Multimodality quantitative ultrasound envelope statistics imaging based support vector machines for characterizing tissue scatterer distribution patterns: Methods and application in detecting microwave-induced thermal lesions.

Ultrasound envelope statistics imaging, including ultrasound Nakagami imaging, homodyned-K imaging, and information entropy imaging, is an important group of quantitative ultrasound techniques for characterizing tissue scatterer distribution patterns, such as scatterer concentrations and arrangements. In this study, we proposed a machine learning approach to integrate the strength of multimodality quantitative ultrasound envelope statistics imaging techniques and applied it to detecting microwave ablation induced thermal lesions in porcine liver ex vivo. The quantitative ultrasound parameters included were homodyned-K α which is a scatterer clustering parameter related to the effective scatterer number per resolution cell, Nakagami m which is a shape parameter of the envelope probability density function, and Shannon entropy which is a measure of signal uncertainty or complexity. Specifically, the homodyned-K log10(α), Nakagami-m, and horizontally normalized Shannon entropy parameters were combined as input features to train a support vector machine (SVM) model to classify thermal lesions with higher scatterer concentrations from normal tissues with lower scatterer concentrations. Through heterogeneous phantom simulations based on Field II, the proposed SVM model showed a classification accuracy above 0.90; the area accuracy and Dice score of higher-scatterer-concentration zone identification exceeded 83% and 0.86, respectively, with the Hausdorff distance <26. Microwave ablation experiments of porcine liver ex vivo at 60-80 W, 1-3 min showed that the SVM model achieved a classification accuracy of 0.85; compared with single log10(α),m, or hNSE parametric imaging, the SVM model achieved the highest area accuracy (89.1%) and Dice score (0.77) as well as the smallest Hausdorff distance (46.38) of coagulation zone identification. We concluded that the proposed multimodality quantitative ultrasound envelope statistics imaging based SVM approach can enhance the capability to characterize tissue scatterer distribution patterns and has the potential to detect the thermal lesions induced by microwave ablation.

Enhancing Fetal Electrocardiogram Signal Extraction Accuracy through a CycleGAN Utilizing Combined CNN-BiLSTM Architecture.

The fetal electrocardiogram (FECG) records changes in the graph of fetal cardiac action potential during conduction, reflecting the developmental status of the fetus in utero and its physiological cardiac activity. Morphological alterations in the FECG can indicate intrauterine hypoxia, fetal distress, and neonatal asphyxia early on, enhancing maternal and fetal safety through prompt clinical intervention, thereby reducing neonatal morbidity and mortality. To reconstruct FECG signals with clear morphological information, this paper proposes a novel deep learning model, CBLS-CycleGAN. The model's generator combines spatial features extracted by the CNN with temporal features extracted by the BiLSTM network, thus ensuring that the reconstructed signals possess combined features with spatial and temporal dependencies. The model's discriminator utilizes PatchGAN, employing small segments of the signal as discriminative inputs to concentrate the training process on capturing signal details. Evaluating the model using two real FECG signal databases, namely "Abdominal and Direct Fetal ECG Database" and "Fetal Electrocardiograms, Direct and Abdominal with Reference Heartbeat Annotations", resulted in a mean MSE and MAE of 0.019 and 0.006, respectively. It detects the FQRS compound wave with a sensitivity, positive predictive value, and F1 of 99.51%, 99.57%, and 99.54%, respectively. This paper's model effectively preserves the morphological information of FECG signals, capturing not only the FQRS compound wave but also the fetal P-wave, T-wave, P-R interval, and ST segment information, providing clinicians with crucial diagnostic insights and a scientific foundation for developing rational treatment protocols.

Ultrasound k-nearest neighbor entropy imaging: Theory, algorithm, and applications.

Ultrasound information entropy is a flexible approach for analyzing ultrasound backscattering. Shannon entropy imaging based on probability distribution histograms (PDHs) has been implemented as a promising method for tissue characterization and diagnosis. However, the bin number affects the stability of entropy estimation. In this study, we introduced the k-nearest neighbor (KNN) algorithm to estimate entropy values and proposed ultrasound KNN entropy imaging. The proposed KNN estimator leveraged the Euclidean distance between data samples, rather than the histogram bins by conventional PDH estimators. We also proposed cumulative relative entropy (CRE) imaging to analyze time-series radiofrequency signals and applied it to monitor thermal lesions induced by microwave ablation (MWA). Computer simulation phantom experiments were conducted to validate and compare the performance of the proposed KNN entropy imaging, the conventional PDH entropy imaging, and Nakagami-m parametric imaging in detecting the variations of scatterer densities and visualizing inclusions. Clinical data of breast lesions were analyzed, and porcine liver MWA experiments ex vivo were conducted to validate the performance of KNN entropy imaging in classifying benign and malignant breast tumors and monitoring thermal lesions, respectively. Compared with PDH, the entropy estimation based on KNN was less affected by the tuning parameters. KNN entropy imaging was more sensitive to changes in scatterer densities and performed better visualizable capability than typical Shannon entropy (TSE) and Nakagami-m parametric imaging. Among different imaging methods, KNN-based Shannon entropy (KSE) imaging achieved the higher accuracy in classification of benign and malignant breast tumors and KNN-based CRE imaging had larger lesion-to-normal contrast when monitoring the ablated areas during MWA at different powers and treatment durations. Ultrasound KNN entropy imaging is a potential quantitative ultrasound approach for tissue characterization.

Examining the low-voltage fast seizure-onset and its response to optogenetic stimulation in a biophysical network model of the hippocampus.

Low-voltage fast (LVF) seizure-onset is one of the two frequently observed temporal lobe seizure-onset patterns. Depth electroencephalogram profile analysis illustrated that the peak amplitude of LVF onset was deep temporal areas, e.g., hippocampus. However, the specific dynamic transition mechanisms between normal hippocampal rhythmic activity and LVF seizure-onset remain unclear. Recently, the optogenetic approach to gain control over epileptic hyper-excitability both in vitro and in vivo has become a novel noninvasive modulation strategy. Here, we combined biophysical modeling to study LVF dynamics following changes in crucial physiological parameters, and investigated the potential optogenetic intervention mechanism for both excitatory and inhibitory control. In an Ammon's horn 3 (CA3) biophysical model with light-sensitive protein channelrhodopsin 2 (ChR2), we found that the cooperative effects of excessive extracellular potassium concentration of parvalbumin-positive (PV+) inhibitory interneurons and synaptic links could induce abundant types of discharges of the hippocampus, and lead to transitions from gamma oscillations to LVF seizure-onset. Simulations of optogenetic stimulation revealed that the LVF seizure-onset and morbid fast spiking could not be eliminated by targeting PV+ neurons, whereas the epileptic network was more sensitive to the excitatory control of principal neurons with strong optogenetic currents. We illustrate that in the epileptic hippocampal network, the trajectories of the normal and the seizure state are in close vicinity and optogenetic perturbations therefore may result in transitions. The network model system developed in this study represents a scientific instrument to disclose the underlying principles of LVF, to characterize the effects of optogenetic neuromodulation, and to guide future treatment for specific types of seizures.

Elucidating Multimodal Imaging Patterns in Accelerated Brain Aging: Heterogeneity through a Discriminant Analysis Approach Using the UK Biobank Dataset.

Accelerated brain aging (ABA) intricately links with age-associated neurodegenerative and neuropsychiatric diseases, emphasizing the critical need for a nuanced exploration of heterogeneous ABA patterns. This investigation leveraged data from the UK Biobank (UKB) for a comprehensive analysis, utilizing structural magnetic resonance imaging (sMRI), diffusion magnetic resonance imaging (dMRI), and resting-state functional magnetic resonance imaging (rsfMRI) from 31,621 participants. Pre-processing employed tools from the FMRIB Software Library (FSL, version 5.0.10), FreeSurfer, DTIFIT, and MELODIC, seamlessly integrated into the UKB imaging processing pipeline. The Lasso algorithm was employed for brain-age prediction, utilizing derived phenotypes obtained from brain imaging data. Subpopulations of accelerated brain aging (ABA) and resilient brain aging (RBA) were delineated based on the error between actual age and predicted brain age. The ABA subgroup comprised 1949 subjects (experimental group), while the RBA subgroup comprised 3203 subjects (control group). Semi-supervised heterogeneity through discriminant analysis (HYDRA) refined and characterized the ABA subgroups based on distinctive neuroimaging features. HYDRA systematically stratified ABA subjects into three subtypes: SubGroup 2 exhibited extensive gray-matter atrophy, distinctive white-matter patterns, and unique connectivity features, displaying lower cognitive performance; SubGroup 3 demonstrated minimal atrophy, superior cognitive performance, and higher physical activity; and SubGroup 1 occupied an intermediate position. This investigation underscores pronounced structural and functional heterogeneity in ABA, revealing three subtypes and paving the way for personalized neuroprotective treatments for age-related neurological, neuropsychiatric, and neurodegenerative diseases.

An Automatic Needle Puncture Path-Planning Method for Thermal Ablation of Lung Tumors.

Computed tomography (CT)-guided thermal ablation is an emerging treatment method for lung tumors. Ablation needle path planning in preoperative diagnosis is of critical importance. In this work, we proposed an automatic needle path-planning method for thermal lung tumor ablation. First, based on the improved cube mapping algorithm, binary classification was performed on the surface of the bounding box of the patient's CT image to obtain a feasible puncture area that satisfied all hard constraints. Then, for different clinical soft constraint conditions, corresponding grayscale constraint maps were generated, respectively, and the multi-objective optimization problem was solved by combining Pareto optimization and weighted product algorithms. Finally, several optimal puncture paths were planned within the feasible puncture area obtained for the clinicians to choose. The proposed method was evaluated with 18 tumors of varying sizes (482.79 mm3 to 9313.81 mm3) and the automatically planned paths were compared and evaluated with manually planned puncture paths by two clinicians. The results showed that over 82% of the paths (74 of 90) were considered reasonable, with clinician A finding the automated planning path superior in 7 of 18 cases, and clinician B in 9 cases. Additionally, the time efficiency of the algorithm (35 s) was much higher than that of manual planning. The proposed method is expected to aid clinicians in preoperative path planning for thermal ablation of lung tumors. By providing a valuable reference for the puncture path during preoperative diagnosis, it may reduce the clinicians' workload and enhance the objectivity and rationality of the planning process, which in turn improves the effectiveness of treatment.

A review of neuroimaging-based data-driven approach for Alzheimer's disease heterogeneity analysis.

Alzheimer's disease (AD) is a complex form of dementia and due to its high phenotypic variability, its diagnosis and monitoring can be quite challenging. Biomarkers play a crucial role in AD diagnosis and monitoring, but interpreting these biomarkers can be problematic due to their spatial and temporal heterogeneity. Therefore, researchers are increasingly turning to imaging-based biomarkers that employ data-driven computational approaches to examine the heterogeneity of AD. In this comprehensive review article, we aim to provide health professionals with a comprehensive view of past applications of data-driven computational approaches in studying AD heterogeneity and planning future research directions. We first define and offer basic insights into different categories of heterogeneity analysis, including spatial heterogeneity, temporal heterogeneity, and spatial-temporal heterogeneity. Then, we scrutinize 22 articles relating to spatial heterogeneity, 14 articles relating to temporal heterogeneity, and five articles relating to spatial-temporal heterogeneity, highlighting the strengths and limitations of these strategies. Furthermore, we discuss the importance of understanding spatial heterogeneity in AD subtypes and their clinical manifestations, biomarkers for abnormal orderings and AD stages, the recent advancements in spatial-temporal heterogeneity analysis for AD, and the emerging role of omics data integration in advancing personalized diagnosis and treatment for AD patients. By emphasizing the significance of understanding AD heterogeneity, we hope to stimulate further research in this field to facilitate the development of personalized interventions for AD patients.

Ultrasound Entropy Imaging Based on the Kernel Density Estimation: A New Approach to Hepatic Steatosis Characterization.

In this paper, we present the kernel density estimation (KDE)-based parallelized ultrasound entropy imaging and apply it for hepatic steatosis characterization. A KDE technique was used to estimate the probability density function (PDF) of ultrasound backscattered signals. The estimated PDF was utilized to estimate the Shannon entropy to construct parametric images. In addition, the parallel computation technique was incorporated. Clinical experiments of hepatic steatosis were conducted to validate the feasibility of the proposed method. Seventy-two participants and 204 patients with different grades of hepatic steatosis were included. The experimental results show that the KDE-based entropy parameter correlates with log10 (hepatic fat fractions) measured by magnetic resonance spectroscopy in the 72 participants (Pearson's r = 0.52, p < 0.0001), and its areas under the receiver operating characteristic curves for diagnosing hepatic steatosis grades ≥ mild, ≥moderate, and ≥severe are 0.65, 0.73, and 0.80, respectively, for the 204 patients. The proposed method overcomes the drawbacks of conventional histogram-based ultrasound entropy imaging, including limited dynamic ranges and histogram settings dependence, although the diagnostic performance is slightly worse than conventional histogram-based entropy imaging. The proposed KDE-based parallelized ultrasound entropy imaging technique may be used as a new ultrasound entropy imaging method for hepatic steatosis characterization.

Exploring Successful Cognitive Aging: Insights Regarding Brain Structure, Function, and Demographics.

In the realm of cognitive science, the phenomenon of "successful cognitive aging" stands as a hallmark of individuals who exhibit cognitive abilities surpassing those of their age-matched counterparts. However, it is paramount to underscore a significant gap in the current research, which is marked by a paucity of comprehensive inquiries that deploy substantial sample sizes to methodically investigate the cerebral biomarkers and contributory elements underpinning this cognitive success. It is within this context that our present study emerges, harnessing data derived from the UK Biobank. In this study, a highly selective cohort of 1060 individuals aged 65 and above was meticulously curated from a larger pool of 17,072 subjects. The selection process was guided by their striking cognitive resilience, ascertained via rigorous evaluation encompassing both generic and specific cognitive assessments, compared to their peers within the same age stratum. Notably, the cognitive abilities of the chosen participants closely aligned with the cognitive acumen commonly observed in middle-aged individuals. Our study leveraged a comprehensive array of neuroimaging-derived metrics, obtained from three Tesla MRI scans (T1-weighted images, dMRI, and resting-state fMRI). The metrics included image-derived phenotypes (IDPs) that addressed grey matter morphology, the strength of brain network connectivity, and the microstructural attributes of white matter. Statistical analyses were performed employing ANOVA, Mann-Whitney U tests, and chi-square tests to evaluate the distinctive aspects of IDPs pertinent to the domain of successful cognitive aging. Furthermore, these analyses aimed to elucidate lifestyle practices that potentially underpin the maintenance of cognitive acumen throughout the aging process. Our findings unveiled a robust and compelling association between heightened cognitive aptitude and the integrity of white matter structures within the brain. Furthermore, individuals who exhibited successful cognitive aging demonstrated markedly enhanced activity in the cerebral regions responsible for auditory perception, voluntary motor control, memory retention, and emotional regulation. These advantageous cognitive attributes were mirrored in the health-related lifestyle choices of the surveyed cohort, characterized by elevated educational attainment, a lower incidence of smoking, and a penchant for moderate alcohol consumption. Moreover, they displayed superior grip strength and enhanced walking speeds. Collectively, these findings furnish valuable insights into the multifaceted determinants of successful cognitive aging, encompassing both neurobiological constituents and lifestyle practices. Such comprehensive comprehension significantly contributes to the broader discourse on aging, thereby establishing a solid foundation for the formulation of targeted interventions aimed at fostering cognitive well-being among aging populations.

One-Stage Detection without Segmentation for Multi-Type Coronary Lesions in Angiography Images Using Deep Learning.

It is rare to use the one-stage model without segmentation for the automatic detection of coronary lesions. This study sequentially enrolled 200 patients with significant stenoses and occlusions of the right coronary and categorized their angiography images into two angle views: The CRA (cranial) view of 98 patients with 2453 images and the LAO (left anterior oblique) view of 176 patients with 3338 images. Randomization was performed at the patient level to the training set and test set using a 7:3 ratio. YOLOv5 was adopted as the key model for direct detection. Four types of lesions were studied: Local Stenosis (LS), Diffuse Stenosis (DS), Bifurcation Stenosis (BS), and Chronic Total Occlusion (CTO). At the image level, the precision, recall, mAP@0.1, and mAP@0.5 predicted by the model were 0.64, 0.68, 0.66, and 0.49 in the CRA view and 0.68, 0.73, 0.70, and 0.56 in the LAO view, respectively. At the patient level, the precision, recall, and F1scores predicted by the model were 0.52, 0.91, and 0.65 in the CRA view and 0.50, 0.94, and 0.64 in the LAO view, respectively. YOLOv5 performed the best for lesions of CTO and LS at both the image level and the patient level. In conclusion, the one-stage model without segmentation as YOLOv5 is feasible to be used in automatic coronary lesion detection, with the most suitable types of lesions as LS and CTO.

MBR-Net: A multi-branch residual network based on ultrasound backscattered signals for characterizing pediatric hepatic steatosis.

The evaluation of pediatric hepatic steatosis and early detection of fatty liver in children are of critical importance. In this paper, a deep learning model based on the convolutional neural network (CNN) of ultrasound backscattered signals, multi-branch residual network (MBR-Net), was proposed for characterizing pediatric hepatic steatosis. The MBR-Net was composed of three convolutional branches. Each branch used different sizes of convolution blocks to enhance the capability of local feature acquisition, and leveraged the residual mechanism with skip connections to guide the network to effectively capture features. A total of 393 frames of ultrasound backscattered signals collected from 131 children were included in the experiments. The hepatic steatosis index was used as the reference standard for diagnosing the steatosis grade, G0-G3. The ultrasound backscattered signals within the liver region of interests (ROIs) were normalized and augmented using a sliding gate method. The gated ROI signals were randomly divided into training, validation, and test sets with the ratio of 8:1:1. The area under the operating characteristic curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE) were used as the evaluation metrics. Experimental results showed that the MBR-Net yields AUCs for diagnosing pediatric hepatic steatosis grade ≥G1, ≥G2, and ≥G3 of 0.94 (ACC: 93.65%; SEN: 89.79%; SPE: 84.48%), 0.93 (ACC: 90.48%; SEN: 87.75%; SPE: 82.65%), and 0.93 (ACC: 87.76%; SEN: 84.84%; SPE: 86.55%), respectively, which were superior to the conventional one-branch CNNs without residual mechanisms. The proposed MBR-Net can be used as a new deep learning method for ultrasound backscattered signal analysis to characterize pediatric hepatic steatosis.

Moderating effects of cognitive reserve on the relationship between brain structure and cognitive abilities in middle-aged and older adults.

The cognitive reserve (CR) hypothesis is reinforced by negative moderating effects, suggesting that those with higher CR are less reliant on brain structure for cognitive function. Previous research on CR's moderating effects yielded inconsistent results, motivating our 3 studies using UK Biobank data. Study I examined five CR proxies' moderating effects on global, lobar, and regional brain-cognition models; study II extended study I by using a larger sample size; and study III investigated age-related moderating effects on the hippocampal regions. In study I, most moderating effects were negative and none survived the multiple comparison correction, but study II identified 13 global-level models with significant negative moderating effects that survived correction. Study III showed age influenced CR proxies' moderating effects in hippocampal regions. Our findings suggest that the effects of CR proxies on brain integrity and cognition varied depending on the proxy used, brain integrity indicators, cognitive domain, and age group. This study offers significant insights regarding the importance of CR for brain integrity and cognitive outcomes.

Three-round learning strategy based on 3D deep convolutional GANs for Alzheimer's disease staging.

Accurately diagnosing of Alzheimer's disease (AD) and its early stages is critical for prompt treatment or potential intervention to delay the the disease's progression. Convolutional neural networks (CNNs) models have shown promising results in structural MRI (sMRI)-based diagnosis, but their performance, particularly for 3D models, is constrained by the lack of labeled training samples. To address the overfitting problem brought on by the insufficient training sample size, we propose a three-round learning strategy that combines transfer learning with generative adversarial learning. In the first round, a 3D Deep Convolutional Generative Adversarial Networks (DCGAN) model was trained with all available sMRI data to learn the common feature of sMRI through unsupervised generative adversarial learning. The second round involved transferring and fine-tuning, and the pre-trained discriminator (D) of the DCGAN learned more specific features for the classification task between AD and cognitively normal (CN). In the final round, the weights learned in the AD versus CN classification task were transferred to the MCI diagnosis. By highlighting brain regions with high prediction weights using 3D Grad-CAM, we further enhanced the model's interpretability. The proposed model achieved accuracies of 92.8%, 78.1%, and 76.4% in the classifications of AD versus CN, AD versus MCI, and MCI versus CN, respectively. The experimental results show that our proposed model avoids overfitting brought on by a paucity of sMRI data and enables the early detection of AD.

Comparison of Machine Learning Models for Brain Age Prediction Using Six Imaging Modalities on Middle-Aged and Older Adults.

Machine learning (ML) has transformed neuroimaging research by enabling accurate predictions and feature extraction from large datasets. In this study, we investigate the application of six ML algorithms (Lasso, relevance vector regression, support vector regression, extreme gradient boosting, category boost, and multilayer perceptron) to predict brain age for middle-aged and older adults, which is a crucial area of research in neuroimaging. Despite the plethora of proposed ML models, there is no clear consensus on how to achieve better performance in brain age prediction for this population. Our study stands out by evaluating the impact of both ML algorithms and image modalities on brain age prediction performance using a large cohort of cognitively normal adults aged 44.6 to 82.3 years old (N = 27,842) with six image modalities. We found that the predictive performance of brain age is more reliant on the image modalities used than the ML algorithms employed. Specifically, our study highlights the superior performance of T1-weighted MRI and diffusion-weighted imaging and demonstrates that multi-modality-based brain age prediction significantly enhances performance compared to unimodality. Moreover, we identified Lasso as the most accurate ML algorithm for predicting brain age, achieving the lowest mean absolute error in both single-modality and multi-modality predictions. Additionally, Lasso also ranked highest in a comprehensive evaluation of the relationship between BrainAGE and the five frequently mentioned BrainAGE-related factors. Notably, our study also shows that ensemble learning outperforms Lasso when computational efficiency is not a concern. Overall, our study provides valuable insights into the development of accurate and reliable brain age prediction models for middle-aged and older adults, with significant implications for clinical practice and neuroimaging research. Our findings highlight the importance of image modality selection and emphasize Lasso as a promising ML algorithm for brain age prediction.

Recent research advances on simulation modeling of temperature distribution in microwave ablation of lung tumors.

Lung tumor is the first malignant tumor with the highest mortality, but only no more than one-third of patients can be treated by surgical resection. Microwave ablation (MWA) has become a new adjuvant therapeutic mean for lung tumors because of its low trauma, short treatment time, large ablation volume and wide application range. However, the treatment parameters of MWA, such as input power and ablation time, still depend on the doctors' experience, which leads to the ineffectiveness of MWA. Therefore, the accurate modeling of temperature distribution of lung tumor MWA has become a significant technical problem to be solved. Recent research was devoted to personalized characterization of lung tumor parameters, finite element analysis of temperature distribution in MWA and accurate ablation effect evaluation. In this paper, a review of the recently obtained results and data will be presented and discussed.

Parallelized ultrasound homodyned-K imaging based on a generalized artificial neural network estimator.

The homodyned-K (HK) distribution model is a generalized backscatter envelope statistical model for ultrasound tissue characterization, whose parameters are of physical meaning. To estimate the HK parameters is an inverse problem, and is quite complicated. Previously, we proposed an artificial neural network (ANN) estimator and an improved ANN (iANN) estimator for estimating the HK parameters, which are fast and flexible. However, a drawback of the conventional ANN and iANN estimators consists in that they use Monte Carlo simulations under known values of HK parameters to generate training samples, and thus the ANN and iANN models have to be re-trained when the size of the test sets (or of the envelope samples to be estimated) varies. In addition, conventional ultrasound HK imaging uses a sliding window technique, which is non-vectorized and does not support parallel computation, so HK image resolution is usually sacrificed to ensure a reasonable computation cost. To this end, we proposed a generalized ANN (gANN) estimator in this paper, which took the theoretical derivations of feature vectors for network training, and thus it is independent from the size of the test sets. Further, we proposed a parallelized HK imaging method that is based on the gANN estimator, which used a block-based parallel computation method, rather than the conventional sliding window technique. The gANN-based parallelized HK imaging method allowed a higher image resolution and a faster computation at the same time. Computer simulation experiments showed that the gANN estimator was generally comparable to the conventional ANN estimator in terms of HK parameter estimation performance. Clinical experiments of hepatic steatosis showed that the gANN-based parallelized HK imaging could be used to visually and quantitatively characterize hepatic steatosis, with similar performance to the conventional ANN-based HK imaging that used the sliding window technique, but the gANN-based parallelized HK imaging was over 3 times faster than the conventional ANN-based HK imaging. The parallelized computation method presented in this work can be easily extended to other quantitative ultrasound imaging applications.

Ultrasound Homodyned-K Contrast-Weighted Summation Parametric Imaging Based on H-scan for Detecting Microwave Ablation Zones.

The homodyned-K (HK) distribution is a generalized model of envelope statistics whose parameters α (the clustering parameter) and k (the coherent-to-diffuse signal ratio) can be used to monitor the thermal lesions. In this study, we proposed an ultrasound HK contrast-weighted summation (CWS) parametric imaging algorithm based on the H-scan technique and investigated the optimal window side length (WSL) of the HK parameters estimated by the XU estimator (an estimation method based on the first moment of the intensity and two log-moments, which was used in the proposed algorithm) through phantom simulations. H-scan diversified ultrasonic backscattered signals into low- and high-frequency passbands. After envelope detection and HK parameter estimation for each frequency band, the α and k parametric maps were obtained, respectively. According to the contrast between the target region and background, the (α or k) parametric maps of the dual-frequency band were weighted and summed, and then the CWS images were yielded by pseudo-color imaging. The proposed HK CWS parametric imaging algorithm was used to detect the microwave ablation coagulation zones of porcine liver ex vivo under different powers and treatment durations. The performance of the proposed algorithm was compared with that of the conventional HK parametric imaging and frequency diversity and compounding Nakagami imaging algorithms. For two-dimensional HK parametric imaging, it was found that a WSL equal to 4 pulse lengths of the transducer was sufficient for estimating the α and k parameters in terms of both parameter estimation stability and parametric imaging resolution. The HK CWS parametric imaging provided an improved contrast-to-noise ratio over conventional HK parametric imaging, and the HK αcws parametric imaging achieved the best accuracy and Dice score of coagulation zone detection.

A review of the application of three-dimensional convolutional neural networks for the diagnosis of Alzheimer's disease using neuroimaging.

Alzheimer's disease (AD) is a degenerative disorder that leads to progressive, irreversible cognitive decline. To obtain an accurate and timely diagnosis and detect AD at an early stage, numerous approaches based on convolutional neural networks (CNNs) using neuroimaging data have been proposed. Because 3D CNNs can extract more spatial discrimination information than 2D CNNs, they have emerged as a promising research direction in the diagnosis of AD. The aim of this article is to present the current state of the art in the diagnosis of AD using 3D CNN models and neuroimaging modalities, focusing on the 3D CNN architectures and classification methods used, and to highlight potential future research topics. To give the reader a better overview of the content mentioned in this review, we briefly introduce the commonly used imaging datasets and the fundamentals of CNN architectures. Then we carefully analyzed the existing studies on AD diagnosis, which are divided into two levels according to their inputs: 3D subject-level CNNs and 3D patch-level CNNs, highlighting their contributions and significance in the field. In addition, this review discusses the key findings and challenges from the studies and highlights the lessons learned as a roadmap for future research. Finally, we summarize the paper by presenting some major findings, identifying open research challenges, and pointing out future research directions.

A Convolutional Neural Network and Graph Convolutional Network Based Framework for AD Classification.

The neuroscience community has developed many convolutional neural networks (CNNs) for the early detection of Alzheimer's disease (AD). Population graphs are thought of as non-linear structures that capture the relationships between individual subjects represented as nodes, which allows for the simultaneous integration of imaging and non-imaging information as well as individual subjects' features. Graph convolutional networks (GCNs) generalize convolution operations to accommodate non-Euclidean data and aid in the mining of topological information from the population graph for a disease classification task. However, few studies have examined how GCNs' input properties affect AD-staging performance. Therefore, we conducted three experiments in this work. Experiment 1 examined how the inclusion of demographic information in the edge-assigning function affects the classification of AD versus cognitive normal (CN). Experiment 2 was designed to examine the effects of adding various neuropsychological tests to the edge-assigning function on the mild cognitive impairment (MCI) classification. Experiment 3 studied the impact of the edge assignment function. The best result was obtained in Experiment 2 on multi-class classification (AD, MCI, and CN). We applied a novel framework for the diagnosis of AD that integrated CNNs and GCNs into a unified network, taking advantage of the excellent feature extraction capabilities of CNNs and population-graph processing capabilities of GCNs. To learn high-level anatomical features, DenseNet was used; a set of population graphs was represented with nodes defined by imaging features and edge weights determined by different combinations of imaging or/and non-imaging information, and the generated graphs were then fed to the GCNs for classification. Both binary classification and multi-class classification showed improved performance, with an accuracy of 91.6% for AD versus CN, 91.2% for AD versus MCI, 96.8% for MCI versus CN, and 89.4% for multi-class classification. The population graph's imaging features and edge-assigning functions can both significantly affect classification accuracy.

Effects of Patchwise Sampling Strategy to Three-Dimensional Convolutional Neural Network-Based Alzheimer's Disease Classification.

In recent years, the rapid development of artificial intelligence has promoted the widespread application of convolutional neural networks (CNNs) in neuroimaging analysis. Although three-dimensional (3D) CNNs can utilize the spatial information in 3D volumes, there are still some challenges related to high-dimensional features and potential overfitting issues. To overcome these problems, patch-based CNNs have been used, which are beneficial for model generalization. However, it is unclear how the choice of a patchwise sampling strategy affects the performance of the Alzheimer's Disease (AD) classification. To this end, the present work investigates the impact of a patchwise sampling strategy for 3D CNN based AD classification. A 3D framework cascaded by two-stage subnetworks was used for AD classification. The patch-level subnetworks learned feature representations from local image patches, and the subject-level subnetwork combined discriminative feature representations from all patch-level subnetworks to generate a classification score at the subject level. Experiments were conducted to determine the effect of patch partitioning methods, the effect of patch size, and interactions between patch size and training set size for AD classification. With the same data size and identical network structure, the 3D CNN model trained with 48 × 48 × 48 cubic image patches showed the best performance in AD classification (ACC = 89.6%). The model trained with hippocampus-centered, region of interest (ROI)-based image patches showed suboptimal performance. If the pathological features are concentrated only in some regions affected by the disease, the empirically predefined ROI patches might be the right choice. The better performance of cubic image patches compared with cuboidal image patches is likely related to the pathological distribution of AD. The image patch size and training sample size together have a complex influence on the performance of the classification. The size of the image patches should be determined based on the size of the training sample to compensate for noisy labels and the problem of the curse of dimensionality. The conclusions of the present study can serve as a reference for the researchers who wish to develop a superior 3D patch-based CNN model with an appropriate patch sampling strategy.