Radio Frequency Data Research Articles

Simultaneous Denoising and Localization Network for Photoacoustic Target Localization.

A significant research problem of recent interest is the localization of targets like vessels, surgical needles, and tumors in photoacoustic (PA) images.To achieve accurate localization, a high photoacoustic signal-to-noise ratio (SNR) is required. However, this is not guaranteed for deep targets, as optical scattering causes an exponential decay in optical fluence with respect to tissue depth. To address this, we develop a novel deep learning method designed to explicitly exhibit robustness to noise present in photoacoustic radio-frequency (RF) data. More precisely, we describe and evaluate a deep neural network architecture consisting of a shared encoder and two parallel decoders. One decoder extracts the target coordinates from the input RF data while the other boosts the SNR and estimates clean RF data. The joint optimization of the shared encoder and dual decoders lends significant noise robustness to the features extracted by the encoder, which in turn enables the network to contain detailed information about deep targets that may be obscured by noise. Additional custom layers and newly proposed regularizers in the training loss function (designed based on observed RF data signal and noise behavior) serve to increase the SNR in the cleaned RF output and improve model performance. To account for depth-dependent strong optical scattering, our network was trained with simulated photoacoustic datasets of targets embedded at different depths inside tissue media of different scattering levels. The network trained on this novel dataset accurately locates targets in experimental PA data that is clinically relevant with respect to the localization of vessels, needles, or brachytherapy seeds. We verify the merits of the proposed architecture by outperforming the state of the art on both simulated and experimental datasets.

Scaled reassigned spectrograms applied to linear transducer signals.

This study evaluates the applicability of scaled reassigned spectrograms (ReSTS) on ultrasound radio frequency data obtained with a clinical linear array ultrasound transducer. The ReSTS's ability to resolve axially closely spaced objects in a phantom is compared to the classical cross-correlation method with respect to the ability to resolve closely spaced objects as individual reflectors using ultrasound pulses with different lengths. The results show that the axial resolution achieved with the ReSTS was superior to the cross-correlation method when the reflected pulses from two objects overlap. A novel B-mode imaging method, facilitating higher image resolution for distinct reflectors, is proposed.

Multidimensional Clutter Filtering of Aperture Domain Data for Improved Blood Flow Sensitivity.

Singular value decomposition (SVD) is a valuable factorization technique used in clutter rejection filtering for power Doppler imaging. Conventionally, SVD is applied to a Casorati matrix of radio frequency data, which enables filtering based on spatial or temporal characteristics. In this article, we propose a clutter filtering method that uses a higher order SVD (HOSVD) applied to a tensor of aperture data, e.g., delayed channel data. We discuss temporal, spatial, and aperture domain features that can be leveraged in filtering and demonstrate that this multidimensional approach improves sensitivity toward blood flow. Further, we show that HOSVD remains more robust to short ensemble lengths than conventional SVD filtering. Validation of this technique is shown using Field II simulations and in vivo data.

A Wireless Intelligent System Construction For Measuring And Control Of Liquid Product Level In Reservoir Parks

The article is devoted to a system construction for measuring and controlling the liquid materials level in remote reservoirs. The hardware solution is based on Wi-Fi technology with radio frequency data transmission and level measurement using an ultrasonic sensor. An internet server is used to collect data in a database and present the data to users in the web pages form. The developed system provides high accuracy, low power consumption and low cost.

Passive Cavitation Mapping by Cavitation Source Localization From Aperture-Domain Signals-Part II: Phantom and In Vivo Experiments.

Passive cavitation mapping (PCM) techniques typically utilize a time-exposure acoustic (TEA) approach, where the received radio frequency data are beamformed, squared, and integrated over time. Such PCM-TEA cavitation maps typically suffer from long-tail artifacts and poor axial resolution with pulse-echo diagnostic arrays. Here, we utilize a recently developed PCM technique based on cavitation source localization (CSL), which fits a hyperbolic function to the received cavitation wavefront. A filtering method based on the root-mean-square error (rmse) of the hyperbolic fit is utilized to filter out spurious signals. We apply a wavefront correction technique to the signals with poor fit quality to recover additional cavitation signals and improve cavitation localization. Validation of the PCM-CSL technique with rmse filtering and wavefront correction was conducted in experiments with a tissue-mimicking flow phantom and an in vivo mouse model of cancer. It is shown that the quality of the hyperbolic fit, necessary for the PCM-CSL, requires an rmse < 0.05 mm2 in order to accurately localize the cavitation sources. A detailed study of the wavefront correction technique was carried out, and it was shown that, when applied to experiments with high noise and interference from multiple cavitating microbubbles, it was capable of effectively correcting noisy wavefronts without introducing spurious cavitation sources, thereby improving the quality of the PCM-CSL images. In phantom experiments, the PCM-CSL was capable of precisely localizing sources on the therapy beam axis and off-axis sources. In vivo cavitation experiments showed that PMC-CSL showed a significant improvement over PCM-TEA and yielded acceptable localization of cavitation signals in mice.

Synthetic Aperture Imaging Using High-Frequency Convex Array for Ophthalmic Ultrasound Applications.

High-frequency ultrasound (HFUS) imaging has emerged as an essential tool for pre-clinical studies and clinical applications such as ophthalmic and dermatologic imaging. HFUS imaging systems based on array transducers capable of dynamic receive focusing have considerably improved the image quality in terms of spatial resolution and signal-to-noise ratio (SNR) compared to those by the single-element transducer-based one. However, the array system still suffers from low spatial resolution and SNR in out-of-focus regions, resulting in a blurred image and a limited penetration depth. In this paper, we present synthetic aperture imaging with a virtual source (SA-VS) for an ophthalmic application using a high-frequency convex array transducer. The performances of the SA-VS were evaluated with phantom and ex vivo experiments in comparison with the conventional dynamic receive focusing method. Pre-beamformed radio-frequency (RF) data from phantoms and excised bovine eye were acquired using a custom-built 64-channel imaging system. In the phantom experiments, the SA-VS method showed improved lateral resolution (>10%) and sidelobe level (>4.4 dB) compared to those by the conventional method. The SNR was also improved, resulting in an increased penetration depth: 16 mm and 23 mm for the conventional and SA-VS methods, respectively. Ex vivo images with the SA-VS showed improved image quality at the entire depth and visualized structures that were obscured by noise in conventional imaging.

High Contrast Ultrasonic Method With Multi-Spatiotemporal Compounding for Monitoring Catheter-Based Ultrasound Thermal Therapy: Development and Ex Vivo Evaluations.

Changes in ultrasound backscatter energy (CBE) imaging can monitor thermal therapy. Catheter-based ultrasound (CBUS) can treat deep tumors with precise spatial control of energy deposition and ablation zones, of which CBE estimation can be limited by low contrast and robustness due to small or inconsistent changes in ultrasound data. This study develops a multi-spatiotemporal compounding CBE (MST-CBE) imaging approach for monitoring specific to CBUS thermal therapy. Ex vivo thermal ablations were performed with stereotactic positioning of a 180° directional CBUS applicator, temperature monitoring probes, endorectal US probe, and subsequent lesion sectioning and measurement. Five frames of raw radiofrequency data were acquired throughout in 15s intervals. Using window-by-window estimation methods, absolute and positive components of MST-CBE images at each point were obtained by the compounding ratio of squared envelope data within an increasing spatial size in each short-time window. Compared with conventional US, Nakagami, and CBE imaging, the detection contrast and robustness quantified by tissue-modification-ratio improved by 37.2 ± 4.7 (p < 0.001), 37.5 ± 5.2 (p < 0.001), and 6.4 ± 4.0 dB (p < 0.05) in the MST-CBE imaging, respectively. Correlation coefficient and bias between cross-sectional dimensions of the ablation zones measured in tissue sections and estimated from MST-CBE were up to 0.91 (p < 0.001) and -0.02 mm2, respectively. The MST-CBE approach can monitor the detailed changes within target tissues and effectively characterize the dimensions of the ablation zone during CBUS energy deposition. The MST-CBE approach could be practical for improved accuracy and contrast of monitoring and evaluation for CBUS thermal therapy.

Ultrasound delta-radiomics during radiotherapy to predict recurrence in patients with head and neck squamous cell carcinoma

PurposeThis study investigated the use of quantitative ultrasound (QUS) obtained during radical radiotherapy (RT) as a radiomics biomarker for predicting recurrence in patients with node-positive head-neck squamous cell carcinoma (HNSCC). MethodsFifty-one patients with HNSCC were treated with RT (70 Gy/33 fractions) (±concurrent chemotherapy) were included. QUS Data acquisition involved scanning an index neck node with a clinical ultrasound device. Radiofrequency data were collected before starting RT, and after weeks 1, and 4. From this data, 31 spectral and related-texture features were determined for each time and delta (difference) features were computed. Patients were categorized into two groups based on clinical outcomes (recurrence or non-recurrence). Three machine learning classifiers were used for the development of a radiomics model. Features were selected using a forward sequential selection method and validated using leave-one-out cross-validation. ResultsThe median follow up for the entire group was 38 months (range 7–64 months). The disease sites involved neck masses in patients with oropharynx (39), larynx (5), carcinoma unknown primary (5), and hypopharynx carcinoma (2). Concurrent chemotherapy and cetuximab were used in 41 and 1 patient(s), respectively. Recurrence was seen in 17 patients. At week 1 of RT, the support vector machine classifier resulted in the best performance, with accuracy and area under the curve (AUC) of 80% and 0.75, respectively. The accuracy and AUC improved to 82% and 0.81, respectively, at week 4 of treatment. ConclusionQUS Delta-radiomics can predict higher risk of recurrence with reasonable accuracy in HNSCC.Clinical trial registration: clinicaltrials.gov.in identifier NCT03908684.

Combining First- and Second-Order Continuity Constraints in Ultrasound Elastography.

Ultrasound elastography is a prominent noninvasive medical imaging technique that estimates tissue elastic properties to detect abnormalities in an organ. A common approximation to tissue elastic modulus is tissue strain induced after mechanical stimulation. To compute tissue strain, ultrasound radio frequency (RF) data can be processed using energy-based algorithms. These algorithms suffer from ill-posedness to tackle. A continuity constraint along with the data amplitude similarity is imposed to obtain a unique solution to the time-delay estimation (TDE) problem. Existing energy-based methods exploit the first-order spatial derivative of the displacement field to construct a regularizer. This first-order regularization scheme alone is not fully consistent with the mechanics of tissue deformation while perturbed with an external force. As a consequence, state-of-the-art techniques suffer from two crucial drawbacks. First, the strain map is not sufficiently smooth in uniform tissue regions. Second, the edges of the hard or soft inclusions are not well-defined in the image. Herein, we address these issues by formulating a novel regularizer taking both first- and second-order derivatives of the displacement field into account. The second-order constraint, which is the principal novelty of this work, contributes both to background continuity and edge sharpness by suppressing spurious noisy edges and enhancing strong boundaries. We name the proposed technique: Second-Order Ultrasound eLastography (SOUL). Comparative assessment of qualitative and quantitative results shows that SOUL substantially outperforms three recently developed TDE algorithms called Hybrid, GLUE, and MPWC-Net++. SOUL yields 27.72%, 62.56%, and 81.37% improvements of the signal-to-noise ratio (SNR) and 72.35%, 54.03%, and 65.17% improvements of the contrast-to-noise ratio (CNR) over GLUE with data pertaining to simulation, phantom, and in vivo tissue, respectively. The SOUL code can be downloaded from code.sonography.ai.

RF-Based UAV Detection and Identification Using Hierarchical Learning Approach.

Unmanned Aerial Vehicles (UAVs) are widely available in the current market to be used either for recreation as a hobby or to serve specific industrial requirements, such as agriculture and construction. However, illegitimate and criminal usage of UAVs is also on the rise which introduces their effective identification and detection as a research challenge. This paper proposes a novel machine learning-based for efficient identification and detection of UAVs. Specifically, an improved UAV identification and detection approach is presented using an ensemble learning based on the hierarchical concept, along with pre-processing and feature extraction stages for the Radio Frequency (RF) data. Filtering is applied on the RF signals in the detection approach to improve the output. This approach consists of four classifiers and they are working in a hierarchical way. The sample will pass the first classifier to check the availability of the UAV, and then it will specify the type of the detected UAV using the second classifier. The last two classifiers will handle the sample that is related to Bebop and AR to specify their mode. Evaluation of the proposed approach with publicly available dataset demonstrates better efficiency compared to existing detection systems in the literature. It has the ability to investigate whether a UAV is flying within the area or not, and it can directly identify the type of UAV and then the flight mode of the detected UAV with accuracy around 99%.

Noninvasive calibrated tissue temperature estimation using backscattered energy of acoustic harmonics

PurposeA real-time and non-invasive thermometry technique is essential in thermal therapies to monitor and control the treatment. Ultrasound is an attractive thermometry modality due to its relatively high sensitivity to change in temperature and fast data acquisition and processing capabilities. A temperature-sensitive acoustic parameter is required for ultrasound thermometry in order to track the changes in that parameter during the treatment. Currently, the main ultrasound thermometry methods are based on variation in the attenuation coefficient, the change in backscattered energy of the signal (CBE), the backscattered radio-frequency (RF) echo-shift due to change in the speed of sound and thermal expansion of the medium, and change in the amplitudes of the acoustic harmonics. In this work, an ultrasound thermometry method based on second harmonic CBE (CBEh2) and combined fundamental and second harmonic CBE (CBEcomb) is used to produce 2D temperature maps, detect localized heated region generated by low intensity focused ultrasound (LIFU), and control temperature in the heated region. Materials and methodsEx vivo pork muscle tissue samples were exposed to localized LIFU heating source and 2D temperature maps were produced from the RF data acquired by a 4.2 MHz linear array probe using a Verasonics Vantage™ ultrasound scanner (Verasonics Inc., Redmond, WA) after the exposure. Calibrated needle thermocouples were also placed in the ex vivo tissue sample close to the LIFU focal zone for temperature calibration purposes. The estimated temperature maps were the established echo-shift technique. A tissue motion compensation algorithm was also used to reduce the susceptibility to motion artifacts. Results2D temperature maps were generated using CBE of acoustic harmonic and echo-shift techniques. The results show a direct correlation between the CBE of acoustic harmonics and focal tissue temperature for a range of temperatures from 37 °C (baseline) to 47 °C. ConclusionsThe findings of this study show that the CBE of acoustic harmonics technique can be used to noninvasively estimate temperature change in tissue in the hyperthermia temperature range.

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 (∼10ms) than the conventional SR method (∼1-5mins), 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 75PWs in reconstruction of in-vivo images of human carotid arteries.

Abstract PS3-29: Pixel-level tissue classification from ultrasound images of breast cancer and direct comparison to matched histological measurements

Abstract Introduction: The use of noninvasive ultrasound (US) for quantitative tissue characterization has been an exciting research prospect for several decades now. Herein the challenge is to find hidden patterns in the US data to reveal more information about tissue function and pathology that cannot be seen in the more conventional US images. A new pixel-level analysis technique has been developed by our group for tissue classification. Termed H-scan US (H stands for hue or Hermite), this imaging approach links a special class of nth-ordered Gaussian-weighted Hermite functions (GHn) to the physics of US scattering and reflection from different tissue structures. The sensitivity of in vivo H-scan US imaging to subtle changes in scatterer size is a question that has not been fully answered. To that end, the purpose of this study was to compare local H-scan US image intensity to direct (co-registered) histological measures made at the cellular level. Methodology Female nude athymic mice were implanted into mammary fatty pad with breast cancer cells (N = 8). Implanted tumors were allowed to grow for about four weeks before study enrollment. Image data was acquired using a programmable US scanner (Vantage 256, Verasonics Inc) equipped with a 256-element L22-8v CMUT linear array transducer (Kolo Medical). Plane wave imaging with 5 angles was performed at a center frequency of 15 MHz. To generate the H-scan US image, three parallel convolution filters (GH2, GH6, and GH10) were applied to the radiofrequency (RF) data sequences to measure the relative strength of the received signals. After envelope detection, the relative strength of the filter outputs is color coded whereby the lower frequency (GH2 = 9 MHz) backscattered US signal components are assigned to the R channel, moderate frequency (GH6 = 15 MHz) signals are assigned to the G channel, and the higher frequency (GH10 = 21 MHz) signals to the B channel. After performing H-scan US, the imaging cross-section was marked, and the tumors were excised and sliced along the same plane for histologic processing. After nuclear staining, tissue sections were scanned and digitized using confocal microscopy and fluorescent filters for DAPI. Automated segmentation of each cancer cell nucleus in the histologic sections was performed using an active contour technique. US images were interpolated to the same number of pixels as the histology image and spatial alignment between the two was performed. Lastly, nucleus size and density from histologic sections was compared to local H-scan US image features. Results: Custom software was developed to compare local US image features at the pixel level to that of co-registered histologic images and measurements of nuclear size and density. In an attempt to properly match the histological and the H-scan US images, three different sized kernels of 0.16 ✗ 0.16 mm, 0.32 ✗ 0.32 mm or 0.5 ✗ 0.5 mm were used to partition both the H-scan US and histology images into 1888, 448 or 62 distinct region-of-interests (ROIs), respectively. Mean nuclear size and density measurements from the histology ROIs was compared to the mean H-scan US image intensity from the many spatially matched ROI locations. A statistically significant linear relationship was found between local H-scan US image intensity and nuclear size (R2 &amp;gt; 0.4, &amp;lt; 0.001) and density (R2 &amp;gt; 0.6, p &amp;lt; 0.001). Conclusions: Preliminary results from use of an animal model of breast cancer reveals that in vivo H-scan US images positively correlated with physical measures of nucleus size and density as quantified from co-registered histologic images. Citation Format: Mawia Khairalseed, Shreya Reddy, Jane Song, Girdhari Rijal, Kenneeth Hoyt. Pixel-level tissue classification from ultrasound images of breast cancer and direct comparison to matched histological measurements [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-29.

Abstract PS3-28: Three-dimensional H-scan ultrasound imaging for acute detection of breast cancer response to neoadjuvant treatment - Initial results using a preclinical animal model

Abstract Introduction: The use of noninvasive ultrasound (US) for tumor tissue characterization is an exciting prospect for anticancer treatment response monitoring. To that end, our group has developed a new technology termed H-scan US imaging, which links differences in the raw backscattered US signals to various-sized tissue structures. The purpose of this research project was to develop a 3-dimensional (3D) H-scan US imaging system and method for tissue characterization in volume space and evaluate using a preclinical animal model of breast cancer. Methodology: Preliminary studies were conducted using female nude athymic mice (N = 20, Charles River Laboratories) implanted in the mammary fat pad with 1 million breast cancer cells (MDA-MB-231, ATCC). Once tumors reached about 1 cm in size, animals were sorted into four groups so that mean tumor size in each was comparable (N = 5 per group). Animals were then US imaged at baseline and before receiving intraperitoneal injections, namely: (1) 0.3 mg sterile saline (control), (2) 0.2 mg of agnostic TRA-8 monoclonal antibody to human death receptor 5 (DR5) + 0.1 mg sterile saline, (3) 0.1 mg paclitaxel + 0.2 mg sterile saline, and (4) 0.1 mg TRA-8 + 0.2 mg paclitaxel. Image data was acquired using a programmable US scanner (Vantage 256, Verasonics Inc) equipped with a volumetric imaging transducer (4DL7, Vermon) at baseline and again every 24 hours for 3 days. To generate the H-scan US images, a set of Gaussian-weighted Hermite filters were convolved with the radiofrequency (RF) data to measure the relative strength of the received signals. The signal envelope for each of the filtered signal then was calculated using a Hilbert transformation. Finally, the lower frequency backscattered signals were assigned to a red (R) channel and the higher frequency components to a blue (B) channel. The unfiltered original RF signal was assigned to the green (G) channel to complete the RGB colormap and 3D H-scan US image display. After US imaging on day 3, animals were humanely euthanized and tumors excised for histological processing using the TUNEL HRP-DAB assay and immunofluorescent staining for activated caspase-3, PARP1, Ki-67, and DAPI. Results: The in vivo results show that 3D H-scan US imaging is considerably more sensitive to tumor changes after neoadjuvant treatment as compared to traditional B-scan US. While there was no difference at baseline (p = 0.52), repeat H-scan US results from treated tumors exhibited progressive increases in image intensity (164.9 ± 23.4%, 180.29 ± 11.6%, and 229.2 ± 14.6% for groups 2, 3, and 4 at day 3, respectively; p &amp;lt; 0.05) due to cancer cell nuclear condensation and apoptotic activity. Moreover, 3D H-scan US exhibited less variance than planar measurements due to increased sample size and statistical averaging. Histological findings also confirmed increased apoptosis and decreased proliferation that matched H-scan US data trends. Conclusions: 3D H-scan US imaging is a promising technique that allows visualization of the heterogenous tissue microenvironment and improves the evaluation of treatment at an early stage of therapy as validated by histologic findings. Citation Format: Haowei Tai, Shreya Reddy, Jane Song, Mawia Khairalseed, Kenneeth Hoyt. Three-dimensional H-scan ultrasound imaging for acute detection of breast cancer response to neoadjuvant treatment - Initial results using a preclinical animal model [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-28.

Non-invasive evaluation of the common carotid artery in normal adults using ultrasound radiofrequency data technology

To evaluate the changes in CCA-IMT and vascular elasticity in healthy subjects in relation to age using radiofrequency data technology, to produce reference values for potential clinical applications. The following variables were measured in 160 subjects using radiofrequency data techniques: CCA-IMT, carotid distensibility (CD), local pulse-wave velocity (PWVβ), and stiffness (β). The subjects were divided into four age groups (30-39, 40-49, 50-59, and 60-69 years), with 20 men and 20 women in each group. The above parameters were compared among the different age groups. CCA-IMT increased significantly with age (p<0.05). CD decreased significantly in the 40-49-years compared with the 30-39-years age group (p<0.05). PWVβ showed a significant increase in the 30-39-, 40-49-, and 50-59-years age groups (p<0.05). β increased significantly in the 50-59- compared with the 40-49-years age group (p<0.05). Compared with the 50-59-years age group, CD, PWVβ, and β showed no significant changes in the 60-69-years age group (p>0.05), with mean values of 329.39±102.06 μm, 7.22±1.91 m/s, and 10.04±4.91, respectively. Ultrasound radiofrequency data technology provides a non-invasive method for quantitatively and accurately assessing increased CCA-IMT and reduced vascular elasticity in healthy subjects in different age groups. The reference values produced in this study will be useful in clinical practice.

Deep learning framework for subject-independent emotion detection using wireless signals.

Emotion states recognition using wireless signals is an emerging area of research that has an impact on neuroscientific studies of human behaviour and well-being monitoring. Currently, standoff emotion detection is mostly reliant on the analysis of facial expressions and/or eye movements acquired from optical or video cameras. Meanwhile, although they have been widely accepted for recognizing human emotions from the multimodal data, machine learning approaches have been mostly restricted to subject dependent analyses which lack of generality. In this paper, we report an experimental study which collects heartbeat and breathing signals of 15 participants from radio frequency (RF) reflections off the body followed by novel noise filtering techniques. We propose a novel deep neural network (DNN) architecture based on the fusion of raw RF data and the processed RF signal for classifying and visualising various emotion states. The proposed model achieves high classification accuracy of 71.67% for independent subjects with 0.71, 0.72 and 0.71 precision, recall and F1-score values respectively. We have compared our results with those obtained from five different classical ML algorithms and it is established that deep learning offers a superior performance even with limited amount of raw RF and post processed time-sequence data. The deep learning model has also been validated by comparing our results with those from ECG signals. Our results indicate that using wireless signals for stand-by emotion state detection is a better alternative to other technologies with high accuracy and have much wider applications in future studies of behavioural sciences.

A priori prediction of response in multicentre locally advanced breast cancer (LABC) patients using quantitative ultrasound and derivative texture methods.

Purpose: We develop a multi-centric response predictive model using QUS spectral parametric imaging and novel texture-derivate methods for determining tumour responses to neoadjuvant chemotherapy (NAC) prior to therapy initiation.Materials and Methods: QUS Spectroscopy provided parametric images of mid-band-fit (MBF), spectral-slope (SS), spectral-intercept (SI), average-scatterer-diameter (ASD), and average-acoustic-concentration (AAC) in 78 patients with locally advanced breast cancer (LABC) undergoing NAC. Ultrasound radiofrequency data were collected from Sunnybrook Health Sciences Center (SHSC), University of Texas MD Anderson Cancer Center (MD-ACC), and St. Michaels Hospital (SMH) using two different systems. Texture analysis was used to quantify heterogeneities of QUS parametric images. Further, a second-pass texture analysis was applied to obtain texture-derivate features. QUS, texture- and texture-derivate parameters were determined from both tumour core and a 5-mm tumour margin and were used in comparison to histopathological analysis for developing a response predictive model to classify responders versus non-responders. Model performance was assessed using leave-one-out cross-validation. Three standard classification algorithms including a linear discriminant analysis (LDA), k-nearest-neighbors (KNN), and support vector machines-radial basis function (SVM-RBF) were evaluated.Results: A combination of tumour core and margin classification resulted in a peak response prediction performance of 88% sensitivity, 78% specificity, 84% accuracy, 0.86 AUC, 84% PPV, and 83% NPV, achieved using the SVM-RBF classification algorithm. Other parameters and classifiers performed less well running from 66% to 80% accuracy.Conclusions: A QUS-based framework and novel texture-derivative method enabled accurate prediction of responses to NAC. Multi-centric response predictive model provides indications of the robustness of the approach to variations due to different ultrasound systems and acquisition parameters.

Liver Fat Quantification by Ultrasound in Children: A Prospective Study.

BACKGROUND. Nonalcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver disease in children in certain regions and is rising in prevalence with increasing obesity. Accurate noninvasive imaging methods for diagnosing and quantifying liver fat are needed to guide NAFLD management. OBJECTIVE. The purpose of this article is to evaluate four ultrasound technologies for quantitative assessment of liver fat content in children using MRI proton density fat fraction (PDFF) as a reference standard. METHODS. This prospective study enrolled children who underwent clinical abdominal MRI without general anesthesia between November 2018 and July 2019. Patients underwent investigational liver ultrasound within a day of 1.5-T or 3-T MRI. Acquired ultrasound radiofrequency data were processed offline to compute the acoustic attenuation coefficient, hepatorenal index (HRI), Nakagami parameter, and shear-wave elastography (SWE) parameters (elasticity, viscosity, and dispersion). Ultrasound parameters were compared with MRI PDFF obtained using a multiecho sequence. A second observer independently performed offline attenuation coefficient and HRI measurements in all patients. RESULTS. A total of 48 patients were enrolled: 22 girls, 26 boys; mean age of 13 years (range, 7-17 years); mean body mass index (weight in kilograms divided by the square of height in meters) of 22.25 (range, 14.5-48.1). A total of 21% (10/48) had steatosis (PDFF ≥ 5%). PDFF was correlated with attenuation coefficient (r = 0.76; 95% CI, 0.60-0.86; p < .001), HRI (r = 0.84; 95% CI, 0.74-0.91; p < .001), and Nakagami parameter (r = 0.55, 95% CI, 0.32-0.72, p < .001), but not SWE parameters (r = 0.05-0.25; p > .05). In patients with no, mild, moderate, and severe steatosis according to PDFF, the mean (± SD) attenuation coefficient was 0.48 ± 0.08, 0.54 ± 0.03, 0.57 ± 0.04, and 0.86 ± 0.07 dB/cm/MHz, respectively, and the mean HRI was 1.28 ± 0.30, 1.59 ± 0.23, 2.25 ± 0.04, and 3.06 ± 0.49, respectively. For the attenuation coefficient, the threshold of 0.54 dB/cm/MHz achieved a sensitivity of 80% and a specificity of 82% for steatosis, and 0.60 dB/cm/MHz achieved a sensitivity of 80% and a specificity of 98% for moderate steatosis. For HRI, the threshold of 1.48 achieved sensitivity of 90% and specificity of 76% for steatosis, and 2.11 achieved sensitivity of 100% and specificity of 100% for moderate steatosis. The interobserver concordance coefficient was 0.92 for attenuation coefficient and 0.91 for HRI. CONCLUSION. Attenuation coefficient and HRI accurately detected and quantified liver fat in this small sample of children. CLINICAL IMPACT. Quantitative ultrasound parameters may guide NAFLD diagnosis and management in children.

Thyroid nodule recognition using a joint convolutional neural network with information fusion of ultrasound images and radiofrequency data.

To develop a deep learning-based method with information fusion of US images and RF signals for better classification of thyroid nodules (TNs). One hundred sixty-three pairs of US images and RF signals of TNs from a cohort of adult patients were used for analysis. We developed an information fusion-based joint convolutional neural network (IF-JCNN) for the differential diagnosis of malignant and benign TNs. The IF-JCNN contains two branched CNNs for deep feature extraction: one for US images and the other one for RF signals. The extracted features are fused at the backend of IF-JCNN for TN classification. Across 5-fold cross-validation, the accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) obtained by using the IF-JCNN with both US images and RF signals as inputs for TN classification were respectively 0.896 (95% CI 0.838-0.938), 0.885 (95% CI 0.804-0.941), 0.910 (95% CI 0.815-0.966), and 0.956 (95% CI 0.926-0.987), which were better than those obtained by using only US images: 0.822 (0.755-0.878; p = 0.0044), 0.792 (0.679-0.868, p = 0.0091), 0.866 (0.760-0.937, p = 0.197), and 0.901 (0.855-0.948, p = .0398), or RF signals: 0.767 (0.694-0.829, p < 0.001), 0.781 (0.685-0.859, p = 0.0037), 0.746 (0.625-0.845, p < 0.001), 0.845 (0.786-0.903, p < 0.001). The proposed IF-JCNN model filled the gap of just using US images in CNNs to characterize TNs, and it may serve as a promising tool for assisting the diagnosis of thyroid cancer. • Raw radiofrequency signals before ultrasound imaging of thyroid nodules provide useful information that is not carried by ultrasound images. • The information carried by raw radiofrequency signals and ultrasound images for thyroid nodules is complementary. • The performance of deep convolutional neural network for diagnosing thyroid nodules can be significantly improved by fusing US images and RF signals in the model as compared with just using US images.

Quantitative Ultrasound Radiofrequency Data Analysis for the Assessment of Hepatic Steatosis in Nonalcoholic Fatty Liver Disease Using Magnetic Resonance Imaging Proton Density Fat Fraction as the Reference Standard.

ObjectiveTo investigate the diagnostic performance of quantitative ultrasound (US) parameters for the assessment of hepatic steatosis in patients with nonalcoholic fatty liver disease (NAFLD) using magnetic resonance imaging proton density fat fraction (MRI-PDFF) as the reference standard.Materials and MethodsIn this single-center prospective study, 120 patients with clinically suspected NAFLD were enrolled between March 2019 and January 2020. The participants underwent US examination for radiofrequency (RF) data acquisition and chemical shift-encoded liver MRI for PDFF measurement. Using the RF data analysis, the attenuation coefficient (AC) based on tissue attenuation imaging (TAI) (AC-TAI) and scatter-distribution coefficient (SC) based on tissue scatter-distribution imaging (TSI) (SC-TSI) were measured. The correlations between the quantitative US parameters (AC and SC) and MRI-PDFF were evaluated using Pearson correlation coefficients. The diagnostic performance of AC-TAI and SC-TSI for detecting hepatic fat contents of ≥ 5% (MRI-PDFF ≥ 5%) and ≥ 10% (MRI-PDFF ≥ 10%) were assessed using receiver operating characteristic (ROC) analysis. The significant clinical or imaging factors associated with AC and SC were analyzed using linear regression analysis.ResultsThe participants were classified based on MRI-PDFF: < 5% (n = 38), 5–10% (n = 23), and ≥ 10% (n = 59). AC-TAI and SC-TSI were significantly correlated with MRI-PDFF (r = 0.659 and 0.727, p < 0.001 for both). For detecting hepatic fat contents of ≥ 5% and ≥ 10%, the areas under the ROC curves of AC-TAI were 0.861 (95% confidence interval [CI]: 0.786–0.918) and 0.835 (95% CI: 0.757–0.897), and those of SC-TSI were 0.964 (95% CI: 0.913–0.989) and 0.935 (95% CI: 0.875–0.972), respectively. Multivariable linear regression analysis showed that MRI-PDFF was an independent determinant of AC-TAI and SC-TSI.ConclusionAC-TAI and SC-TSI derived from quantitative US RF data analysis yielded a good correlation with MRI-PDFF and provided good performance for detecting hepatic steatosis and assessing its severity in NAFLD.