Radio Frequency Data Research Articles

In Vivo Adaptive Bayesian Regularized Lagrangian Carotid Strain Imaging for Murine Carotid Arteries and Its Associations With Histological Findings

ObjectivesNon-invasive methods for monitoring arterial health and identifying early injury to optimize treatment for patients are desirable. The objective of this study was to demonstrate the use of an adaptive Bayesian regularized Lagrangian carotid strain imaging (ABR-LCSI) algorithm for monitoring of atherogenesis in a murine model and examine associations between the ultrasound strain measures and histology. MethodsUltrasound radiofrequency (RF) data were acquired from both the right and left common carotid artery (CCA) of 10 (5 male and 5 female) ApoE tm1Unc/J mice at 6, 16 and 24 wk. Lagrangian accumulated axial, lateral and shear strain images and three strain indices—maximum accumulated strain index (MASI), peak mean strain of full region of interest (ROI) index (PMSRI) and strain at peak axial displacement index (SPADI)—were estimated using the ABR-LCSI algorithm. Mice were euthanized (n = 2 at 6 and 16 wk, n = 6 at 24 wk) for histology examination. ResultsSex-specific differences in strain indices of mice at 6, 16 and 24 wk were observed. For male mice, axial PMSRI and SPADI changed significantly from 6 to 24 wk (mean axial PMSRI at 6 wk = 14.10 ± 5.33% and that at 24 wk = –3.03 ± 5.61%, p < 0.001). For female mice, lateral MASI increased significantly from 6 to 24 wk (mean lateral MASI at 6 wk = 10.26 ± 3.13% and that at 24 wk = 16.42 ± 7.15%, p = 0.048). Both cohorts exhibited strong associations with ex vivo histological findings (male mice: correlation between number of elastin fibers and axial PMSRI: rs = 0.83, p = 0.01; female mice: correlation between shear MASI and plaque score: rs = 0.77, p = 0.009). ConclusionThe results indicate that ABR-LCSI can be used to measure arterial wall strain in a murine model and that changes in strain are associated with changes in arterial wall structure and plaque formation.

Microbubble contrast agent as a minimally invasive pressure sensor at body temperature and reduced concentrations in an ultrasound flow phantom

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): MRC. Background Cardiac catheterisation is the gold standard for assessing cardiac pressures. However, this is an invasive and costly procedure. A minimally invasive alternative may be to use the subharmonic signal of microbubble ultrasound contrast agents to estimate pressure using echocardiography (Sub-Harmonic Assisted Pressure Estimation: SHAPE). Purpose The aim of this work was to investigate the ability of SHAPE to track dynamic diastolic pressures in a flow phantom at body temperature and at reduced microbubble concentrations. Methods A priming dose of 0.4 mL/L of sulphur hexafluoride microbubbles was circulated through a silicone tube using a peristaltic pump at a flow rate of 375 mL/min (generating 376 pressure cycles/min and maximum pressure gradient of 1.9 mmHg/ms; Fig. 1). The ULtrasound Advanced Open Platform (ULA-OP) was used with a cardiac phased array transducer to transmit a pulse-inversion sequence at 2.1 MHz. Radiofrequency data were recorded at five acoustic pressures (71–228 kPa peak-negative). To investigate the effects of temperature, the subharmonic-pressure relationship was measured over 60 min at both room and body temperature (37°C). Microbubble concentration was maintained by regular top-up (0.15 mL/L every 3–5 min). To investigate the effect of lower concentrations on the subharmonic-pressure relationship, a priming dose of 0.4, 0.3 or 0.2 mL/L was administered before a single acquisition was made. The system was immediately purged before testing the next concentration. The subharmonic (1.05 MHz) amplitude of the microbubbles was calculated over a 40% bandwidth. A solid-state catheter recorded ground-truth pressures. Correlation coefficient (r) and sensitivity (dB/mmHg) of the subharmonic signal to the pressure values were calculated. Results The mean correlation coefficient (r = 0.91 ± 0.04; Fig. 2) and sensitivity (0.20 ± 0.04 dB/mmHg) at body temperature (n = 20) were significantly greater than those for the same batch of microbubbles at room temperature (r = 0.81 ± 0.07 and 0.14 ± 0.02; n = 18; p &amp;lt; 0.05) at the optimal acoustic pressure (Fig. 2). No significant differences in the subharmonic-pressure relationship were observed between the three different microbubble concentrations (Correlation: r = 0.96 ± 0.00, 0.96 ± 0.01, 0.97 ± 0.01. Sensitivity: 0.26 ± 0.02, 0.26 ± 0.03, 0.28 ± 0.02 mmHg/dB for 0.4, 0.3 and 0.2 mL/L, respectively; n = 3) at the optimal acoustic pressure. Conclusions SHAPE can accurately track diastolic pressures five times faster than the average resting heart rate at body temperature and with reduced microbubble concentrations. SHAPE is a potential method for assessing cardiac diastolic pressures using ultrasound in a minimally invasive fashion, being suitable for use in patients with faster microbubble clearance or in the event of delayed acquisition. The use of SHAPE fits into routine clinical procedures and has the potential to improve the diagnosis and monitoring of cardiovascular diseases.

Enhanced full-inversion-based ultrasound elastography for evaluating tumor response to neoadjuvant chemotherapy in patients with locally advanced breast cancer

PurposeAn enhanced ultrasound elastography technique is proposed for early assessment of locally advanced breast cancer (LABC) response to neoadjuvant chemotherapy (NAC). MethodsThe proposed elastography technique inputs ultrasound radiofrequency data obtained through tissue quasi-static stimulation and adapts a strain refinement algorithm formulated based on fundamental principles of continuum mechanics, coupled with an iterative inverse finite element method to reconstruct the breast Young’s modulus (E) images. The technique was explored for therapy response assessment using data acquired from 25 LABC patients before and at weeks 1, 2, and 4 after the NAC initiation (100 scans). The E ratio of tumor to the surrounding tissue was calculated at different scans and compared to the baseline for each patient. Patients’ response to NAC was determined many months later using standard clinical and histopathological criteria. ResultsReconstructed E ratio changes obtained as early as one week after the NAC onset demonstrate very good separation between the two cohorts of responders and non-responders to NAC. Statistically significant differences were observed in the E ratio changes between the two patient cohorts at weeks 1 to 4 after treatment (p-value < 0.001; statistical power greater than 97%). A significant difference in axial strain ratio changes was observed only at week 4 (p-value = 0.01; statistical power = 76%). No significant difference was observed in tumor size changes at weeks 1, 2 or 4. ConclusionThe proposed elastography technique demonstrates a high potential for chemotherapy response monitoring in LABC patients and superior performance compared to strain imaging.

Super-Resolution Ultrasound Localization Microscopy Using High-Frequency Ultrasound to Measure Ocular Perfusion Velocity in the Rat Eye.

Imaging of the ocular vasculature can provide new insights into the pathophysiology of ocular diseases. This study proposes a novel high-frequency super-resolution ultrasound localization microscopy (SRULM) technique and evaluates its ability to measure in vivo perfusion changes in the rat eye at elevated intraocular pressure (IOP). A 38.4 MHz center frequency linear array transducer on a VisualSonics Vevo F2 imaging platform was used to collect high frame rate (1 kHz) radiofrequency data of the posterior rat eye following systemic microbubble contrast injection. Following clutter and spatiotemporal non-local means filtering, individual microbubbles were localized and tracked. The microbubble tracks were accumulated over 10,000 frames to generate vascular images quantifying perfusion velocity and direction. Experiments were performed using physiologic relevant controlled flow states for algorithm validation and subsequently performed in vivo on the rat eye at 10 mm Hg IOP increments from 10 to 60 mm Hg. The posterior vasculature of the rat eye, including the ophthalmic artery, long posterior ciliary arteries and their branches, central retinal artery and retinal arterioles and venules were successfully visualized, and velocities quantified at each IOP level. Significant reductions in arterial flow were measured as IOP was elevated. High-frequency SRULM can be used to visualize and quantify the perfusion velocity of the rat eye in both the retrobulbar and intraocular vasculature simultaneously. The ability to detect ocular perfusion changes throughout the depth of the eye may help elucidate the role ischemia has in the pathophysiology of ocular diseases such as glaucoma.

Calibrating Data Mismatches in Deep Learning-Based Quantitative Ultrasound Using Setting Transfer Functions.

Deep learning (DL) can fail when there are data mismatches between training and testing data distributions. Due to its operator-dependent nature, acquisition-related data mismatches, caused by different scanner settings, can occur in ultrasound imaging. As a result, it is crucial to mitigate the effects of these mismatches to enable wider clinical adoption of DL-powered ultrasound imaging and tissue characterization. To address this challenge, we propose an inexpensive and generalizable method that involves collecting a large training set at a single setting and a small calibration set at each scanner setting. Then, the calibration set will be used to calibrate data mismatches by using a signals and systems perspective. We tested the proposed solution to classify two phantoms using an L9-4 array connected to a SonixOne scanner. To investigate generalizability of the proposed solution, we calibrated three types of data mismatches: pulse frequency mismatch, focus mismatch, and output power mismatch. Two well-known convolutional neural networks (CNNs), i.e., ResNet-50 and DenseNet-201, were trained using the ultrasound radio frequency (RF) data. To calibrate the setting mismatches, we calculated the setting transfer functions. The CNNs trained without calibration resulted in mean classification accuracies of around 52%, 84%, and 85% for pulse frequency, focus, and output power mismatches, respectively. By using the setting transfer functions, which allowed a matching of the training and testing domains, we obtained the mean accuracies of 96%, 96%, and 98%, respectively. Therefore, the incorporation of the setting transfer functions between scanner settings can provide an economical means of generalizing a DL model for specific classification tasks where scanner settings are not fixed by the operator.

Safety and durability of PFA for pulmonary vein isolation: comparison to a reference RF data base

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Pulse field ablation (PFA) is a promising ablation technique for pulmonary vein isolation (PVI) with appealing advantages over RF including speed, tissue selectivity and the promise of enhanced durability. We determined procedural efficiency, safety and durability of PFA for PVI and compared its performance to a dataset of optimized radiofrequency (RF) ablation. Methods 180 consecutive patients with paroxysmal or persistent AF undergoing PFA-guided PVI (conventional basket and flower applications for each vein) were included and compared to 180 patients who underwent optimized high-power RF-guided PVI in experienced hands in the POWER-PLUS study. Endpoints were procedural efficiency, safety, and the incidence of 4 isolated veins as evaluated by comprehensive activation mapping at repeat procedure for recurrence. Results Median age was 66 years with 72% paroxysmal AF (clinical characteristics comparable to the RF group). PVI was obtained in all patients with a procedure time of 47min (vs 71min in RF, p &amp;lt; 0.0001) and fluoroscopy time of 14min (vs 11min in RF, p &amp;lt; 0.0001). One serious adverse event (TIA) occurred in a patient with thrombocytosis (1/180, 0.6% vs 0% in RF). No patient experienced pericarditis (vs 1 in RF). During a median follow-up in the PFA group of 4.8 [3.0; 6.4] months, 8 patients underwent repeat for recurrence which revealed a status of 4 isolated veins in a quarter of the patients (2/8, 25%). In the POWER-PLUS study, repeat ablation for recurrence revealed durable isolation of all veins in half of the patients (7/13, 54%, p between groups 0.2). By the time of the EHRA symposium, results will be presented for 6 months in both groups. Conclusion PFA fulfills the promise of offering a short and safe PVI procedure, even when compared to optimized RF in experienced hands. Our data, showing PV reconnection as the dominant cause for recurrence, however temper the expectation of 100% durability. It remains to be seen, how operator expertise affects efficiency, safety, and durability of PFA.

Boosting multi‐target recognition performance with multi‐input multi‐output radar‐based angular subspace projection and multi‐view deep neural network

AbstractCurrent radio frequency (RF) classification techniques assume only one target in the field of view. Multi‐target recognition is challenging because conventional radar signal processing results in the superposition of target micro‐Doppler signatures, making it difficult to recognise multi‐target activity. This study proposes an angular subspace projection technique that generates multiple radar data cubes (RDC) conditioned on angle (RDC‐ω). This approach enables signal separation in the raw RDC, making possible the utilisation of deep neural networks taking the raw RF data as input or any other data representation in multi‐target scenarios. When targets are in closer proximity and cannot be separated by classical techniques, the proposed approach boosts the relative signal‐to‐noise ratio between targets, resulting in multi‐view spectrograms that boosts the classification accuracy when input to the proposed multi‐view DNN. Our results qualitatively and quantitatively characterise the similarity of multi‐view signatures to those acquired in a single‐target configuration. For a nine‐class activity recognition problem, 97.8% accuracy in a 3‐person scenario is achieved, while utilising DNN trained on single‐target data. We also present the results for two cases of close proximity (sign language recognition and side‐by‐side activities), where the proposed approach has boosted the performance.

The Ultrasound Window Into Vascular Ageing: A Technology Review by the VascAgeNet COST Action.

Non-invasive ultrasound (US) imaging enables the assessment of the properties of superficial blood vessels. Various modes can be used for vascular characteristics analysis, ranging from radiofrequency (RF) data, Doppler- and standard B/M-mode imaging, to more recent ultra-high frequency and ultrafast techniques. The aim of the present work was to provide an overview of the current state-of-the-art non-invasive US technologies and corresponding vascular ageing characteristics from a technological perspective. Following an introduction about the basic concepts of the US technique, the characteristics considered in this review are clustered into: 1) vessel wall structure; 2) dynamic elastic properties, and 3) reactive vessel properties. The overview shows that ultrasound is a versatile, non-invasive, and safe imaging technique that can be adopted for obtaining information about function, structure, and reactivity in superficial arteries. The most suitable setting for a specific application must be selected according to spatial and temporal resolution requirements. The usefulness of standardization in the validation process and performance metric adoption emerges. Computer-based techniques should always be preferred to manual measures, as long as the algorithms and learning procedures are transparent and well described, and the performance leads to better results. Identification of a minimal clinically important difference is a crucial point for drawing conclusions regarding robustness of the techniques and for the translation into practice of any biomarker.

Laser diode photoacoustic point source detection: machine learning-based denoising and reconstruction.

A new development in photoacoustic (PA) imaging has been the use of compact, portable and low-cost laser diodes (LDs), but LD-based PA imaging suffers from low signal intensity recorded by the conventional transducers. A common method to improve signal strength is temporal averaging, which reduces frame rate and increases laser exposure to patients. To tackle this problem, we propose a deep learning method that will denoise point source PA radio-frequency (RF) data before beamforming with a very few frames, even one. We also present a deep learning method to automatically reconstruct point sources from noisy pre-beamformed data. Finally, we employ a strategy of combined denoising and reconstruction, which can supplement the reconstruction algorithm for very low signal-to-noise ratio inputs.

Hemoglobin Microbubbles and the Prediction of Different Oxygen Levels Using RF Data and Deep Learning.

Ultrasound contrast agents (UCA) are gas-encapsulated microspheres that oscillate volumetrically when exposed to an ultrasound field producing backscattered signals efficiently, which can be used for improved ultrasound imaging and drug delivery applications. We developed a novel oxygen-sensitive hemoglobin-shell microbubble designed to acoustically detect blood oxygen levels. We hypothesize that structural change in hemoglobin caused due to varying oxygen levels in the body can lead to mechanical changes in the shell of the UCA. This can produce detectable changes in the acoustic response that can be used for measuring oxygen levels in the body. In this study, we have shown that oxygenated hemoglobin microbubbles can be differentiated from deoxygenated hemoglobin microbubbles using a 1D convolutional neural network using radiofrequency (RF) data. We were able to classify RF data from oxygenated and deoxygenated hemoglobin microbubbles into the two classes with a testing accuracy of 90.15%. The results suggest that oxygen content in hemoglobin affects the acoustical response and may be used for determining oxygen levels and thus could open many applications, including evaluating hypoxic regions in tumors and the brain, among other blood-oxygen-level-dependent imaging applications.

High-frequency ultrasound point-of-care device to quantify myopia-induced microstructural changes in the anterior sclera.

To develop a point-of-care (POC) device using high-frequency ultrasound (US) for evaluating microstructural changes in the anterior sclera associated with myopia. The proposed POC device must satisfy four primary requirements for effective clinical use: the measurement component is handheld; the software must be simple and provide real-time feedback; patient safety and health data security requirements set forth by relevant governing bodies must be satisfied and the measurement data must have sufficient signal-to-noise ratio (SNR) and repeatability. Radiofrequency (RF) echo data acquired by the POC device will be processed using our quantitative US methods to characterise tissue microstructure and biomechanical properties. All stated requirements have been met in the developed POC device. The high-frequency transducer is housed in a custom, 3D-printed, pen-like holder that allows for easy measurements of the anterior sclera. Custom software provides a simple interface for data acquisition, real-time data display and secure data storage. Exposimetry measurements of the US pressure field indicate device compliance with United States Food and Drug Administration limits for ophthalmic US. In vivo measurements on a volunteer suggest the RF data SNR and acquisition consistency are suitable for quantitative analysis. A fully functioning POC device using high-frequency US has been created for evaluating the microstructure of the anterior sclera. Planned studies using the POC device to scan the eyes of myopia patients will help clarify how the anterior sclera microstructure may be affected by myopia. If effective, this portable, inexpensive and user-friendly system could be an important part of routine eye examinations.

Implementation of Efficiency Tracking and Evaluation Algorithm for Agricultural Product Circulation Based on Radio Frequency Data Tags and Android Platform

The article believes that the circulation efficiency of agricultural products, especially the efficiency of the circulation system of agricultural products, is the ratio of circulation output to circulation cost in the process of circulation of agricultural products. The circulation efficiency of agricultural products cannot be measured by a single indicator. The article argues that the PCA-DEA combination evaluation model can be constructed by using the market integration degree, market concentration degree, technical efficiency, consumer satisfaction, circulation spread, transaction costs, and using the principal component weight constraint cone to objectively assign DEA values. At the same time, it is necessary to develop a circulation mode of agricultural products led by leading enterprises, promote the intensive processing of agricultural products, and improve the circulation efficiency.

A robust cascaded deep neural network for image reconstruction of single plane wave ultrasound RF data

Reconstruction of ultrasound data from single plane wave Radio Frequency (RF) data is a challenging task. The traditional Delay and Sum (DAS) method produces an image with low resolution and contrast, if employed with RF data from only a single plane wave. A Coherent Compounding (CC) method that reconstructs the image by coherently summing the individual DAS images was proposed to enhance the image quality. However, CC relies on a large number of plane waves to accurately sum the individual DAS images, hence it produces high quality images but with low frame rate that may not be suitable for time-demanding applications. Therefore, there is a need for a method that can create a high quality image with higher frame rates. Furthermore, the method needs to be robust against the input transmission angle of the plane wave. To reduce the method’s dependence on the input angle, we propose to unify the RF data at different angles by learning a linear data transformation from different angled data to a common, 0° data. We further propose a cascade of two independent neural networks to reconstruct an image, similar in quality to CC, by making use of a single plane wave. The first network, denoted as “PixelNet”, is a fully Convolutional Neural Network (CNN) which takes in the transformed time-delayed RF data as input. PixelNet learns optimal pixel weights that get element-wise multiplied with the single angle DAS image. The second network is a conditional Generative Adversarial Network (cGAN) which is used to further enhance the image quality. Our networks were trained on the publicly available PICMUS and CPWC datasets and evaluated on a completely separate, CUBDL dataset obtained from different acquisition settings than the training dataset. The results thus obtained on the testing dataset, demonstrate the networks’ ability to generalize well on unseen data, with frame rates better than the CC method. This paves the way for applications that require high-quality images reconstructed at higher frame rates.

RadioMic: Sound Sensing via Radio Signals

Voice interfaces have become an integral part of our lives with the proliferation of smart devices. Today, Internet of Things devices mainly rely on microphones to sense sound. Microphones, however, have fundamental limitations, such as weak source separation, limited range in the presence of acoustic insulation, and being prone to multiple side-channel attacks. In this article, we propose RadioMic, a radio-based sound sensing system to mitigate these issues and enrich sound applications. RadioMic constructs sound based on tiny vibrations on active sources (e.g., a speaker diaphragm) or object surfaces (e.g., paper bag), and can work through walls, even a soundproof one. To convert the extremely weak sound vibration in the radio signals into sound signals, RadioMic introduces radio acoustics, and presents training-free approaches for robust sound detection and high-fidelity sound recovery. It then exploits a neural network to further enhance the recovered sound by expanding the recoverable frequencies and reducing the noises. RadioMic translates massive online audios to synthesized data to train the network and, thus, minimizes the need for radio-frequency (RF) data. We thoroughly evaluate different components of RadioMic under different scenarios using a commodity mmWave radar. The results show RadioMic outperforms the state-of-the-art systems significantly. We believe RadioMic provides new horizons for sound sensing and inspires attractive sensing capabilities of mmWave sensing devices.

Quantitative lung ultrasound spectroscopy, an in vivo clinical study conducted over 101 patients

Lung ultrasound (LUS) is widely adopted to assess the state of the lung surface. However, as standard ultrasound imaging assumptions are unmet in the lung due to the presence of air, essentially this organ cannot be anatomically investigated. Indeed, LUS is mainly based on the analysis of imaging artifacts (horizontal and vertical). Of particular interest is the analysis of vertical artifacts, as they correlate with several pathologies. More specifically, their dependence on frequency was demonstrated to carry important diagnostic information capable to improve LUS specificity. In this study, a new dataset was generated acquiring raw radiofrequency (RF) data from 101 patients (affected by different pathologies, e.g., COVID-19 pneumonia and cardiogenic edema). The ULA-OP research platform was used to collect the data, and a multifrequency approach implemented on both linear and convex probes was adopted. Ultrasound data were thus acquired utilizing orthogonal sub-bands with a 1-MHz bandwidth and with different center frequencies (2–6 MHz). About 12 000 and 3600 multifrequency frames were acquired with convex and linear probes, respectively. Clinical parameters (e.g., FIO2 and LDH) were also stored for all patients. During this meeting, the dataset will be presented together with preliminary results, and its potential applications discussed.

Improving lateral resolution and contrast by combining coherent plane-wave compounding with adaptive weighting for medical ultrasound imaging

Due to the severe lateral lobe artifact by coherent plane-wave compounding (CPWC) and the low signal-to-noise ratio of radiofrequency (RF) data collected from the plane wave, the adaptive beamforming methods based on focused wave imaging (FWI) are improper to be directly applied to CPWC. To obtain a high-quality image with high resolution and contrast, this study combined the threshold phase coherence factor (THR-PCF) with the reconstructed covariance matrix minimum variance (RCM-MV) and then proposed a novel CPWC-based adaptive beamforming algorithm, THR-PCF + RCM-MV. The simulation, phantom, and in-vivo experiments were performed to investigate the performance of the proposed methods in comparison with the CPWC and the classical adaptive methods including the minimum variance (MV), generalized coherence factor (GCF) and their combination GCF + MV. The simulation results demonstrated that the THR-PCF + RCM-MV beamformer improved contrast ratio (CR) by 28.14%, contrast noise ratio (CNR) by 22.01%, speckle signal-to-noise ratio (s_SNR) by 23.58%, generalized contrast-to-noise ratio (GCNR) by 0.3%, and the full width at half maximum (FWHM) by 43.38% on average, compared with the GCF + MV method. The phantom experimental results showed a better performance of the THR-PCF + RCM-MV beamformer with an average improvement by 21.95% in CR, 2.62% in s_SNR, and 48.64% in FWHM compared with the GCF + MV. Meanwhile, the results showed that the image quality of the near and far fields was enhanced by the THR-PCF + RCM-MV. The in-vivo imaging results showed that our new method had potential for clinical application. In conclusion, the lateral resolution and contrast of medical ultrasound imaging could be improved greatly with our proposed method.

Improving diagnostic accuracy of ultrasound texture features in detecting and quantifying hepatic steatosis using various beamforming sound speeds

Objective. While ultrasound image texture has been utilized to detect and quantify hepatic steatosis, the texture features extracted using a single (conventionally 1540 m s−1) beamforming speed of sound (SoS) failed to achieve reliable diagnostic performance. This study aimed to investigate if the texture features extracted using various beamforming SoSs can improve the accuracy of hepatic steatosis detection and quantification. Approach. Patients with suspected non-alcoholic fatty liver disease underwent liver biopsy or MRI proton density fat fraction (PDFF) as part of standard of care, were prospectively enrolled. The radio-frequency data from subjects’ right and left liver lobes were collected using 6 beamforming SoSs: 1300, 1350, 1400, 1450, 1500 and 1540 m s−1 and analyzed offline. The texture features, i.e. Contrast, Correlation, Energy and Homogeneity from gray-level co-occurrence matrix of normalized envelope were obtained from a region of interest in the liver parenchyma. Main results. Forty-three subjects (67.2%) were diagnosed with steatosis while 21 had no steatosis. Homogeneity showed the area under the curve (AUC) of 0.75–0.82 and 0.58–0.81 for left and right lobes, respectively with varying beamforming SoSs. The combined Homogeneity value over 1300–1540 m s−1 from left and right lobes showed the AUC of 0.90 and 0.81, respectively. Furthermore, the combined Homogeneity values from left and right lobes over 1300–1540 m s−1 improved the AUC to 0.94. The correlation between texture features and steatosis severity was improved by using the images from various beamforming SoSs. The combined Contrast values over 1300–1540 m s−1 from left and right lobes demonstrated the highest correlation (r = 0.90) with the MRI PDFF while the combined Homogeneity values over 1300–1540 m s−1 from left and right lobes showed the highest correlation with the biopsy grades (r = −0.81). Significance. The diagnostic accuracy of ultrasound texture features in detecting and quantifying hepatic steatosis was improved by combining its values extracted using various beamforming SoSs.

Contrast analysis in ultrafast ultrasound blood flow imaging of jugular vein

PurposeThe contrasts of flowing blood in in vitro experiments using porcine blood and in vivo measurements of human jugular veins were analyzed to demonstrate that the hemorheological property was dependent on the shear rate.MethodsBlood samples (45% hematocrit) suspended in saline or plasma were compared with examine the difference in viscoelasticity. Ultrafast plane-wave imaging at an ultrasonic center frequency of 7.5 MHz was performed on different steady flows in a graphite-agar phantom. Also, in vivo measurement was performed in young, healthy subjects and patients with diabetes. A spatiotemporal matrix of beamformed radio-frequency data was used for the singular value decomposition (SVD) clutter filter. The clutter-filtered B-mode image was calculated as the amplitude envelope normalized at the first frame in the diastolic phase to evaluate contrast. The shear rate was estimated as the velocity gradient perpendicular to the lateral axis.ResultsAlthough nonaggregated erythrocytes at a high shear rate exhibited a low echogenicity, the echogenicity in the plasma sample overall increased due to erythrocyte aggregation at a low shear rate. In addition, the frequency of detection of specular components, defined as components beyond twice the standard deviation of a contrast map obtained from a clutter-filtered B-mode image, increased in the porcine blood at a high shear rate and the venous blood in healthy subjects versus patients with diabetes.ConclusionThe possibility of characterizing hemorheological properties dependent on the shear rate and diabetes condition was indicated using ultrafast plane-wave imaging with an SVD-based clutter filter.

Finding Explanations in AI Fusion of Electro-Optical/Passive Radio-Frequency Data.

In the Information Age, the widespread usage of blackbox algorithms makes it difficult to understand how data is used. The practice of sensor fusion to achieve results is widespread, as there are many tools to further improve the robustness and performance of a model. In this study, we demonstrate the utilization of a Long Short-Term Memory (LSTM-CCA) model for the fusion of Passive RF (P-RF) and Electro-Optical (EO) data in order to gain insights into how P-RF data are utilized. The P-RF data are constructed from the in-phase and quadrature component (I/Q) data processed via histograms, and are combined with enhanced EO data via dense optical flow (DOF). The preprocessed data are then used as training data with an LSTM-CCA model in order to achieve object detection and tracking. In order to determine the impact of the different data inputs, a greedy algorithm (explainX.ai) is implemented to determine the weight and impact of the canonical variates provided to the fusion model on a scenario-by-scenario basis. This research introduces an explainable LSTM-CCA framework for P-RF and EO sensor fusion, providing novel insights into the sensor fusion process that can assist in the detection and differentiation of targets and help decision-makers to determine the weights for each input.

Comparison of Optimizers for Drone Signal Detection Using Convolutional Neural Networks (CNN)

Convolutional Neural Networks (CNNs) have emerged as a promising approach for drone signal detection from radio frequency (RF) data. The choice of optimizer significantly affects the performance of CNNs, as it influences the network's ability to learn complex patterns. In this paper, we conduct a comprehensive comparison of popular optimizers, including SGD, Adam, RMSprop, and Adagrad, and their variants, for drone signal detection using CNNs on an RF dataset. The results demonstrated that RMSprop outperformed other optimizers, particularly for larger datasets, maintaining consistently high accuracy with increased test data. Adam, Adamax, and AdamW also achieved impressive accuracy levels above 95%, with the exception of the Adadelta optimized model, which showed lower accuracy due to suboptimal performance during training. In conclusion, the RMSprop optimizer is the most suitable for drone signal detection using a CNN model, although other optimizers such as Adam, Adamax, and AdamW can also provide satisfactory results.