Background— The identification of sites of dominant activation frequency during atrial fibrillation (AF) in humans and the effect of ablation at these sites have not been reported. Methods and Results— Thirty-two patients undergoing AF ablation (19 paroxysmal, 13 permanent) during ongoing arrhythmia were studied. Electroanatomic mapping was performed, acquiring 126±13 points per patient throughout both atria and coronary sinus. At each point, 5-second electrograms were obtained to determine the highest-amplitude frequency on spectral analysis and to construct 3D dominant frequency (DF) maps. The temporal stability of the recording interval was confirmed in a subset. Ablation was performed with the operator blinded to the DF maps. The effect of ablation at sites with or without high-frequency DF sites (maximal frequencies surrounded by a decreasing frequency gradient ≥20%) was evaluated by determining the change in AF cycle length (AFCL) and the termination and inducibility of AF. The spatial distribution ...

Spectral Analysis Identifies Sites of High-Frequency Activity Maintaining Atrial Fibrillation in Humans

Low back pain during pregnancy.

Compared with common deep learning methods (e.g., convolutional neural networks), transfer learning is characterized by simplicity, efficiency and its low training cost, breaking the curse of small datasets. Medical image analysis plays an indispensable role in both scientific research and clinical diagnosis. Common medical image acquisition methods include Computer Tomography (CT), Magnetic Resonance Imaging (MRI), Ultrasound (US), X-Ray, etc. Although these medical imaging methods can be applied for non-invasive qualitative and quantitative analysis of patients—compared with image datasets in other computer vision fields such like faces—medical images, especially its labeling, is still scarce and insufficient. Therefore, more and more researchers adopted transfer learning for medical image processing. In this study, after reviewing one hundred representative papers from IEEE, Elsevier, Google Scholar, Web of Science and various sources published from 2000 to 2020, a comprehensive review is presented, including (i) structure of CNN, (ii) background knowledge of transfer learning, (iii) different types of strategies performing transfer learning, (iv) application of transfer learning in various sub-fields of medical image analysis, and (v) discussion on the future prospect of transfer learning in the field of medical image analysis. Through this review paper, beginners could receive an overall and systematic knowledge of transfer learning application in medical image analysis. And policymaker of related realm will benefit from the summary of the trend of transfer learning in medical imaging field and may be encouraged to make policy positive to the future development of transfer learning in the field of medical image analysis.

A Review of Deep Learning on Medical Image Analysis

Electrocardiograms (ECGs) play a vital role in the clinical diagnosis of heart diseases. An ECG record of the heart signal over time can be used to discover numerous arrhythmias. Our work is based on 15 different classes from the MIT-BIH arrhythmia dataset. But the MIT-BIH dataset is strongly imbalanced, which impairs the accuracy of deep learning models. We propose a novel data-augmentation technique using generative adversarial networks (GANs) to restore the balance of the dataset. Two deep learning approaches—an end-to-end approach and a two-stage hierarchical approach—based on deep convolutional neural networks (CNNs) are used to eliminate hand-engineering features by combining feature extraction, feature reduction, and classification into a single learning method. Results show that augmenting the original imbalanced dataset with generated heartbeats by using the proposed techniques more effectively improves the performance of ECG classification than using the same techniques trained only with the original dataset. Furthermore, we demonstrate that augmenting the heartbeats using GANs outperforms other common data augmentation techniques. Our experiments with these techniques achieved overall accuracy above 98.0%, precision above 90.0%, specificity above 97.4%, and sensitivity above 97.7% after the dataset had been balanced using GANs, results that outperform several other ECG classification methods.

/pdf/generalization-of-convolutional-neural-networks-for-ecg-2gyl7sadq0.pdf

Generalization of Convolutional Neural Networks for ECG Classification Using Generative Adversarial Networks

BACKGROUND: Coronavirus disease 2019 (COVID-19) has become a pandemic since its first appearance in late 2019 Deaths caused by COVID-19 are still increasing day by day and early diagnosis has become crucial Since current diagnostic methods have many disadvantages, new investigations are needed to improve the performance of diagnosis METHODS: A novel method is proposed to automatically diagnose COVID-19 by using Electrocardiogram (ECG) data with deep learning for the first time Moreover, a new and effective method called hexaxial feature mapping is proposed to represent 12-lead ECG to 2D colorful images Gray-Level Co-Occurrence Matrix (GLCM) method is used to extract features and generate hexaxial mapping images These generated images are then fed into a new Convolutional Neural Network (CNN) architecture to diagnose COVID-19 RESULTS: Two different classification scenarios are conducted on a publicly available paper-based ECG image dataset to reveal the diagnostic capability and performance of the proposed approach In the first scenario, ECG data labeled as COVID-19 and No-Findings (normal) are classified to evaluate COVID-19 classification ability According to results, the proposed approach provides encouraging COVID-19 detection performance with an accuracy of 9620% and F1-Score of 9630% In the second scenario, ECG data labeled as Negative (normal, abnormal, and myocardial infarction) and Positive (COVID-19) are classified to evaluate COVID-19 diagnostic ability The experimental results demonstrated that the proposed approach provides satisfactory COVID-19 prediction performance with an accuracy of 9300% and F1-Score of 9320% Furthermore, different experimental studies are conducted to evaluate the robustness of the proposed approach CONCLUSION: Automatic detection of cardiovascular changes caused by COVID-19 can be possible with a deep learning framework through ECG data This not only proves the presence of cardiovascular changes caused by COVID-19 but also reveals that ECG can potentially be used in the diagnosis of COVID-19 We believe the proposed study may provide a crucial decision-making system for healthcare professionals SOURCE CODE: All source codes are made publicly available at: https://githubcom/mkfzdmr/COVID-19-ECG-Classification

/pdf/classification-of-covid-19-electrocardiograms-by-using-4w5tdi4bfm.pdf

Classification of COVID-19 electrocardiograms by using hexaxial feature mapping and deep learning.

Electrocardiogram (ECG) is one of the most important and effective tools in clinical routine to assess the cardiac arrhythmias. In this research higherorder spectral estimations, bispectrum and third-order cumulants, are evaluated, saved, and pre-trained using convolutional neural networks (CNN) algorithm. CNN is transferred in this study to carry out automatic ECG arrhythmia diagnostics after employing the higher-order spectral algorithms. Transfer learning strategies are applied on pre-trained convolutional neural network, namely AlexNet and GoogleNet, to carry out the final classification. Five different arrhythmias of ECG waveform are chosen from the MIT-BIH arrhythmia database to evaluate the proposed approach. The main contribution of this study is to utilize the pre-trained convolutional neural networks with a combination of higher-order spectral estimations of arrhythmias ECG signal to implement a reliable and applicable deep learning classification technique. The Highest average accuracy obtained is 97.8 % when using third cumulants and GoogleNet. As is evident from these results, the proposed approach is an efficient automatic cardiac arrhythmia classification method and provides a reliable recognition system based on well-established CNN architectures instead of training a deep CNN from scratch.

/pdf/ecg-classification-using-higher-order-spectral-estimation-4xk0foafsr.pdf

Ecg classification using higher order spectral estimation and deep learning techniques

Sound Source Localization refers to the ability to identify the location of a certain sound in direction and distance from a reference point. This paper introduces a sound source direction estimation system for an acoustical signal of interest in a 360°sweep in the horizontal plane. In this paper, an array of microphones is used as receivers for the sound waves which are recorded with prescribed blocks size of samples and interfaced to a PC for analysis using digital signal processing techniques. Spectral analysis and wavelet transform are applied on a reference microphone to test if a signal of interest has been received. If so, further processing for signals received on all microphones is initiated to estimate the sound source direction. Generalized Cross Correlation is used in this research for time delay estimation among the microphones since it has low computation complexity, sufficient estimation precision and it works very well in reverberant environments.

Sound source direction estimation in horizontal plane using microphone array

Modeling of biomedical signals is crucial and vital for compression, transmission, understanding, feature extraction, and prediction. Researchers devoted their efforts on modeling cardiac related biosignals such as electrocardiography (ECG), photoplethysmography (PPG), and arterial blood pressure(APP) without concerning other types of non-cardiac related biosignals such as electrooculogram (EOG). This paper focus on the modeling of EOG signal using a novel method that exploits a linear combination of seven gaussian basis functions. The proposed method succeeded on modeling of fifteen different eye movements directions combined from right, left, and center with average root mean squared (RMS) error of I%.

Gaussian Model of Electrooculogram Signals

The purpose of this work is to investigate the effect of anteriorly-added mass to simulate pregnancy on lower extremities kinematic and lumbar and thoracic angles during stair ascending and descending. 18 healthy females ascended and descended, with and without a pseudo-pregnancy sac of 12 kg (experimental and control groups, respectively), a costume-made wooden staircase while instrumented with 20 reflective markers placed on the lower extremities and the spine. The movements were captured by 12 infrared cameras surrounding the staircase. Tracked position data were exported to MATLAB to calculate the required joints angles. SPSS was used to compare the ascent and descent phases of control group, and to find if there are any significant differences between control and experimental groups in the ascent phase as well as in the descent phase. When comparing the ascent and descent phases of control group, data revealed a higher hip flexion during ascending and greater ankle planter-flexion and dorsiflexion, lumbar, and thoracic angles during descending; however, no significant difference was shown in the knee flexion angle between ascending and descending. Non-pregnant data showed greater maximum hip flexion and ankle dorsiflexion during stair ascending compared to simulated-pregnant group; while ankle planter-flexion, knee flexion, and lumbar angle were greater for simulated-pregnant status. During stair descending, non-pregnant group had greater minimum hip flexion and ankle dorsiflexion compared to simulated pregnant group; while ankle planter-flexion, knee flexion, and maximum hip flexion were greater for simulated-pregnant group. However, the lumbar and thoracic angles were found to be similar for simulated-pregnant and non-pregnant groups during stair descending. In conclusion, the current study revealed important kinematic modifications pregnant women adopt while ascending and descending stairs at their final stage of pregnancy to increase their stability.

Spinal Curvature and Lower Extremities Kinematics of Simulated Pregnancy during Stair Ascending and Descending

Background: Handgrip or Grip strength (GS) is a common method used to evaluate muscle strength and affected by different factors, including age, gender, and arm’s positions. Objective: This study aims to investigate the effect of both the gender and arm’s positions on the handgrip strength and the fatigue resistance (FR), which is the time needed for the handgrip strength to drop to 75% (FR75), 50% (FR50), and 25% (FR25) of its maximum strength during sustained maximal handgrip effort. Material and Methods: In this experimental study, 59 male and 41 female participants were asked to grip forcefully on a dynamometer for the longest period. GS and FR75, FR50, and FR25 values were recorded for 7 different arm positions. Factorial ANOVA was used to find the main effect of gender and position and the interaction between them. Sidak and Tukey’s HSD tests were used to find the gender and arm position effects, respectively. Results: The results showed a significant effect for gender and arm position on GS and FR and a significant interaction effect for GS that was significantly higher in males than females for all positions. The gender difference in FR depends on arm’s positions and the level at which the FR was measured. GS was higher when arm adduction with 90 ͦ forward at the elbow as compared to arm abduction with 180 ͦ at the shoulder and 90 ͦ at the elbow. Conclusion: The results confirmed the significant effect of the gender and arm’s positions on the maximal handgrip strength and fatigue resistance during sustained maximal handgrip effort.

/pdf/the-effect-of-gender-and-arm-anatomical-position-on-the-hand-1x7ye08b.pdf

Sami Almashaqbeh

Papers

Ecg classification using higher order spectral estimation and deep learning techniques

Sound source direction estimation in horizontal plane using microphone array

Gaussian Model of Electrooculogram Signals

Spinal Curvature and Lower Extremities Kinematics of Simulated Pregnancy during Stair Ascending and Descending

The Effect of Gender and Arm Anatomical Position on the Hand Grip Strength and Fatigue Resistance during Sustained Maximal Handgrip Effort