SCANED: Siamese collateral assessment network for evaluation of collaterals from ischemic damage.

This study conducts collateral evaluation from ischemic damage using a deep learning-based Siamese network, addressing the challenges associated with a small and imbalanced dataset. The collateral network provides an alternative oxygen and nutrient supply pathway in ischemic stroke cases, influencing treatment decisions. Research in this area focuses on automated collateral assessment using deep learning (DL) methods to expedite decision-making processes and enhance accuracy. Our study employed a 3D ResNet-based Siamese network, referred to as SCANED, to classify collaterals as good/intermediate or poor. Utilizing non-contrast computed tomography (NCCT) images, the network automates collateral identification and assessment by analyzing tissue degeneration around the ischemic site. Relevant features from the left/right hemispheres were extracted, and Euclidean Distance (ED) was employed for similarity measurement. Finally, dichotomized classification of good/intermediate or poor collateral is performed by SCANED using an optimal threshold derived from ROC analysis. SCANED provides a sensitivity of 0.88, a specificity of 0.63, and a weighted F1 score of 0.86 in the dichotomized classification.

Homodyned K-Distribution Parameter Estimation in Quantitative Ultrasound: Autoencoder and Bayesian Neural Network Approaches.

Quantitative ultrasound (QUS) analyzes the ultrasound (US) backscattered data to find the properties of scatterers that correlate with the tissue microstructure. Statistics of the envelope of the backscattered radio frequency (RF) data can be utilized to estimate several QUS parameters. Different distributions have been proposed to model envelope data. The homodyned K-distribution (HK-distribution) is one of the most comprehensive distributions that can model US backscattered envelope data under diverse scattering conditions (varying scatterer number density and coherent scattering). The scatterer clustering parameter ( α ) and the ratio of the coherent to diffuse scattering power ( k ) are the parameters of this distribution that have been used extensively for tissue characterization in diagnostic US. The estimation of these two parameters (which we refer to as HK parameters) is done using optimization algorithms in which statistical features such as the envelope point-wise signal-to-noise ratio (SNR), skewness, kurtosis, and the log-based moments have been utilized as input to such algorithms. The optimization methods minimize the difference between features and their theoretical value from the HK model. We propose that the true value of these statistical features is a hyperplane that covers a small portion of the feature space. In this article, we follow two approaches to reduce the effect of sample features' error. We propose a model projection neural network based on denoising autoencoders to project the noisy features into this space based on this assumption. We also investigate if the noise distribution can be learned by the deep estimators. We compare the proposed methods with conventional methods using simulations, an experimental phantom, and data from an in vivo animal model of hepatic steatosis. The network weight and a demo code are available online at ht.tp://code.sonography.ai.

Exploiting Mechanics-Based Priors for Lateral Displacement Estimation in Ultrasound Elastography.

Tracking the displacement between the pre- and post-deformed radio-frequency (RF) frames is a pivotal step of ultrasound elastography, which depicts tissue mechanical properties to identify pathologies. Due to ultrasound's poor ability to capture information pertaining to the lateral direction, the existing displacement estimation techniques fail to generate an accurate lateral displacement or strain map. The attempts made in the literature to mitigate this well-known issue suffer from one of the following limitations: 1) Sampling size is substantially increased, rendering the method computationally and memory expensive. 2) The lateral displacement estimation entirely depends on the axial one, ignoring data fidelity and creating large errors. This paper proposes exploiting the effective Poisson's ratio (EPR)-based mechanical correspondence between the axial and lateral strains along with the RF data fidelity and displacement continuity to improve the lateral displacement and strain estimation accuracies. We call our techniques MechSOUL (Mechanically-constrained Second-Order Ultrasound eLastography) and L1 -MechSOUL ( L1 -norm-based MechSOUL), which optimize L2 - and L1 -norm-based penalty functions, respectively. Extensive validation experiments with simulated, phantom, and in vivo datasets demonstrate that MechSOUL and L1 -MechSOUL's lateral strain and EPR estimation abilities are substantially superior to those of the recently-published elastography techniques. We have published the MATLAB codes of MechSOUL and L1 -MechSOUL at https://code.sonography.ai.

Lateral Strain Imaging Using Self-Supervised and Physically Inspired Constraints in Unsupervised Regularized Elastography.

Convolutional Neural Networks (CNN) have shown promising results for displacement estimation in UltraSound Elastography (USE). Many modifications have been proposed to improve the displacement estimation of CNNs for USE in the axial direction. However, the lateral strain, which is essential in several downstream tasks such as the inverse problem of elasticity imaging, remains a challenge. The lateral strain estimation is complicated since the motion and the sampling frequency in this direction are substantially lower than the axial one, and a lack of carrier signal in this direction. In computer vision applications, the axial and the lateral motions are independent. In contrast, the tissue motion pattern in USE is governed by laws of physics which link the axial and lateral displacements. In this paper, inspired by Hooke's law, we, first propose Physically Inspired ConsTraint for Unsupervised Regularized Elastography (PICTURE), where we impose a constraint on the Effective Poisson's ratio (EPR) to improve the lateral strain estimation. In the next step, we propose self-supervised PICTURE (sPICTURE) to further enhance the strain image estimation. Extensive experiments on simulation, experimental phantom and in vivo data demonstrate that the proposed methods estimate accurate axial and lateral strain maps.

Deep Estimation of Speckle Statistics Parametric Images.

Quantitative Ultrasound (QUS) provides important information about the tissue properties. QUS parametric image can be formed by dividing the envelope data into small overlapping patches and computing different speckle statistics such as parameters of the Nakagami and Homodyned K-distributions (HK-distribution). The calculated QUS parametric images can be erroneous since only a few independent samples are available inside the patches. Another challenge is that the envelope samples inside the patch are assumed to come from the same distribution, an assumption that is often violated given that the tissue is usually not homogenous. In this paper, we propose a method based on Convolutional Neural Networks (CNN) to estimate QUS parametric images without patching. We construct a large dataset sampled from the HK-distribution, having regions with random shapes and QUS parameter values. We then use a well-known network to estimate QUS parameters in a multi-task learning fashion. Our results confirm that the proposed method is able to reduce errors and improve border definition in QUS parametric images.

Robust Scatterer Number Density Segmentation of Ultrasound Images.

Quantitative ultrasound (QUS) aims to reveal information about the tissue microstructure using backscattered echo signals from clinical scanners. Among different QUS parameters, scatterer number density is an important property that can affect the estimation of other QUS parameters. Scatterer number density can be classified into high or low scatterer densities. If there are more than ten scatterers inside the resolution cell, the envelope data are considered as fully developed speckle (FDS) and, otherwise, as underdeveloped speckle (UDS). In conventional methods, the envelope data are divided into small overlapping windows (a strategy here we refer to as patching), and statistical parameters, such as SNR and skewness, are employed to classify each patch of envelope data. However, these parameters are system-dependent, meaning that their distribution can change by the imaging settings and patch size. Therefore, reference phantoms that have known scatterer number density are imaged with the same imaging settings to mitigate system dependency. In this article, we aim to segment regions of ultrasound data without any patching. A large dataset is generated, which has different shapes of scatterer number density and mean scatterer amplitude using a fast simulation method. We employ a convolutional neural network (CNN) for the segmentation task and investigate the effect of domain shift when the network is tested on different datasets with different imaging settings. Nakagami parametric image is employed for multitask learning to improve performance. Furthermore, inspired by the reference phantom methods in QUS, a domain adaptation stage is proposed, which requires only two frames of data from FDS and UDS classes. We evaluate our method for different experimental phantoms and in vivo data.

Bi-Directional Semi-Supervised Training of Convolutional Neural Networks for Ultrasound Elastography Displacement Estimation.

The performance of ultrasound elastography (USE) heavily depends on the accuracy of displacement estimation. Recently, convolutional neural networks (CNNs) have shown promising performance in optical flow estimation and have been adopted for USE displacement estimation. Networks trained on computer vision images are not optimized for USE displacement estimation since there is a large gap between the computer vision images and the high-frequency radio frequency (RF) ultrasound data. Many researchers tried to adopt the optical flow CNNs to USE by applying transfer learning to improve the performance of CNNs for USE. However, the ground-truth displacement in real ultrasound data is unknown, and simulated data exhibit a domain shift compared to the real data and are also computationally expensive to generate. To resolve this issue, semisupervised methods have been proposed in which the networks pretrained on computer vision images are fine-tuned using real ultrasound data. In this article, we employ a semisupervised method by exploiting the first- and second-order derivatives of the displacement field for regularization. We also modify the network structure to estimate both forward and backward displacements and propose to use consistency between the forward and backward strains as an additional regularizer to further enhance the performance. We validate our method using several experimental phantom and in vivo data. We also show that the network fine-tuned by our proposed method using experimental phantom data performs well on in vivo data similar to the network fine-tuned on in vivo data. Our results also show that the proposed method outperforms current deep learning methods and is comparable to computationally expensive optimization-based algorithms.

Ultrasound Scatterer Density Classification Using Convolutional Neural Networks and Patch Statistics.

Quantitative ultrasound (QUS) can reveal crucial information on tissue properties, such as scatterer density. If the scatterer density per resolution cell is above or below 10, the tissue is considered as fully developed speckle (FDS) or underdeveloped speckle (UDS), respectively. Conventionally, the scatterer density has been classified using estimated statistical parameters of the amplitude of backscattered echoes. However, if the patch size is small, the estimation is not accurate. These parameters are also highly dependent on imaging settings. In this article, we adapt convolutional neural network (CNN) architectures for QUS and train them using simulation data. We further improve the network's performance by utilizing patch statistics as additional input channels. Inspired by deep supervision and multitask learning, we propose a second method to exploit patch statistics. We evaluate the networks using simulation data and experimental phantoms. We also compare our proposed methods with different classic and deep learning models and demonstrate their superior performance in the classification of tissues with different scatterer density values. The results also show that we are able to classify scatterer density in different imaging parameters with no need for a reference phantom. This work demonstrates the potential of CNNs in classifying scatterer density in ultrasound images.

A Pilot Study on Scatterer Density Classification of Ultrasound Images Using Deep Neural Networks.

Quantitative ultrasound estimates different intrinsic tissue properties, which can be used for tissue characterization. Among different tissue properties, the effective number of scatterers per resolution cell is an important parameter, which can be estimated by the echo envelope. Assuming the signal is stationary and coherent, if the number of scatterers per resolution cell is above approximately 10, envelope signal is considered to be fully developed speckle (FDS) and otherwise they are from low scatterer number density (LSND). Two statistical parameters named R and S are often calculated from envelope intensity to classify FDS from LSND. The main problem is that limited data from small patches often renders this classification inaccurate. Herein, we propose two techniques based on neural networks to estimate the effective number of scatterers. The first network is a multi-layer perceptron (MLP) that uses the hand-crafted features of R and S for classification. The second network is a convolutional neural network (CNN) that does not need hand-crafted features and instead utilizes spectrum and the envelope intensity directly. We show that the proposed MLP works very well for large patches wherein a reliable estimation of R and S can be made. However, its classification becomes inaccurate for small patches, where the proposed CNN provides accurate classifications.

Segmentation of Ultrasound Images based on Scatterer Density using U-Net.

Quantitative ultrasound can provide an objective estimation of different tissue properties, which may be used for tissue characterization and detection of abnormal tissue. The effective number of scatterers in different parts of a tissue is one of the important tissue properties that can be estimated by quantitative ultrasound techniques. The envelope echo is the signal which is usually used to estimate the scatterer density. In this study, we proposed using deep learning to estimate the effective number of scatterers. We generated 2000 simulated phantom data containing randomly distributed inclusions with three different values for number of scatterers per resolution cell. We used U-Net to segment the envelope data and to distinguish three different values of scatterer densities. We show that U-Net can discriminate different scattering regimes, particularly, when the difference between the number of scatterers is substantial. The overall accuracy of the network is 83.9%, and the average sensitivity and specificity among the three classes are 83.1% and 92.3% respectively. This study confirms the potential of deep learning framework in quantitative ultrasound and estimation of tissue properties using ultrasound images.

Real-time and High Quality Ultrasound Elastography Using Convolutional Neural Network by Incorporating Analytic Signal.

Convolutional Neural Networks (CNN) have been extensively used for many computer vision applications including optical flow estimation. Although CNNs have been very successful in optical flow problem, they have been rarely used for displacement estimation in Ultrasound Elastography (USE) due to vast differences between ultrasound data and computer vision images. In USE, a main goal is to obtain the strain image which is the derivative of the axial displacement in axial direction; therefore, a very accurate displacement estimation is required. Radio Frequency (RF) data is needed to obtain accurate displacement estimation. RF data contains high frequency contents which cannot be downsampled without significant loss of information, in contrast to computer vision images. We propose a novel technique to utilize LiteFlowNet for USE. For the first time, we incorporate analytic signal to improve the quality of the displacement estimation. We show that this network with the designed inputs is more suitable for USE compared to more complex networks such as FlowNet2. The network is adopted to our application and it is compared with FlowNet2 and a state-of-the-art elastography method (GLUE). The results show that this network performs well and comparable to GLUE. Furthermore, not only this network is faster and has lower memory footprint compared to FlowNet2, but also it obtains higher quality strain images which makes it suitable for portable and real-time elastography devices.

Displacement Estimation in Ultrasound Elastography Using Pyramidal Convolutional Neural Network.

In this article, two novel deep learning methods are proposed for displacement estimation in ultrasound elastography (USE). Although convolutional neural networks (CNNs) have been very successful for displacement estimation in computer vision, they have been rarely used for USE. One of the main limitations is that the radio frequency (RF) ultrasound data, which is crucial for precise displacement estimation, has vastly different frequency characteristics compared with images in computer vision. Top-rank CNN methods used in computer vision applications are mostly based on a multilevel strategy, which estimates finer resolution based on coarser ones. This strategy does not work well for RF data due to its large high-frequency content. To mitigate the problem, we propose modified pyramid warping and cost volume network (MPWC-Net) and RFMPWC-Net, both based on PWC-Net, to exploit information in RF data by employing two different strategies. We obtained promising results using networks trained only on computer vision images. In the next step, we constructed a large ultrasound simulation database and proposed a new loss function to fine-tune the network to improve its performance. The proposed networks and well-known optical flow networks as well as state-of-the-art elastography methods are evaluated using simulation, phantom, and in vivo data. Our two proposed networks substantially outperform current deep learning methods in terms of contrast-to-noise ratio (CNR) and strain ratio (SR). Also, the proposed methods perform similar to the state-of-the-art elastography methods in terms of CNR and have better SR by substantially reducing the underestimation bias.