Sparse matrix beamforming and image reconstruction for 2-D HIFU monitoring using harmonic motion imaging for focused ultrasound (HMIFU) with in vitro validation

Beamforming-integrated neural networks for ultrasound imaging

Sparse matrix beamforming (SMB) is a computationally efficient reformulation of delay-and-sum (DAS) beamforming as a single sparse matrix multiplication. This reformulation can potentially dovetail with machine learning platforms like TensorFlow and PyTorch that already support sparse matrix operations. In this work, using SMB principles, we present the development of beamforming-integrated neural networks (BINNs) that can rationally infer ultrasound images directly from pre-beamforming channel-domain radiofrequency (RF) datasets. To demonstrate feasibility, a toy BINN was first designed with two 2D-convolution layers that were respectively placed both before and after an SMB layer. This toy BINN correctly updated kernel weights in all convolution layers, demonstrating efficiency in both training (PyTorch - 133 ms, TensorFlow - 22 ms) and inference (PyTorch - 4 ms, TensorFlow - 5 ms). As an application demonstration, another BINN with two RF-domain convolution layers, an SMB layer, and three image-domain convolution layers was designed to infer high-quality B-mode images in vivo from single-shot plane-wave channel RF data. When trained using 31-angle compounded plane wave images (3000 frames from 22 human volunteers), this BINN showed mean-square logarithmic error improvements of 21.3 % and 431 % in the inferred B-mode image quality respectively comparing to an image-to-image convolutional neural network (CNN) and an RF-to-image CNN with the same number of layers and learnable parameters (3,777). Overall, by including an SMB layer to adopt prior knowledge of DAS beamforming, BINN shows potential as a new type of informed machine learning framework for ultrasound imaging.

Amplitude-Modulation Frequency Optimization for Enhancing Harmonic Motion Imaging Performance of Breast Tumors in the Clinic

OBJECTIVE

Elastography images tissue mechanical responses and infers the underlying properties to aid diagnosis and treatment response monitoring. The estimation of absolute or relative tumor properties may vary with dimensions even when the mechanical properties remain constant. Harmonic motion imaging (HMI) uses amplitude-modulated (AM) focused ultrasound to interrogate the targeted tissue's viscoelastic properties. In this study, effects of AM frequencies on HMI were investigated in terms of inclusion relative stiffness and size estimation.

METHODS

AM frequencies from 200 to 600 Hz in steps of 100 Hz were considered using a 5.3-kPa phantom with cylindrical inclusions (Young's modulus: 22, 31, 44, 56 kPa, and diameter: 4.8, 8.1, 13.6, 19.8 mm) to optimize the performance of HMI in characterizing tumors with the same mechanical properties and of different dimensions.

RESULTS

Consistent displacement ratios (DRs) (17.5% variation) of the inclusion to background were obtained with 200-Hz AM for breast-tumor-mimicking inclusions albeit a suboptimal inclusion size estimation obtained. 400-Hz was otherwise used for small and low-contrast inclusions (4.8 mm, 22 or 31 kPa). A linear relationship (R2 = 0.9043) was found between the inverse DR at these frequencies and the Young's modulus ratio. 400 Hz obtained the most accurate inclusion size estimation with an overall estimation error on the lateral dimension of 0.5 mm. In vivo imaging of breast cancer patients (n = 5) was performed at 200 or 400 Hz.

CONCLUSION

The results presented herein indicate that the HMI AM frequency could be optimized adaptively in cases of different applications, i.e., at 200 or 400 Hz, depending on whether aimed for consistent DR measurement for tumor response assessment or tumor margin delineation for surgical planning. HMI may thus be capable of predicting the pathologic endpoint of tumors in response to neoadjuvant chemotherapy (NACT) as early as 3 weeks into treatment.

An efficient ultrasonic wavenumber-domain plane wave imaging method towards the inspection of curved structures

Ultrasonic phased array testing is commonly employed for inspecting curved structures. Conventional plane wave imaging techniques, based on delay-and-sum in the time-domain, offer high image quality and inspection accuracy but suffer from low frame rates due to their high computational complexity. In this work, an efficient wavenumber-domain imaging method that combines non-stationary wavefield extrapolation and f-k migration is proposed for curved structure inspection. Special emission focal laws are designed to generate a sequence of steered plane waves through the curved interface. The raw data is then extrapolated to the top boundary of the region of interest, followed by f-k migration to reconstruct images with high time efficiency. Simulation and experimental evaluations demonstrate a time reduction by a factor of up to 32.24 compared to conventional time-domain plane wave image reconstruction with equivalent image quality, highlighting its potential for monitoring flaws in real-time.

Real-Time Speed-of-Sound Estimation In Vivo via Steered Plane Wave Ultrasound

Speed-of-sound (SoS) is an intrinsic acoustic property of human tissues and has been regarded as a potential biomarker of tissue health. To foster the clinical use of this emerging biomarker in medical diagnostics, it is important for SoS estimates to be derived and displayed in real time. Here, we demonstrate that concurrent global SoS estimation and B-mode imaging can be achieved live on a portable ultrasound scanner. Our innovation is hinged upon the design of a novel pulse-echo SoS estimation framework that is based on steered plane wave imaging. It has accounted for the effects of refraction and imaging depth when the medium SoS differs from the nominal value of 1540 m/s that is conventionally used in medical imaging. The accuracy of our SoS estimation framework was comparatively analyzed with through-transmit time-of-flight measurements in vitro on 15 custom agar phantoms with different SoS values (1508-1682 m/s) and in vivo on human calf muscles ( N = 9 ; SoS range: 1560-1586 m/s). Our SoS estimation framework has a mean signed difference (MSD) of - 0.6 ± 2.3 m/s in vitro and - 2.2 ± 11.2 m/s in vivo relative to the reference measurements. In addition, our real-time system prototype has yielded simultaneous SoS estimates and B-mode imaging at an average frame rate of 18.1 fps. Overall, by realizing real-time tissue SoS estimation with B-mode imaging, our innovation can foster the use of tissue SoS as a biomarker in medical ultrasound diagnostics.

Optimization of the Tracking Beam Sequence in Harmonic Motion Imaging

Harmonic motion imaging (HMI) is an ultrasound elastography technique that estimates the viscoelastic properties of tissues by inducing localized oscillatory motion using focused ultrasound (FUS). The resulting displacement, assumed to be inversely proportional to the tissue local stiffness, is estimated using an imaging array based on RF speckle tracking. In conventional HMI, this is accomplished with plane-wave (PW) imaging, which inherently suffers from low lateral resolution. Coherent PW compounding (PWC) leverages spatial and temporal resolution using synthetic focusing in transmit. In this study, we introduced focused imaging with parallel tracking in HMI and compared parallel tracking of various transmit F-numbers (F/2.6, 3, 4, and 5) qualitatively and quantitatively with PW and PWC imaging at various compounded angle ranges (6°, 12°, and 18°). An in silico model of a 56-kPa spherical inclusion (diameter: 3.6 mm) embedded in a 5.3-kPa background and a 5.3-kPa elastic phantom with cylindrical inclusions (Young's moduli: 22-56 kPa, diameters: 2.0-8.6 mm) were imaged to assess different tracking beam sequences. Speckle biasing in displacement estimation associated with parallel tracking was also investigated and concluded to be negligible in HMI. Parallel tracking in receive (Rx) resulted in 2%-7% and 8%-12% increase compared to PW imaging ( ) in HMI contrast and contrast-to-noise ratio in silico and phantoms. Focused imaging with parallel tracking in Rx was concluded to be most robust among PW and PWC imaging for displacement estimation, and its preclinical feasibility was demonstrated in postsurgical human cancerous breast tissue specimens and in vivo murine models of breast cancer.

Localized motion imaging for monitoring HIFU therapy: Comparison of modulating frequencies and utilization of square modulating wave

High-intensity focused ultrasound (HIFU) has been successfully used as a minimally invasive cancer therapy method. For monitoring the therapy, the amplitude-modulated (AM) localized motion imaging (LMI) method had been proposed. This paper compares the performance of AM-LMI while using different sine modulating wave frequencies and proposes the utilization of square modulating waves to gain the advantages of both high and low modulating frequencies. A single element therapy transducer with a 2 MHz central frequency was driven by sine modulating waves with different frequencies (approximate 34, 67, 102, 168, and 201 Hz) and by square modulating waves with two frequencies (34 and 67 Hz). An imaging probe with a 5 MHz central frequency and a 20 MHz sampling frequency was mounted in the center hole of the therapy transducer to acquire pulse-echo data, which were used to estimate the tissue oscillation amplitude induced by the acoustic radiation force of the HIFU beam. The decrease ratio of the oscillation amount was then utilized to estimate the coagulated lesion length during the therapy. The comparison of modulating frequencies demonstrated that a higher frequency could bring higher sensitivity to small lesions, while a lower frequency not only gives greater noise robustness but also promotes the ability to estimate lengths of larger lesions. The utilization of a square modulating wave demonstrated its utility to produce tissue oscillation with multiple frequencies and gain the advantages of both high and low modulating frequencies.

In Vivo assessment of hypertensive nephrosclerosis using ultrasound localization microscopy

PURPOSE

As a typical chronic kidney disease (CKD), hypertensive nephrosclerosis (HN) is a common syndrome of hypertension, characterized by chronic kidney microvascular damage. Early diagnosis of microvascular damage using conventional ultrasound imaging encounters challenges in sensitivity and specificity owing to the inherent diffraction limit. Ultrasound localization microscopy (ULM) has been developed to obtain microvasculature and microvascular hemodynamics within the kidney, and would be a promising tool for early diagnosis of CKD.

METHODS

In this study, the advantage of quantitative indexes obtained by using ULM (mean arterial blood flow speeds of different segments of interlobular arteries) over indexes obtained using conventional clinical serum (β2-microglobulin, serum urea nitrogen and creatinine) and urine (24-hour urine volume and urine protein) tests and ultrasound Doppler imaging [peak systolic velocity (PSV), end-diastolic velocity (EDV) and resistance index (RI)] and contrast-enhanced ultrasound imaging [CEUS; rise time (RT), peak intensity (IMAX), mean transit time (mTT) and area under the time-intensity curve (AUC)] for early diagnosis of HN was investigated. Examinations were carried out on 6 spontaneously hypertensive rats (SHR) and 5 normal Wistar-Kyoto (WKY) rats at the age of 10 weeks.

RESULTS

The experimental results showed that the indicators derived from conventional clinical inspections (serum and urine tests) and ultrasound imaging (PSV, EDV, RI, RT, IMAX, mTT and AUC) did not show significant difference between hypertensive and healthy rats (p > 0.05), while the TTP of the SHR group (28.52 ± 5.52 s) derived from CEUS is significantly higher than that of the WKY group (18.68 ± 7.32 s; p < 0.05). The mean blood flow speed in interlobular artery of SHR (12.47 ± 1.06 mm/s) derived from ULM is significantly higher than that of WKY rats (10.13 ± 1.17 mm/s; p < 0.01).

CONCLUSION

The advantages of ULM over conventional clinical inspections and ultrasound imaging methods for early diagnosis of HN were validated. The quantitative results showed that ULM can effectively diagnose HN at the early stage by detecting the blood flow speed changes of interlobular arteries. ULM may promise a reliable technique for early diagnosis of HN in the future. This article is protected by copyright. All rights reserved.

So you think you can DAS? A viewpoint on delay-and-sum beamforming

Delay-and-sum (DAS) is the most widespread digital beamformer in high-frame-rate ultrasound imaging. Its implementation is simple and compatible with real-time applications. In this viewpoint article, we describe the fundamentals of DAS beamforming. The underlying theory and numerical approach are detailed so that users can be aware of its functioning and limitations. In particular, we discuss the importance of the f-number and speed of sound on image quality, and propose one solution to set their values from a physical viewpoint. We suggest determining the f-number from the directivity of the transducer elements and the speed of sound from the phase dispersion of the delayed signals. Simplified Matlab codes are provided for the sake of clarity and openness. The effect of the f-number and speed of sound on the lateral resolution and contrast-to-noise ratio was investigated in vitro and in vivo. If not properly preset, these two factors had a substantial negative impact on standard metrics of image quality (namely CNR and FWHM). When beamforming with DAS in vitro or in vivo, it is recommended to optimize these parameters in order to use it wisely and prevent image degradation.

Ultrasound image reconstruction from plane wave radio-frequency data by self-supervised deep neural network

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.

An analytical model of full-field displacement and strain induced by amplitude-modulated focused ultrasound in harmonic motion imaging

The majority of disease processes involves changes in the micro-structure of the affected tissue, which can translate to changes in the mechanical properties of the corresponding tissue. Harmonic motion imaging (HMI) is an elasticity imaging technique that allows the study of the mechanical parameters of tissue by detecting the tissue response by a harmonic motion field, which is generated by oscillatory acoustic radiation force (ARF). HMI has been demonstrated in tumor detection and characterization as well as monitoring of ablation procedures. In this study, an analytical HMI model is demonstrated and compared with a finite element model (FEM), allowing rapid and accurate computation of the displacement, strain, and shear wave velocity (SWV) at any location in a homogeneous linear elastic material. Average absolute differences between the analytical model and the FEM were respectively 1.2 % for the displacements and 0.5 % for the strains for 41940 force voxels at 0.22 seconds per displacement evaluation. A convergence study showed that the average difference could be further decreased to 1.0 % and 0.15 % for the displacements and strains, respectively, if force resolution is increased. SWV fields, as calculated with the FEM and the analytical model, have regional differences in velocities up to 0.57 m/s with an average absolute difference of 0.11±0.07 m/s, primarily due to imperfections in the non-reflecting FEM boundary conditions. The apparent SWV differed from the commonly used plane-wave approximation by up to 1.2 m/s due to near and intermediate field effects. Maximum displacement amplitudes for a model with an inclusion stabilize within 10 % of the homogeneous model at an inclusion radius of 10 mm while the maximum strain reacts faster, stabilizing at an inclusion radius of 3 mm. In conclusion, an analytical model for HMI stiffness estimation is presented in this paper. The analytical model has advantages over FEM as the full-field displacements do not need to be calculated to evaluate the model at a single measurement point. This advantage, together with the computational speed, makes the analytical model useful for real-time imaging applications. However, the analytical model was found to have restrictive assumptions on tissue homogeneity and infinite dimensions, while the FEM approaches were shown adaptable to variable geometry and non-homogeneous properties.

Real-Time Passive Acoustic Mapping Using Sparse Matrix Multiplication

Passive acoustic mapping enables the spatiotemporal monitoring of cavitation with circulating microbubbles during focused ultrasound (FUS)-mediated blood-brain barrier opening. However, the computational load for processing large data sets of cavitation maps or more complex algorithms limit the visualization in real-time for treatment monitoring and adjustment. In this study, we implemented a graphical processing unit (GPU)-accelerated sparse matrix-based beamforming and time exposure acoustics in a neuronavigation-guided ultrasound system for real-time spatiotemporal monitoring of cavitation. The system performance was tested in silico through benchmarking, in vitro using nonhuman primate (NHP) and human skull specimens, and demonstrated in vivo in NHPs. We demonstrated the stability of the cavitation map for integration times longer than 62.5 [Formula: see text]. A compromise between real-time displaying and cavitation map quality obtained from beamformed RF data sets with a size of 2000 ×128 ×30 (axial [Formula: see text]) was achieved for an integration time of [Formula: see text], which required a computational time of 0.27 s (frame rate of 3.7 Hz) and could be displayed in real-time between pulses at PRF = 2 Hz. Our benchmarking tests show that the GPU sparse-matrix algorithm processed the RF data set at a computational rate of [Formula: see text]/pixel/sample, which enables adjusting the frame rate and the integration time as needed. The neuronavigation system with real-time implementation of cavitation mapping facilitated the localization of the cavitation activity and helped to identify distortions due to FUS phase aberration. The in vivo test of the method demonstrated the feasibility of GPU-accelerated sparse matrix computing in a close to a clinical condition, where focus distortions exemplify problems during treatment. These experimental conditions show the need for spatiotemporal monitoring of cavitation with real-time capability that enables the operator to correct or halt the sonication in case substantial aberrations are observed.

Deep Learning for Ultrasound Localization Microscopy

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 ( [Formula: see text] in the lateral direction and [Formula: see text] 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 [Formula: see text] can be detected and adjacent microvessels with a distance of [Formula: see text] 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.

Noninvasive Young's modulus visualization of fibrosis progression and delineation of pancreatic ductal adenocarcinoma (PDAC) tumors using Harmonic Motion Elastography (HME) in vivo

Background and aims: Poor specificity and predictive values of current cross-sectional radiological imaging methods in evaluation of pancreatic adenocarcinoma (PDAC) limit the clinical capability to accurately stage the tumor pre-operatively and provide optimal surgical treatment and improve patient outcomes.

Methods: In this study, we applied Harmonic Motion Elastography (HME), a quantitative ultrasound-based imaging method to calculate Young's modulus (YM) in PDAC mouse models (n = 30) and human pancreatic resection specimens of PDAC (n=32). We compared the YM to the collagen assessment by Picrosirius red (PSR) stain on corresponding histologic sections.

Results: HME is capable of differentiating between different levels of fibrosis in transgenic mice. In mice without pancreatic fibrosis, the measured YM was 4.2 ± 1.3 kPa, in fibrotic murine pancreata, YM was 5.5 ± 2.0 kPa and in murine PDAC tumors, YM was 11.3 ± 1.7 kPa. The corresponding PSR values were 2.0 ± 0.8 %, 9.8 ± 3.4 %, and 13.2 ± 1.2%, respectively. In addition, three regions within each human surgical PDAC specimen were assessed: tumor, which had both the highest Young's modulus (YM > 40 kPa) and collagen density (PSR > 40 %); non-neoplastic adjacent pancreas, which had the lowest Young's modulus (YM < 15 kPa) and collagen density (PSR < 10%) and a transitional peri-lesional region between the tumor and non-neoplastic pancreas with an intermediate value of measured Young's modulus (15 kPa < YM < 40 kPa) and collagen density (15% < PSR < 35 %).

Conclusion: In conclusion, a non-invasive, quantitative imaging tool for detecting, staging and delineating PDAC tumor margins based on the change in collagen density was developed.

Harmonic Motion Imaging of Pancreatic Tumor Stiffness Indicates Disease State and Treatment Response

PURPOSE

Pancreatic ductal adenocarcinoma (PDA) is a common, deadly cancer that is challenging both to diagnose and to manage. Its hallmark is an expansive, desmoplastic stroma characterized by high mechanical stiffness. In this study, we sought to leverage this feature of PDA for two purposes: differential diagnosis and monitoring of response to treatment.

EXPERIMENTAL DESIGN

Harmonic motion imaging (HMI) is a functional ultrasound technique that yields a quantitative relative measurement of stiffness suitable for comparisons between individuals and over time. We used HMI to quantify pancreatic stiffness in mouse models of pancreatitis and PDA as well as in a series of freshly resected human pancreatic cancer specimens.

RESULTS

In mice, we learned that stiffness increased during progression from preneoplasia to adenocarcinoma and also effectively distinguished PDA from several forms of pancreatitis. In human specimens, the distinction of tumors versus adjacent pancreatitis or normal pancreas tissue was even more stark. Moreover, in both mice and humans, stiffness increased in proportion to tumor size, indicating that tuning of mechanical stiffness is an ongoing process during tumor progression. Finally, using a brca2-mutant mouse model of PDA that is sensitive to cisplatin, we found that tissue stiffness decreases when tumors respond successfully to chemotherapy. Consistent with this observation, we found that tumor tissues from patients who had undergone neoadjuvant therapy were less stiff than those of untreated patients.

CONCLUSIONS

These findings support further development of HMI for clinical applications in disease staging and treatment response assessment in PDA.

Technical Note: In vivo Young's modulus mapping of pancreatic ductal adenocarcinoma during HIFU ablation using harmonic motion elastography (HME)

PURPOSE

Noninvasive quantitative assessment of coagulated tissue during high-intensity focused ultrasound (HIFU) ablation is one of the essential steps for tumor treatment, especially in such cases as the Pancreatic Ductal Adenocarcinoma (PDA) that has low probability of diagnosis at the early stages and high probability of forming solid carcinomas resistant to chemotherapy treatment at the late stages.

METHODS

Harmonic motion elastography (HME) is a technique for the localized estimation of tumor stiffness. This harmonic motion imaging (HMI)-based technique is designed to map the tissue Young's modulus or stiffness noninvasively. A focused ultrasound (FUS) transducer generates an oscillating, acoustic radiation force in its focal region. The two-dimensional (2D) shear wave speed, and consequently the Young's modulus maps, is generated by tracking the radio frequency (RF) signals acquired at high frame rates. By prolonging the sonication for more than 50 s using the same methodology, the 2D Young's modulus maps are reconstructed while HIFU is applied and ablation is formed on PDA murine tumors.

RESULTS

The feasibility of this technique in measuring the regional Young's modulus was first assessed in tissue-mimicking phantoms. The contrast-to-noise ratio (CNR) was found to be higher than 11.7 dB for each 2D reconstructed Young's modulus map. The mean error in this validation study was found to be equal to less than 19%. Then HME was applied on two transgenic mice with pancreatic ductal adenocarcinoma tumors. The Young's modulus median value of this tumor at the start of the HIFU application was equal to 2.1 kPa while after 45 s of sonication it was found to be approximately three times stiffer (6.7 kPa).

CONCLUSIONS

The HME was described herein and showed its capability of measuring tissue stiffness noninvasively by measuring the shear wave speed propagation inside the tissue and reconstructing a 2D Young's modulus map. Application of the methodology in vivo and during HIFU were thus reported here for the first time.

Spatiotemporal matrix image formation for programmable ultrasound scanners

As programmable ultrasound scanners become more common in research laboratories, it is increasingly important to develop robust software-based image formation algorithms that can be obtained in a straightforward fashion for different types of probes and sequences with a small risk of error during implementation. In this work, we argue that as the computational power keeps increasing, it is becoming practical to directly implement an approximation to the matrix operator linking reflector point targets to the corresponding radiofrequency signals via thoroughly validated and widely available simulations software. Once such a spatiotemporal forward-problem matrix is constructed, standard and thus highly optimized inversion procedures can be leveraged to achieve very high quality images in real time. Specifically, we show that spatiotemporal matrix image formation produces images of similar or enhanced quality when compared against standard delay-and-sum approaches in phantoms and in vivo, and show that this approach can be used to form images even when using non-conventional probe designs for which adapted image formation algorithms are not readily available.

Fast lesion mapping during HIFU treatment using harmonic motion imaging guided focused ultrasound (HMIgFUS) in vitro and in vivo

The successful clinical application of high intensity focused ultrasound (HIFU) ablation depends on reliable monitoring of the lesion formation. Harmonic motion imaging guided focused ultrasound (HMIgFUS) is an ultrasound-based elasticity imaging technique, which monitors HIFU ablation based on the stiffness change of the tissue instead of the echo intensity change in conventional B-mode monitoring, rendering it potentially more sensitive to lesion development. Our group has shown that predicting the lesion location based on the radiation force-excited region is feasible during HMIgFUS. In this study, the feasibility of a fast lesion mapping method is explored to directly monitor the lesion map during HIFU. The harmonic motion imaging (HMI) lesion map was generated by subtracting the reference HMI image from the present HMI peak-to-peak displacement map, as streamed on the computer display. The dimensions of the HMIgFUS lesions were compared against gross pathology. Excellent agreement was found between the lesion depth (r ²  =  0.81, slope  =  0.90), width (r ²  =  0.85, slope  =  1.12) and area (r ²  =  0.58, slope  =  0.75). In vivo feasibility was assessed in a mouse with a pancreatic tumor. These findings demonstrate that HMIgFUS can successfully map thermal lesions and monitor lesion development in real time in vitro and in vivo. The HMIgFUS technique may therefore constitute a novel clinical tool for HIFU treatment monitoring.

Mechanics of ultrasound elastography

Ultrasound elastography enables in vivo measurement of the mechanical properties of living soft tissues in a non-destructive and non-invasive manner and has attracted considerable interest for clinical use in recent years. Continuum mechanics plays an essential role in understanding and improving ultrasound-based elastography methods and is the main focus of this review. In particular, the mechanics theories involved in both static and dynamic elastography methods are surveyed. They may help understand the challenges in and opportunities for the practical applications of various ultrasound elastography methods to characterize the linear elastic, viscoelastic, anisotropic elastic and hyperelastic properties of both bulk and thin-walled soft materials, especially the in vivo characterization of biological soft tissues.

3D Quasi-Static Ultrasound Elastography With Plane Wave In Vivo

In biological tissue, an increase in elasticity is often a marker of abnormalities. Techniques such as quasi-static ultrasound elastography have been developed to assess the strain distribution in soft tissues in two dimensions using a quasi-static compression. However, as abnormalities can exhibit very heterogeneous shapes, a three dimensional approach would be necessary to accurately measure their volume and remove operator dependency. Acquisition of volumes at high rates is also critical to performing real-time imaging with a simple freehand compression. In this study, we developed for the first time a 3D quasi-static ultrasound elastography method with plane waves that estimates axial strain distribution in vivo in entire volumes at high volume rate. Acquisitions were performed with a 2D matrix array probe of 2.5 MHz frequency and 256 elements. Plane waves were emitted at a volume rate of 100 volumes/s during a continuous motorized and freehand compression. 3D B-mode volumes and 3D cumulative axial strain volumes were successfully estimated in inclusion phantoms and in ex vivo canine liver before and after a high intensity focused ultrasound ablation. We also demonstrated the in vivo feasibility of the method using freehand compression on the calf muscle of a human volunteer and were able to retrieve 3D axial strain volume at a high volume rate depicting the differences in stiffness of the two muscles which compose the calf muscle. 3D ultrasound quasi-static elastography with plane waves could become an important technique for the imaging of the elasticity in human bodies in three dimensions using simple freehand scanning.

Elasticity mapping of murine abdominal organs in vivo using harmonic motion imaging (HMI)

Recently, ultrasonic imaging of soft tissue mechanics has been increasingly studied to image otherwise undetectable pathologies. However, many underlying mechanisms of tissue stiffening remain unknown, requiring small animal studies and adapted elasticity mapping techniques. Harmonic motion imaging (HMI) assesses tissue viscoelasticity by inducing localized oscillation from a periodic acoustic radiation force. The objective of this study was to evaluate the feasibility of HMI for in vivo elasticity mapping of abdominal organs in small animals. Pathological cases, i.e. chronic pancreatitis and pancreatic cancer, were also studied in vivo to assess the capability of HMI for detection of the change in mechanical properties. A 4.5 MHz focused ultrasound transducer (FUS) generated an amplitude-modulated beam resulting in 50 Hz harmonic tissue oscillations at its focus. Axial tissue displacement was estimated using 1D-cross-correlation of RF signals acquired with a 7.8 MHz diagnostic transducer confocally aligned with the FUS. In vitro results in canine liver and kidney showed the correlation between HMI displacement and Young's moduli measured by rheometry compression testing. HMI was capable of providing reproducible elasticity maps of the mouse abdominal region in vivo allowing the identification of, from stiffest to softest, the murine kidney, pancreas, liver, and spleen. Finally, pancreata affected by pancreatitis and pancreatic cancer showed HMI displacements 1.7 and 2.2 times lower than in the control case, respectively, indicating higher stiffness. The HMI displacement amplitude was correlated with the extent of fibrosis as well as detecting the very onset of stiffening even before fibrosis could be detected on H&E. This work shows that HMI can produce reliable elasticity maps of mouse abdominal region in vivo, thus providing a potentially critical tool to assess pathologies affecting organ elasticity.

Tumor characterization and treatment monitoring of postsurgical human breast specimens using harmonic motion imaging (HMI)

BACKGROUND

High-intensity focused ultrasound (HIFU) is a noninvasive technique used in the treatment of early-stage breast cancer and benign tumors. To facilitate its translation to the clinic, there is a need for a simple, cost-effective device that can reliably monitor HIFU treatment. We have developed harmonic motion imaging (HMI), which can be used seamlessly in conjunction with HIFU for tumor ablation monitoring, namely harmonic motion imaging for focused ultrasound (HMIFU). The overall objective of this study was to develop an all ultrasound-based system for real-time imaging and ablation monitoring in the human breast in vivo.

METHODS

HMI was performed in 36 specimens (19 normal, 15 invasive ductal carcinomas, and 2 fibroadenomas) immediately after surgical removal. The specimens were securely embedded in a tissue-mimicking agar gel matrix and submerged in degassed phosphate-buffered saline to mimic in vivo environment. The HMI setup consisted of a HIFU transducer confocally aligned with an imaging transducer to induce an oscillatory radiation force and estimate the resulting displacement.

RESULTS

3D HMI displacement maps were reconstructed to represent the relative tissue stiffness in 3D. The average peak-to-peak displacement was found to be significantly different (p = 0.003) between normal breast tissue and invasive ductal carcinoma. There were also significant differences before and after HMIFU ablation in both the normal (53.84 % decrease) and invasive ductal carcinoma (44.69 % decrease) specimens.

CONCLUSIONS

HMI can be used to map and differentiate relative stiffness in postsurgical normal and pathological breast tissues. HMIFU can also successfully monitor thermal ablations in normal and pathological human breast specimens. This HMI technique may lead to a new clinical tool for breast tumor imaging and HIFU treatment monitoring.

Real-time Monitoring of High Intensity Focused Ultrasound (HIFU) Ablation of In Vitro Canine Livers Using Harmonic Motion Imaging for Focused Ultrasound (HMIFU)

Harmonic Motion Imaging for Focused Ultrasound (HMIFU) is a technique that can perform and monitor high-intensity focused ultrasound (HIFU) ablation. An oscillatory motion is generated at the focus of a 93-element and 4.5 MHz center frequency HIFU transducer by applying a 25 Hz amplitude-modulated signal using a function generator. A 64-element and 2.5 MHz imaging transducer with 68kPa peak pressure is confocally placed at the center of the HIFU transducer to acquire the radio-frequency (RF) channel data. In this protocol, real-time monitoring of thermal ablation using HIFU with an acoustic power of 7 W on canine livers in vitro is described. HIFU treatment is applied on the tissue during 2 min and the ablated region is imaged in real-time using diverging or plane wave imaging up to 1,000 frames/second. The matrix of RF channel data is multiplied by a sparse matrix for image reconstruction. The reconstructed field of view is of 90° for diverging wave and 20 mm for plane wave imaging and the data are sampled at 80 MHz. The reconstruction is performed on a Graphical Processing Unit (GPU) in order to image in real-time at a 4.5 display frame rate. 1-D normalized cross-correlation of the reconstructed RF data is used to estimate axial displacements in the focal region. The magnitude of the peak-to-peak displacement at the focal depth decreases during the thermal ablation which denotes stiffening of the tissue due to the formation of a lesion. The displacement signal-to-noise ratio (SNRd) at the focal area for plane wave was 1.4 times higher than for diverging wave showing that plane wave imaging appears to produce better displacement maps quality for HMIFU than diverging wave imaging.

Harmonic motion imaging for abdominal tumor detection and high-intensity focused ultrasound ablation monitoring: an in vivo feasibility study in a transgenic mouse model of pancreatic cancer

Harmonic motion imaging (HMI) is a radiationforce- based elasticity imaging technique that tracks oscillatory tissue displacements induced by sinusoidal ultrasonic radiation force to assess the resulting oscillatory displacement denoting the underlying tissue stiffness. The objective of this study was to evaluate the feasibility of HMI in pancreatic tumor detection and high-intensity focused ultrasound (HIFU) treatment monitoring. The HMI system consisted of a focused ultrasound transducer, which generated sinusoidal radiation force to induce oscillatory tissue motion at 50 Hz, and a diagnostic ultrasound transducer, which detected the axial tissue displacements based on acquired radio-frequency signals using a 1-D cross-correlation algorithm. For pancreatic tumor detection, HMI images were generated for pancreatic tumors in transgenic mice and normal pancreases in wild-type mice. The obtained HMI images showed a high contrast between normal and malignant pancreases with an average peak-to-peak HMI displacement ratio of 3.2. Histological analysis showed that no tissue damage was associated with HMI when it was used for the sole purpose of elasticity imaging. For pancreatic tumor ablation monitoring, the focused ultrasound transducer was operated at a higher acoustic power and longer pulse length than that used in tumor detection to simultaneously induce HIFU thermal ablation and oscillatory tissue displacements, allowing HMI monitoring without interrupting tumor ablation. HMI monitoring of HIFU ablation found significant decreases in the peak-to-peak HMI displacements before and after HIFU ablation with a reduction rate ranging from 15.8% to 57.0%. The formation of thermal lesions after HIFU exposure was confirmed by histological analysis. This study demonstrated the feasibility of HMI in abdominal tumor detection and HIFU ablation monitoring.

High intensity focused ultrasound (HIFU) focal spot localization using harmonic motion imaging (HMI)

Several ultrasound-based imaging modalities have been proposed for image guidance and monitoring of high-intensity focused ultrasound (HIFU) treatment. However, accurate localization and characterization of the effective region of treatment (focal spot) remain important obstacles in the clinical implementation of HIFU ablation. Harmonic motion imaging for focused ultrasound (HMIFU) is a HIFU monitoring technique that utilizes radiation-force-induced localized oscillatory displacement. HMIFU has been shown to correctly identify the formation and extent of HIFU thermal ablation lesions. However a significant problem remains in identifying the location of the HIFU focus, which is necessary for treatment planning. In this study, the induced displacement was employed to localize the HIFU focal spot inside the tissue prior to treatment. Feasibility was shown with two separate systems. The 1D HMIFU system consisted of a HIFU transducer emitting an amplitude-modulated HIFU beam for mechanical excitation and a confocal single-element, pulse-echo transducer for simultaneous RF acquisition. The 2D HIFU system consists of a HIFU phased array, and a co-axial imaging phased array for simultaneous imaging. Initial feasibility was first performed on tissue-mimicking gelatin phantoms and the focal zone was defined as the region corresponding to the -3dB full width at half maximum of the HMI displacement. Using the same parameters, in vitro experiments were performed in canine liver specimens to compare the defined focal zone with the lesion. In vitro measurements showed good agreement between the HMI predicted focal zone and the induced HIFU lesion location. HMIFU was experimentally shown to be capable of predicting and tracking the focal region in both phantoms and in vitro tissues. The accuracy of focal spot localization was evaluated by comparing with the lesion location in post-ablative tissues, with a R(2) = 0.821 at p < 0.002 in the 2D HMI system. We demonstrated the feasibility of using this HMI-based technique to localize the HIFU focal spot without inducing thermal changes during the planning phase. The focal spot localization method has also been applied on ex vivo human breast tissue ablation and can be fully integrated into any HMI system for planning purposes.

High-intensity focused ultrasound monitoring using harmonic motion imaging for focused ultrasound (HMIFU) under boiling or slow denaturation conditions

Harmonic motion imaging for focused ultrasound (HMIFU) is a recently developed high-intensity focused ultrasound (HIFU) treatment monitoring method that utilizes an amplitude-modulated therapeutic ultrasound beam to induce an oscillatory radiation force at the HIFU focus and estimates the focal tissue displacement to monitor the HIFU thermal treatment. In this study, the performance of HMIFU under acoustic, thermal, and mechanical effects was investigated. The performance of HMIFU was assessed in ex vivo canine liver specimens (n = 13) under slow denaturation or boiling regimes. A passive cavitation detector (PCD) was used to assess the acoustic cavitation activity, and a bare-wire thermocouple was used to monitor the focal temperature change. During lesioning with slow denaturation, high quality displacements (correlation coefficient above 0.97) were observed under minimum cavitation noise, indicating the tissue initial-softening-then- stiffening property change. During HIFU with boiling, HMIFU monitored a consistent change in lesion-to-background displacement contrast (0.46 ± 0.37) despite the presence of strong cavitation noise due to boiling during lesion formation. Therefore, HMIFU effectively monitored softening-then-stiffening during lesioning under slow denaturation, and detected lesioning under boiling with a distinct change in displacement contrast under boiling in the presence of cavitation. In conclusion, HMIFU was shown under both boiling and slow denaturation regimes to be effective in HIFU monitoring and lesioning identification without being significantly affected by cavitation noise.

Thermometry and ablation monitoring with ultrasound

In this review we present the current status of ultrasound thermometry and ablation monitoring, with emphasis on the diverse approaches published in the literature and with an eye on which methods are closest to clinical reality. It is hoped that this review will serve as a guide to the expansion of sonographic methods for treatment monitoring and thermometry since the last brief review in 2007.