Showing papers in "Biocybernetics and Biomedical Engineering in 2022"

TL;DR: In this paper , the authors provided the Sperm Videos and Images Analysis (SVIA) dataset, including three different subsets, including subset-A, subset-B and subset-C, to test and evaluate different computer vision techniques in computer aided sperm analysis.

TL;DR: In this paper , transfer learning has been applied to a wide range of medical imaging tasks, such as segmentation, object identification, disease categorization, severity grading, to name a few.

TL;DR: In this paper , a new stacked convolutional neural network model was designed for the automatic diagnosis of COVID-19 disease from the chest X-ray and CT images, which achieved a sensitivity of 97.62% for the multi-class classification of Xray images into COVID19, Normal and Pneumonia Classes and 98.31% sensitivity for binary classification of CT images into CT images and no-Finding classes.

TL;DR: In this article , a novel combination of Stationary Wavelet transforms (SWT) and a two-stage median filter with Savitzky-Golay (SG) filter were used for preprocessing of the ECG signal followed by segmentation and z-score normalisation process.

TL;DR: In this paper , the role of artificial intelligence (AI) in recent advances on IoMT is reviewed and a comprehensive list of major benefits and challenges is presented as well. And the WMDs classification is also performed based on their technology.

TL;DR: In this paper , a hybrid CNN and bidirectional LSTM (Bi-LSTM) based EMGHandNet architecture was proposed to encode the inter-channel and temporal dependencies of sEMG signals for hand activity classification.

TL;DR: In this article , a novel hybridization of the oscillatory modes decomposition, features mining based on the Second Order Difference Plots (SODPs) of oscillatory mode, and machine learning algorithms is devised for an effective identification of alcoholism.

TL;DR: In this paper , a capsule network is applied to ECG signal classification to improve the classification accuracy of electrocardiogram (ECG) signals for diagnosing heart abnormalities and arrhythmias and preventing cardiovascular diseases (CVDs).

TL;DR: In this paper , the authors presented a novel mix of Level-Crossing Sampling (LCS), Enhanced Activity Selection (EAS) based QRS complex selection, multirate processing, Wavelet Decomposition (WD), Metaheuristic Optimization (MO), and machine learning.

TL;DR: In this paper , the authors proposed a novel biometric authentication scheme based on ECG detection called BAED, which was developed based on deep learning algorithms, including a convolutional neural network (CNN) and a long-term memory (LSTM) network with a customized activation function.

TL;DR: In this paper , a two-stage framework is proposed to perform an accurate classification of diverse voice pathologies, which considers impaired voice as a noisy signal and uses the noise lestral harmonic-to-noise ratio (CHNR) to put this hypothesis into practice, the second stage consists of a CNN-LSTM architecture designed to learn complex features from spectrograms of the first-stage enhanced signals.

TL;DR: Li et al. as mentioned in this paper used dropout in the convolutional part of the network to detect pneumonia in chest X-ray images from retrospective cohorts of pediatric patients from Guangzhou Women and Children's Medical Center, Guangzhou, China.

TL;DR: In this article , the authors proposed a new framework for the accurate classification of cervical cells, which comprises three phases: segmentation, localization of nucleus, and classification, which achieved an accuracy of 99.12 %, specificity of 0.45 %, and sensitivity of 1.25 % with an execution time 99.6248 on SIPaKMed.

TL;DR: In this article , a method based on the combination of signal decomposition and statistical methods was proposed to realize the detection and classification of epileptic seizure, which achieved an accuracy of 98.86%, 98.37, 98.41%, 99.41% and 99.57% for the TUSZ corpus in the TUH EEG corpus.

TL;DR: In this paper , three state-of-the-art deep learning (DL) architectures, experimenting with convolutional, residual, and attention (Transformers) approaches to classify subjects with DM from diabetic foot thermography images, were trained under three conditions of data augmentation.

TL;DR: In this paper , an ensemble of two CNN architectures integrated with Channel and Spatial attention was proposed for breast cancer classification, which achieved a classification accuracy of 99.55% on the BreakHis dataset.

TL;DR: In this article , a cyber-attack and anomaly detection model based on recursive feature elimination (RFE) and multilayer perceptron (MLP) was developed for the Internet of Medical Things (IoMT).

TL;DR: In this paper , Deep CNN and LSTM Architecture (DCNN-LSTM)-based automated diagnosis system is proposed for detecting Congestive Heart Failure (CHF) using ECG signals.

TL;DR: In this paper , a deep learning model was proposed for detecting Parkinson's disease from speech signals. But the performance of the proposed model was verified on two databases, and the accuracy of up to 92% was obtained on the speech tasks, which included reading simple (/loslibros/) and complex (/viste/) sentences in Spanish.

TL;DR: In this paper , a fuzzy c-median clustering method is used to determine the amplitude threshold values for each sliding window across the composed detection function waveform, and then the identified peaks are evaluated on the basis of the speed of rising and falling slopes of detection function peak.

TL;DR: In this article , the authors formulated primary protein sequences using evolutionary and sequence-based numerical descriptors, whereas, evolutionary features are collected using a bigram Position-specific scoring matrix, besides, K-space amino acid pair (KSAAP) and dipeptide composition are utilized to extract sequential information.

TL;DR: In this paper , a completely automated HPT detection system is proposed using time-frequency (T-F) spectral images and a convolutional neural network (CNN) for the accurate detection of HPT using BCG signals.

TL;DR: In this paper , a dual-channel asymmetric convolutional neural network (CNN) was proposed to segment retinal blood vessels based on the morphological differences in retinal vessels.

TL;DR: Wang et al. as discussed by the authors proposed an efficient U-Net (e-UNet) for segmenting chronic stroke lesions, which incorporates a depth-wise convolution-based e-block designed to efficiently reduce the trainable parameters.

TL;DR: In this article , an electromyography (EMG) controlled hand exoskeleton for basic movements in assisted bilateral therapy, where bimanual work is required by the user, is presented.

TL;DR: In this article , a pyramidal one-dimensional convolutional neural network (1DCNN) is designed to deal with 1D EEG signals and trained on the augmented training set that consists of both original and generated EEG data.

TL;DR: In this article , the authors proposed an end-to-end deep learning approach, where the input images are fed directly to the deep model for final decision, which achieved an accuracy of 99.99% compared with other deep learning models.

TL;DR: Zhang et al. as discussed by the authors proposed a two-stage transfer learning recognition model for medical images of COVID-19 (TL-Med), which can effectively learn the information between different modal images.

TL;DR: Wang et al. as mentioned in this paper combined SHAP value with four classifiers, namely deep forest (gcForest), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and random forest (RF), respectively, and applied them to Parkinson's disease diagnosis.

TL;DR: In this article , a Focal-Generalized classification method is proposed that compares traditional classification algorithms and deep learning methods. And the proposed method was applied on 23 subjects in the Temple University Hospital (TUH) scalp EEG data set, and a classification success rate of 95,92% for case (I) and 98,08% was successfully achieved with deep learning architecture LSTM.