Advanced Time-Frequency Methods for ECG Waves Recognition

Article Computer Science, Artificial Intelligence

Deep learning for single-lead ECG beat arrhythmia-type detection using novel iris spectrogram representation

Ali Mohammad Alqudah et al.

Summary: This paper introduces a new deep learning methodology for detecting one of up to 17 classes of cardiac arrhythmia based on beat-wise ECG signal analysis, showing high recognition accuracy and efficiency.

SOFT COMPUTING (2022)

Add to Collection

Article Chemistry, Multidisciplinary

LiverNet: Diagnosis of Liver Tumors in Human CT Images

Khaled Alawneh et al.

Summary: This study presents a computer-aided diagnosis system for classifying hepatic tumors as benign or malignant using computed tomography scans. By utilizing deep learning classification and a Support Vector Machine classifier, this system achieves high accuracy and provides a lightweight, fast, reliable, and accurate diagnostic approach.

APPLIED SCIENCES-BASEL (2022)

Add to Collection

Article Medicine, General & Internal

Intelligent Diagnosis and Classification of Keratitis

Hiam Alquran et al.

Summary: This study proposes a deep learning approach for diagnosing corneal ulcers, which has the potential to improve treatment outcomes. Experimental results demonstrate the successful discrimination of ulcers using both manual and automatic feature extraction techniques, with higher accuracy achieved for severity grading using automatic feature extraction. This research has important clinical implications and practical applications.

DIAGNOSTICS (2022)

Add to Collection

Article Multidisciplinary Sciences

Super-resolved time-frequency measurements of coupled phonon dynamics in a 2D quantum material

Christian Gentry et al.

Summary: In this study, a newly-developed superlet transform method was used to analyze and extract phonon dynamics in a laser-excited two-dimensional quantum material. The results showed that the superlet transform method is superior in both time and frequency resolution compared to traditional methods, and revealed changes in the coupled dynamics of phonon modes with new physics. This study opens up a new avenue for capturing the coherent evolution and couplings of strongly-coupled materials and quantum systems.

SCIENTIFIC REPORTS (2022)

Add to Collection

Article Biotechnology & Applied Microbiology

Cervical Net: A Novel Cervical Cancer Classification Using Feature Fusion

Hiam Alquran et al.

Summary: This study proposes a computer-aided diagnosis method for cervical cancer using deep learning and feature fusion. The proposed method includes image processing, feature extraction and selection, as well as classification using machine learning algorithms. Experimental results show that the proposed method achieves high accuracy in multiple categories.

BIOENGINEERING-BASEL (2022)

Add to Collection

Article Computer Science, Information Systems

A Self-Contained STFT CNN for ECG Classification and Arrhythmia Detection at the Edge

Mohammed M. Farag

Summary: This study proposes a lightweight Convolutional Neural Network model for real-time ECG classification and arrhythmia detection at the edge. By interpreting the convolutional layer as a FIR filter, the input ECG signal is transformed into a spectrogram for classification. After optimization, the proposed classifier achieves up to 99.1% classification accuracy and 95% F1-score at the edge.

IEEE ACCESS (2022)

Add to Collection

Article Physics, Multidisciplinary

Automatic ECG Classification Using Continuous Wavelet Transform and Convolutional Neural Network

Tao Wang et al.

Summary: The study introduces an automatic ECG classification method utilizing CWT and CNN, achieving solid performance in diagnosing arrhythmia. With its simplicity and high accuracy, the method has the potential to be a clinical auxiliary diagnostic tool.

ENTROPY (2021)

Add to Collection

Article Chemistry, Analytical

An Automated High-Accuracy Detection Scheme for Myocardial Ischemia Based on Multi-Lead Long-Interval ECG and Choi-Williams Time-Frequency Analysis Incorporating a Multi-Class SVM Classifier

Ahmed Faeq Hussein et al.

Summary: This study proposes a new scheme for myocardial ischemia detection using multi-lead long-interval ECG, with observation of changes in ischemic-related ECG components and extraction of ST and PR features for classification. The scheme achieved high accuracy, sensitivity, and specificity using data from normal and patient populations, reflecting its flexibility and reliability across different sources. Additionally, it can assist cardiologists in detecting signal abnormality with precision and be utilized in emergency cases for rapid evaluation.

SENSORS (2021)

Add to Collection

Article Engineering, Biomedical

Deep convolutional neural networks based ECG beats classification to diagnose cardiovascular conditions

Md Rashed-Al-Mahfuz et al.

Summary: In this study, a novel ECG beat classifier based on a customized VGG16-based Convolution Neural Network (CNN) was proposed to accurately identify cardiovascular conditions by utilizing time-frequency representation of ECG wave-forms. The model achieved high classification accuracy for different classes and could potentially be applied to automate cardiovascular diagnosis system for clinical use.

BIOMEDICAL ENGINEERING LETTERS (2021)

Add to Collection

Article Chemistry, Analytical