 m/s, n=17) aortas. The correlation coefficient of the linear regression was significantly higher ( p < 0.005) in the sham aortas (0.89 +/- 0.03, n=5 ) than in the AngII-treated ones (0.61 +/- 0.15, n=17). The wall velocities induced by the pulse wave were lower and the pulse wave moved nonuniformly along the AngII-treated aorta ( p < 0.005), with the lowest velocities at the aneurysmal regions. The discrepancy in the regional wall velocity and the nonuniform pulse-wave propagation along the AngII-treated aorta indicated the inhomogeneities in the aortic wall properties, and the reduced wall velocities indicated stiffening of the aneurysmal wall. This novel technique may thus constitute an early detection tool of vascular degeneration as well as serve as a suitable predictor of AAA rupture, complementary to the current clinical screening practice.