Strain Parameters Research Articles

A Survey on Efficient Vision Transformers: Algorithms, Techniques, and Performance Benchmarking.

Vision Transformer (ViT) architectures are becoming increasingly popular and widely employed to tackle computer vision applications. Their main feature is the capacity to extract global information through the self-attention mechanism, outperforming earlier convolutional neural networks. However, ViT deployment and performance have grown steadily with their size, number of trainable parameters, and operations. Furthermore, self-attention's computational and memory cost quadratically increases with the image resolution. Generally speaking, it is challenging to employ these architectures in real-world applications due to many hardware and environmental restrictions, such as processing and computational capabilities. Therefore, this survey investigates the most efficient methodologies to ensure sub-optimal estimation performances. More in detail, four efficient categories will be analyzed: compact architecture, pruning, knowledge distillation, and quantization strategies. Moreover, a new metric called Efficient Error Rate has been introduced in order to normalize and compare models' features that affect hardware devices at inference time, such as the number of parameters, bits, FLOPs, and model size. Summarizing, this paper first mathematically defines the strategies used to make Vision Transformer efficient, describes and discusses state-of-the-art methodologies, and analyzes their performances over different application scenarios. Toward the end of this paper, we also discuss open challenges and promising research directions.

From data to dynamics: Reconstructing soliton collision phenomena in optical fibers using a convolutional autoencoder

In this study, a convolutional autoencoder is constructed to extract and reconstruct the dynamical processes of soliton collisions in optical fibers. The model demonstrates exceptional reconstruction capabilities, accurately capturing the evolution of optical event horizons and reproducing nonlinear phenomena such as complex frequency conversions and energy exchange processes. The reconstruction results show high consistency with the numerical simulations, with RMSE values of 0.0220 and 0.0174 in the temporal and frequency domains, respectively. Additionally, by adjusting the training parameters of the convolutional autoencoder model, its reconstruction performance for nonlinear dynamic processes was further validated. This method is expected to provide a different perspective for studying nonlinear phenomena in optical fibers while reducing the consumption of computational resources.

The relationship between crustal deformation characteristics and strong earthquakes in western China revealed by GNSS.

To clarify the relationship between GNSS (Global Navigation Satellite System) crustal deformation and the preparation of strong earthquakes in the western Chinese Mainland, a series of methods including velocity field, strain rate field, and strain parameter time series are adopted based on the GNSS observation data from the CMONOC (Crustal Movement Observation Network of China). The abnormal characteristics of crustal deformation movement before the M ≥ 5 earthquakes in the area since 1999 are studied. The corresponding variety of the crustal movement before earthquakes is discussed in combination with the seismogenic structure of earthquakes. The abnormalities of the GNSS velocity field, strain field, and strain parameter time series before earthquakes with different magnitudes are proposed. Finally, the anomaly criterion before medium-strong-large earthquakes in the western Chinese Mainland is proposed according to the GNSS and its combination with the characteristics of geological structure. The results can provide a basis for the forecasting of the trend and location of medium-strong-large earthquakes in the western Chinese Mainland.

Analysis of deformation evolution characteristics of the northern Lajishan fault before the Jishishan MS6.2 earthquake based on GNSS observations

A MS6.2 earthquake struck Jishishan Country in Gansu Province on 18 December 2023. The earthquake occurred on the eastern branch of the Northern Lajishan fault (NLJSF) which has never experienced such a violent seismic event in recorded history. In this study, multiple periods of GNSS campaign observation data and continuous observation data are collected to reveal the crustal deformation evolution characteristics of the eastern branch of the NLJSF before the earthquake, focusing on the dynamic behavior of the causative fault, regional strain parameter features, and the variation of GNSS baselines across the NLJSF. Velocity field data show that the differential movement on both sides of the fault zone before the earthquake accumulated energy for the occurrence of the earthquake. Cross-fault profile analyses indicate a noticeable weakening of horizontal shortening on both fault flanks, suggesting a nearing fault lockup. The baseline time series and strain parameter time series results both show that the study area is mainly NEE-trending compression deformation. In addition, before the earthquake, multiple baseline results and strain parameter time series deviated from the linear trend and gradually flattened, indicating a strain accumulation in the study area. The overall crustal deformation evolution shows a relatively high earthquake risk before the Lajishan earthquake.

Abstract 4138589: Prognostic implications of Biventricular systolic and diastolic strain in diabetes without coronary artery diseases.

Background: 2D-speckle-tracking echocardiography (2D-STE) has emerged as useful to assess subclinical myocardial dysfunction, however its utility in diabetes milletus (DM) has been limited to the evaluation of left ventricular (LV) longitudinal systolic function. Aim: To assess the differences and prognostic implications of left (LV) and right (RV)ventricular systolic and diastolic strain parameters in DM. Methods: Of the initial 1721 DM patients screened for this study only 513 consecutive patients with DM were eligible and retrospectively enrolled, and 2D-STE strain (longitudinal, circumferential and radial) parameters assessed on the 2D-TTE images. All patients were prospectively followed for development of new outcome/cardiac events which included hospitalization for acute HF, CVS-related death, heart transplantation, and also diabetic complications. The 2D-STE analysis was used to measure the peak LV (longitudinal, radial and circumferential) and RV (longitudinal and radial) strain parameters. The long-term outcomes were evaluated for all-cause CVS outcomes and recurrent hospitalization. Results: For the 513 eligible patients [mean age: 51±8; 62 % male sex], 263 reached the combined endpoint during a mean follow-up of 1.4±0.6years. Conventional TTE was performed in all subjects and reviewed. Both LV and RV strain parameters were significantly impaired compared to normal standard references (p<.0001), where RV was more impaired than LV strain parameters. Th e best performance parameters predictive of CVS events were RV systolic and diastolic strain than global LV longitudinal systolic and diastolic strain. The strongest association between the degree of myocardial dysfunction and CVS risk was more evident for RV strain (free wall > global strain) and lesser extend to LV strain. More myocardial strain impairments were demonstrated in those with higher left atrial volume index, impaired diastolic function, lower strain values ( LV, RV), and reduced tricuspid annular plane systolic excursion, p<.001). RV strain and to a lesser extend LV strain impairment were independent predictors of combined outcomes after multiple adjustments using different models. Conclusion: Patients with Diabetes, RV and LV strain parameters are stronger (RV>LV strain) predictors of outcome than other conventional parameters and should be included for regular clinical screening DM patients as these parameters provide a stronger prognostic stratification

Abstract 4138593: Sex-, Age-and Racial- specific disparities, echocardiographic myocardial mechanics and biomarkers, and 1-year survival among patients with heart failure.

Background: Heart failure with reduced ejection fraction (HFrEF) is a global health burden affecting millions worldwide, particularly among resource constrained communities. Information is limited regarding addative prognostic relations of sex -, age- and racial- differences, myocardial mechanics and biomarkers, in HFrEF. Aim: Assess prognostic implications of sex-, age- and racial- related disparity in HFrEF; and additive prognostic implications of these parameters to other markers including echocardiography (myocardial mechanics). Methodology: The HFrEF patients’ data were prospectively evaluated during the initial assessment and those hospitalized, and subseguently followed for at least 12months. Patients' KCCQ and EQ-5D-5L Questionnaires were assessed during the initial evaluation and follow-up. Sex, age and racial/ethnic disparities, in-hospital management, follow-ups, health status and 1year mortality were assessed. These parameters were further assessed in addition to biomakers and echocardiographic parameters including LV and RV systolic and distolic strain. Multivariable Cox regression models were fitted to investigate additive prognostic differences. Results: The study population consisted mainly of blacks and major factors were hypertension, diabetes, mitral valve disease, and older age (p<0.0001). Of the initial 2642 screened, only 1883 completed the 1year follow-up visit (mean LVEF 38%). Women were likely to have hypertension, atrial fibrillation and diabetes (p<.001). Rheumatic or valvular heart diseases were more commonly identified among younger, black ethnicity, and males patients. No demonstrated sex differences in functional class and biomarkers including NT-proBNP levels. Men demonstrated significant risk for one-year mortality than women, as was black ethnicity and older age. Impaired strain parameters (RV>LV systolic) were independent risk factors for poor survival, more so when added to sex, racial ethnicity, age and biomarkers i.e. the NT-proBNP levels. Conclusion: The prevalence of HFrEF was higher than anticipated. Men, black ethnicity and older age were associated with a high one-year mortality. Impaired strain (RV>LV systolic) parameters were independent risk factors for poor survival, with additive prognostic implications to sex, racial ethnicity, age and traditional biomarkers. These parameters should be assessed routunely to risk stratify HFrEF patients earlier during their disease course.

Abstract 4135061: Myocardial work in different patients with electrical dysynchrony due to left bundle branch block

Left bundle branch block (LBBB) in patients with dilated cardiomyopathy (DCM) is defined by huge wasted work and decreased myocardial constructive work. The aim of this study is to assess the myocardial work in patients in LBBB and other clinical situations. Materials and methods: The echocardiographic study was performed in 25 patients with LBBB and DCM (LBBB-DCM); 20 patients with DCM and narrow QRS complex (nonLBBB-DCM group); 15 patients with LBBB developed after transcatheter aortic valve implantation (TAVI group); 12 patients with idiopathic LBBB, normal systolic LV function and excluded cardiovascular pathology (LBBB group); 27 patients with right ventricular pacing due to atrioventricular block of high grade (RVP group) and 20 healthy volunteers (HV group). In all of them global longitudinal strain (GLS) and myocardial work parameters including global constructive work (GCW), global wasted work (GWW), global work index (GWI) and global work efficiency (GWE) were assessed. In all patients the segments with maximum and minimum GWI value were determined. Results: GLS, GWI and GCW values were comparable in LBBB and RVP groups, whereas in TAVI and in both DCM groups they was significantly lower than in HV (p = 0.049, 0.006, 0.025; p < 0,001 respectively, (Fig.1A,B,C)). Patients from LBBB, TAVI, RVP and LBBB-DCM groups were characterized by increased amount of GWW (261 [203.5; 291.5], 273 [184.8; 385,3] and 237 [149.5; 445.3] vs 108 [74.3; 137.3] mmHg%, p = 0.033, 0.006, < 0.001, < 0.001 respectively). GWW value in LBBB-DCM group was the highest (346.5 [255.5; 437.8] mmHg% (Fig.1D). There was moderate strength of relation between GWW and QRS duration (ρ=0.456, p<0.001). A huge amount of wasted work affected the GWE value and led to its considerable reduction in LBBB, TAVI, RVP and LBBB-DCM groups compared to HV (p = 0.033, 0.007, 0.023, < 0,001, respectively, Fig.1E). The zone of maximum GWI localized in posterolateral wall, the zone of minimum GWI was in interventricular septum in most LBBB patients (Fig.1F). At the same time, these zones turned out to be very variable in RVP group. Conclusion: Although all patients with LBBB pattern of electrical activation are characterized by significant amount of wasted work, its contribution to global myocardial work manifested only in DCM patients. Wasted work doesn’t lead to reduced constructive work in patients with idiopathic LBBB and RVP and normal systolic function probably due to compensatory mechanisms.

Muscular Endurance Training: Bridging Gym Practices with Science Recommendations

This study investigated the effectiveness of training methodologies for improving muscular endurance in young adult males and whether these practices align with scientific recommendations. A cross-sectional descriptive comparative study was conducted with 115 male participants aged 18 to 26. Data on various training parameters were collected using a self-administered questionnaire. The results indicated that participants exhibited discrepancies in their training practices compared to established scientific recommendations. An important percentage of study participants did not exercise according to the recommended guidelines regarding exercise sequencing, intensity, repetitions, rest periods, and training supervision. Participants who followed the recommendations regarding intensity, sets, repetitions, weekly training frequency, and supervision achieved higher objective accomplishment scores. Thus, there are notable discrepancies between everyday gym practices and scientific recommendations for improving muscular endurance in young adult men. Adherence to current recommendations and professional supervision are essential for maximizing training effectiveness, preventing injuries, and facilitating future adaptations.

Myocardial Strain Assessment for Early Duchenne Muscular Dystrophy Diagnosis in Pediatric Patients Using Cardiac MRI

Assessing myocardial strain remains challenging, particularly in the pediatric population, due to the smaller heart sizes, higher heart rates, and variability in strain parameters compared to adult populations. This study aimed to investigate the utility of myocardial strain measurements using cardiac magnetic resonance-feature tracking (CMR-FT) for early diagnosis of Duchenne muscular dystrophy (DMD) in pediatric patients. Twenty-eight DMD patients and 20 healthy controls were involved in this study. Global circumferential, longitudinal, and radial strain (GCS, GLS, and GRS) were measured for the left ventricle (LV) using CMR-FT. Segmental strain values only of the inferolateral and anterolateral LV segments in DMD patients without late gadolinium enhancement (LGE) and DMD patients with LGE were compared to the healthy controls. Strain measurements using CMR-FT in DMD patients were considerably lower than those of healthy controls, with all p-values lower than 0.001. DMD patients without LGE showed decreased inferolateral and anterolateral segmental values only relative to healthy controls. The same behavior was maintained for the LV geometry. Multivariable linear regression demonstrated that the end-systole (ES) wall thicknesses and thickening were associated with decreased GCS and GLS. CMR-FT is crucial in detecting cardiac abnormalities in patients with DMD. It represents an innovative imaging biomarker that can detect initial myocardial alterations in DMD cardiomyopathy without relying on gadolinium.

Acromegaly in Speckle Tracking Echocardiography—A New Cardiac Hypertrophy Phenotype? Case Report and Review

Background: This study aims to provide a comprehensive overview of speckle tracking echocardiography (STE) findings in patients diagnosed with acromegaly, exploring a potential application for the differential diagnosis of cardiac hypertrophy and guiding clinicians in patient management. To our knowledge, this is the first review showcasing changes in the bull’s-eye pattern in myocardial function after acromegaly treatment, suggesting a possible pattern in this aetiology of left ventricular hypertrophy. Methods: A review of PubMed articles using the search term “speckle tracking echocardiography acromegaly” yielded 11 relevant papers published between 2017 and 2023. The final analysis evaluated each paper’s substantial value and summarised pertinent information. A clinical case of myocardial strain changes before and after pharmacological and neurosurgical treatment is also described. Results: The 11 analysed papers, with patient groups ranging from 19 to 50 individuals, revealed varying results in STE parameters between acromegalic and control groups. Left ventricular strain parameters were commonly assessed, showing discrepancies in different studies. Notably, the MAGYAR-Path Study emphasised left ventricular twist and radial strain abnormalities in acromegaly patients. Furthermore, the interplay between acromegaly and diabetes influenced myocardial function. Drug treatment with somatostatin receptor ligands demonstrated a favourable impact on left ventricular systolic function. The case study we describe in this manuscript showed changes in the posterior basal LV segment, which may be a specific pattern of acromegaly remodelling. Somatostatin therapy and neurosurgery led to the normalisation of global longitudinal strain (GLS) and improvement in myocardial work, as well as improved diastolic function, including enhanced left atrial strain (LAS) as well as a visible elastic recoil sign (ERS). Conclusions: While the available literature on STE in acromegaly is limited, our analysis suggests potential applications in differentiating hypertrophy aetiologies and monitoring cardiac function post-treatment. The results underscore the need for more interdisciplinary research to optimise patient management. The bull’s-eye pattern with posterior basal segment strain impairment we describe may be used to differentiate this entity.

DeepRFQC: automating quality control for P-wave receiver function analysis using a U-net inspired network

This paper introduces DeepRFQC, an automated method for quality control in P-wave receiver function analysis. Leveraging a U-Net inspired deep learning model, which has previously shown promise in denoising and phase detection, DeepRFQC efficiently distinguishes usable from noisy receiver functions. We examine a Proterozoic Trans-Hudson Orogen dataset from northern Canada, including seismic events from 1990 to 2023, which is expanded for training purposes by data augmentation techniques. With 1,508,449 trainable parameters, the DeepRFQC model attains a commendable 96.6% validation accuracy, on a test dataset from the X5 seismic network; tests on stations from different tectonic environments indicate that the model is effective even in environments very different from the training set. Validation through the H-κ stacking method shows consistent and plausible results. As manual quality control is a major bottleneck in receiver-function processing, automated methods such as this one will allow for efficient examination of large data sets.

Forecasting extremes of football players’ performance in matches

This study evaluates the use of simple linear or piecewise linear predictive models to predict extreme performance metrics in soccer matches, based on historical training and to match data of soccer players from RKS Raków Częstochowa football club. The data were collected from January to June 2023. The collected training and matched data average is 9000 records per month. A standard workweek at the RKS Academy consisted of 5 training units and at least 1 match. The best individual models found predict selected game performance metrics with a relative error of 2.3%, suggesting an excellent model fit between prediction and the actual value. This is illustrated by input data metric called “Metabolic Time Zone 5 and 6 Per Distance”, and output data by “Decelarations Total Distance in Zone 5 and 6 Per Distance”—calculated for in 3 min sliding window and characterized by the highest value of the generated parameter based on High Metabolic Load Distance (HMLD). The result concerns models run on aggregated performance metrics developed in APEX-PRO system using expert knowledge in soccer training, while raw GPS location-based models performing worse but still acceptably. Although we believe that the accuracy of the models still has limited reliability, their clarity and up-to-date quality make them useful in the daily planning of training activities and the management of workloads that affect player performance in the upcoming match, as well as the tactical decisions of the coach. More accurate predictions are given by individual models compared to aggregated models (player position), but there are exceptions where group models also perform very well. Adding a second metric to the input did not show a significant difference in the analyzed examples (the results are very similar). Our findings indicate that the model based on metrics from the last match also effectively predict extreme motor performances occurring in the game. In the case of the analyzed player, it was at the input “Accelerations Total Time Per Distance in Zone 6” at the output “Distance in Zone 6”. Specific training or match parameters can be key in predicting exceptional soccer performance, but they can also vary depending on the analyzed player. This confirms the need for further analysis of this issue.

Left Atrial Structure and Function Following Ethanol Infusion into Vein of Marshall (MR-SHALL Study).

Ethanol infusion into the vein of Marshall (EI-VOM) is an adjunctive therapy to pulmonary vein isolation (PVI), which improves the efficacy of persistent atrial fibrillation (AF) ablation procedures. It is unknown how EI-VOM scar formation impacts left atrium (LA) structure and function. To characterize scar formation after EI-VOM, and its impact on LA function, by cardiac magnetic resonance (CMR). Consecutive patients referred for persistent AF ablation, or any repeat procedure, were included. Ablation strategy included PVI, EI-VOM and linear lesions when deemed necessary. LA was assessed by late gadolinium enhancement (LGE) CMR at baseline and 3-month follow-up. CMR was post-processed off-line with dedicated software. The LA was regionalized into 17 segments, and global and regional scarring were calculated. Strain parameters were analyzed-including LA reservoir, conduit, and booster strain; LA ejection fraction; and LA passive and active emptying function. Follow-up LGE-CMR reconstruction was coregistered with the voltage map, ablation RF points, and fluoroscopy to compare the scar location with the VOM trajectory. This study included 21 patients (65 ± 8 years; 76% men; 57% ablation of persistent AF). After EI-VOM, 95% of patients presented new low-voltage areas, all of whom exhibited new LGE-CMR scarring at follow-up. Voltage map and LGE-CMR coregistration showed comparable scar areas (p = 0.287). LGE-CMR scarring frequently involved the whole VOM trajectory, predominantly affecting the LA ridge, mitral isthmus, and left posterior wall. LA volume was significantly reduced after ablation, without significant changes in LA functional parameters. EI-VOM causes permanent scarring at 3-month follow-up, with no adverse impact on CMR LA functional parameters.

Correcting the distribution of batch normalization signals for Trojan mitigation

Backdoor (Trojan) attacks represent a significant adversarial threat to deep neural networks (DNNs). In such attacks, the presence of an attacker’s backdoor trigger causes a test instance to be misclassified into the attacker’s chosen target class. Post-training mitigation methods aim to rectify these misclassifications, ensuring that poisoned models correctly classify backdoor-triggered samples. These methods require the defender to have access to a small, clean dataset and the potentially compromised DNN. However, most defenses rely on parameter fine-tuning, making their effectiveness dependent on the dataset size available to the defender. To overcome the limitations of existing approaches, we propose a method that rectifies misclassifications by correcting the altered distribution of internal layer activations of backdoor-triggered instances. Distribution alterations are corrected by applying simple transformations to internal activations. Notably, our method does not modify any trainable parameters of the DNN, yet it achieves generally good mitigation performance against various backdoor attacks and benchmarks. Consequently, our approach demonstrates robustness even with a limited amount of clean data, making it highly practical for real-world applications. The effectiveness of our approach is validated through both theoretical analysis and extensive experimentation. The appendix is provided as an electronic component and can be accessed via the link in the footnote.22https://arxiv.org/pdf/2308.09850. The source codes can be found in the link33https://github.com/lixi1994/BNA. at the footnote.

Non-small cell lung cancer detection through knowledge distillation approach with teaching assistant.

Non-small cell lung cancer (NSCLC) exhibits a comparatively slower rate of metastasis in contrast to small cell lung cancer, contributing to approximately 85% of the global patient population. In this work, leveraging CT scan images, we deploy a knowledge distillation technique within teaching assistant (TA) and student frameworks for NSCLC classification. We employed various deep learning models, CNN, VGG19, ResNet152v2, Swin, CCT, and ViT, and assigned roles as teacher, teaching assistant and student. Evaluation underscores exceptional model performance in performance metrics achieved via cost-sensitive learning and precise hyperparameter (alpha and temperature) fine-tuning, highlighting the model's efficiency in lung cancer tumor prediction and classification. The applied TA (ResNet152) and student (CNN) models achieved 90.99% and 94.53% test accuracies, respectively, with optimal hyperparameters (alpha = 0.7 and temperature = 7). The implementation of the TA framework improves the overall performance of the student model. After obtaining Shapley values, explainable AI is applied with a partition explainer to check each class's contribution, further enhancing the transparency of the implemented deep learning techniques. Finally, a web application designed to make it user-friendly and classify lung types in recently captured images. The execution of the three-stage knowledge distillation technique proved efficient with significantly reduced trainable parameters and training time applicable for memory-constrained edge devices.

The Impact of Different Revascularization Strategies Implemented in Acute Myocardial Infarction on the Recovery of Left Ventricular Functional and Deformational Parameters

Background: Despite well-established therapeutic techniques, such as direct revascularization through percutaneous coronary intervention (PCI), acute myocardial infarction (AMI) remains a leading cause of mortality and morbidity. Objectives: To determine if two-dimensional speckle tracking echocardiography (STE) deformation parameters and the early recovery of left ventricular (LV) functions are affected by the timing of PCI in AMI. Methods: A total of 200 cases with newly-onset acute myocardial infarction (AMI) who had a baseline left ventricular ejection fraction (LVEF) higher than 40% and received effective therapy with percutaneous coronary intervention (PCI) were included in this investigation. cases were categorized as either ST-elevation myocardial infarction (STEMI) or non-ST-elevation myocardial infarction (NSTEMI). cases were grouped into four groups according to the time between presentation and PCI. Using standard echocardiography and two-dimensional (2D) ST-elevation STE, individuals were re-evaluated three months later to find out if remodeling had taken place or if the LV function had returned. Results: Of the 200 AMI patients, including 140 males (70%), improvement in global longitudinal strain (GLS) and harmed longitudinal strain (HLS) were better in STEMI and NSTEMI patients received urgent revascularization with PCI (groups I and III) versus patients with pharmacoinvasive strategy or routine invasive strategy (Groups II and IV) (P < 0.05) while there was an insignificant difference between group I and III (P = 0.79). Of the 200 patients, 47 patients (23.5%) presented signs of LV remodeling at 3 months follow up. Age, smoking history, hypertension, dyslipidemia, Killip class, peak creatine phosphokinase - MB level, baseline left ventricular end diastolic volume (LVEDV), HLS, and harmed longitudinal strain rate (HLSR) were all factors that were found to be significantly associated with left ventricular remodeling (P<0.05) in the univariate logistic regression analysis. The following factors were identified as independent predictors of left ventricular remodeling in multivariate logistic regression analysis: damaged left ventricular ejection fraction (EF) and end-systolic volume, peak troponin I, Killip class, culprit left anterior descending (LAD), 2 and 3-vessel coronary artery disease (CAD), and wall motion score index (WMSI). Conclusion: Earlier PCI in AMI helps earlier improvement in myocardial strain parameters. HLS and HLSR are excellent predictors for LV remodeling and may do better than global parameters.

Global, segmental, and layer-specific two-dimensional speckle tracking echocardiography immediately after acute myocardial infarction as a predictive tool to assess myocardial viability and scar size.

The identification of myocardial scar is key in clinical decision-making after acute myocardial infarction (AMI). However, the gold standard that is cardiac magnetic resonance imaging (CMR) encounters limitations in terms of availability. Two-dimensional speckle tracking echocardiography (2D-STE) may be an accessible alternative in detecting scar and assessing scar transmurality. We aim to evaluate the predictive value of 2D-STE, encompassing measures of global, segmental and layer-specific strain, with respect to myocardial viability and scar size at 6 months follow-up. In 43 patients admitted for primary AMI, we conducted a comparative analysis of strain parameters (including global longitudinal strain (GLS), segmental longitudinal strain (SLS), layer-specific GLS and SLS and the transmural strain gradient from endocardium to epicardium) in relation to conventional echocardiographic parameters at baseline in predicting for scar size and the transmurality index, as measured by CMR, 6 months post enrollment. We demonstrate a moderate correlation between both GLS and conventional echocardiographic parameters, and scar size as well as transmurality index. Wall motion score index exhibited superior predictive performance over GLS and left ventricular ejection fraction in anticipating scar formation. At a cut-off of -13.3% for any scar and -11.5% for transmural scar, SLS can predict scar formation. Layer-specific strain did not provide added predictive value. SLS, but not layer-specific strain, during admission after AMI is an easy and accessible quantitative tool for predicting scar formation and transmurality extent at 6 months follow-up. GLS correlates well with scar size, suggesting its potential utility as a predictive tool.

Intelligent Fault Diagnosis Method Based on Neural Network Compression for Rolling Bearings

Rolling bearings are often exposed to high speeds and pressures, leading to the symmetry in their rotating structure being disrupted, which can lead to serious failures. Intelligent rolling bearing fault diagnosis is a critical part of ensuring operation of machinery, and it has been facilitated by the growing popularity of convolutional neural networks (CNNs). The outstanding performance of fault diagnosis CNNs results from complex and redundant network structures and parameters, resulting in huge storage and computational requirements, which makes it challenging to implement these models in resource-limited industrial devices. This study aims to address this problem by proposing a comprehensive compression method for CNNs that is applied to intelligent fault diagnosis. It involves several different compression methods, including tensor train decomposition, parameter quantization, and knowledge distillation for deep network compression. This results in a significant decrease in redundancy and speeding up the training of CNN models. Firstly, tensor train decomposition is applied to reduce redundant connections in both convolutional and fully connected layers. The next step is to perform parameter quantization to minimize the bits needed for parameter representation and storage. Finally, knowledge distillation is used to restore accuracy to the compressed model. The effectiveness of the proposed approach is confirmed by an experiment and ablation study with different models on several datasets. The results show that it can significantly reduce redundant information and floating-point operations with little degradation in accuracy. Notably, on the CWRU dataset, with about 60% parameter reduction, there is no degradation in our model’s accuracy. The proposed approach is a new attempt at the intelligent fault diagnosis of rolling bearings in industrial equipment.

Generative artificial intelligence (GAI): From large language models (LLMs) to multimodal applications towards fine tuning of models, implications, investigations

This research explores the transformative integration of artificial intelligence (AI), robotics, and language models, with a particular emphasis on the PaLM-E model. The exploration aims to assess PaLM-E’s decision-making processes and adaptability across various robotic environments, demonstrating its capacity to convert textual prompts into very precise robotic actions. In addition, the research investigates Parameter-Efficient Fine-Tuning (PEFT) techniques, such as Low-Rank Adaptation (LoRA) and Quantized Low-Rank Adaptation (QLoRA), providing a historical overview of PEFT and highlighting their significance in enhancing task performance while reducing the number of trainable parameters. The broader scope of Generative AI is examined through an analysis of influential models like GPT-3, GPT-4, Copilot, Bard, LLaMA, Stable Diffusion, Midjourney, and DALL-E. These models’ abilities to process natural language prompts and generate a wide range of outputs are thoroughly investigated. The research traces the historical evolution of AI, from its roots in science fiction to its practical applications today, with a focus on the rise of Generative AI in the 21st century. Furthermore, the research delves into the various modalities of Generative AI, covering applications in text, code, images, and more, and assesses their real-world impact on robotics, planning, and business intelligence. The implications of synthetic data generation for business analytics are also explored. The research inspects within both software and hardware landscapes, comparing local deployment on consumer-grade hardware along with cloud-based services, and underscores the benefits of local model deployment in terms of privacy protection, intellectual property security, and censorship resistance. Ethical considerations are central to this research, addressing concerns related to privacy, security, societal impact, biases, and misinformation. The research proposes ethical guidelines for the responsible development and deployment of AI technologies. Ultimately, this work reveals the deep interconnections between vision, language, and robotics, pushing the boundaries of AI capabilities and providing crucial insights for future AI model development and technological innovation. These findings are intended to guide the field through the emerging challenges of the rapidly evolving Generative AI landscape.

Insights into the Associations Between Systolic Left Ventricular Rotational Mechanics and Left Atrial Peak Reservoir Strains in Healthy Adults from the MAGYAR-Healthy Study

Introduction: In systole, when the left ventricle (LV) twists, the left atrium (LA) behaves like a reservoir, having a special wall contractility pattern opposite to that of the LV wall. Accordingly, the objective of the present study was to investigate the associations between LV rotational mechanics and LA peak (reservoir) strains as assessed simultaneously by three-dimensional speckle-tracking echocardiography (3DSTE) under healthy conditions. Methods: In the present study, 157 healthy adults (mean age: 33.2 ± 12.7 years, 73 men) were involved. Complete two-dimensional Doppler echocardiography with 3DSTE-derived data acquisition were performed in all cases. The 3DSTE-derived LV rotational and LA strain parameters were determined at a later date. Results: Global LA peak reservoir circumferential (22.7 ± 6.4% vs. 27.6 ± 6.8%, p < 0.05) and area (57.8 ± 20.0% vs. 66.0 ± 22.7%, p < 0.05) strains proved to be reduced in the case of the highest vs. lowest basal LV rotation; other LA peak reservoir strains were not associated with increasing basal LV rotation. Global LA peak radial strain was highest in the case of the lowest vs. highest apical LV rotation (−19.2 ± 9.4% vs. −13.0 ± 8.2%, p < 0.05). Global LA peak reservoir 3D strain was lowest in the case of the highest vs. lowest apical LV rotation (−9.9 ± 6.8% vs. −5.0 ± 4.2%, p < 0.05). Only apical LV rotation proved to be significantly reduced in the case of the highest vs. lowest global LA peak reservoir 3D strain (8.12 ± 3.23° vs. 10.50 ± 3.44°, p < 0.05). Other global LA peak reservoir strains were not associated with basal and apical LV rotations. Conclusions: In LV systole, LV rotational mechanics is associated with LA deformation represented by LA peak (reservoir) strains even in healthy circumstances. While basal LV rotation is associated with LA widening, apical LV rotation is associated with LA thinning, suggesting the close cooperation of the LV and LA in systole even in healthy adults.