scholarly journals Electrocardiographic Fragmented Activity (II): A Machine Learning Approach to Detection

2019 ◽  
Vol 9 (17) ◽  
pp. 3565 ◽  
Author(s):  
Francisco-Manuel Melgarejo-Meseguer ◽  
Francisco-Javier Gimeno-Blanes ◽  
María-Eladia Salar-Alcaraz ◽  
Juan-Ramón Gimeno-Blanes ◽  
Juan Martínez-Sánchez ◽  
...  
Keyword(s):  
Negative Predictive Value ◽  
Predictive Value ◽  
Companion Paper ◽  
Gaussian Kernel ◽  
Support Vector ◽  
Common Disease ◽  
Ecg Signals ◽  
Fragmented Qrs ◽  
Machine Learning Approach ◽  
Electrocardiogram Ecg

Hypertrophic cardiomyopathy, according to its prevalence, is a comparatively common disease related to the risk of suffering sudden cardiac death, heart failure and stroke. This illness is characterized by the excessive deposition of collagen among healthy myocardium cells. This situation, which is medically known as fibrosis, constitutes effective conduction obstacles in the myocardium electrical path, and when severe enough, it can be outlined as additional peaks or notches in the QRS, clinically entitled as fragmentation. Nowadays, the fragmentation detection is performed by visual inspection, but the fragmented QRS can be confused with the noise present in the electrocardiogram (ECG). On the other hand, fibrosis detection is performed by magnetic resonance imaging with late gadolinium enhancement, the main drawback of this technique being its cost in terms of time and money. In this work, we propose two automatic algorithms, one for fragmented QRS detection and another for fibrosis detection. For this purpose, we used four different databases, including the subrogated database described in the companion paper and incorporating three additional ones, one compounded by more accurate subrogated ECG signals and two compounded by real and affected subjects as labeled by expert clinicians. The first real-world database contains QRS fragmented records and the second one contains records with fibrosis and both were recorded in Hospital Clínico Universitario Virgen de la Arrixaca (Spain). To deeply analyze the scope of these datasets, we benchmarked several classifiers such as Neural Networks, Support Vector Machines (SVM), Decision Trees and Gaussian Naïve Bayes (NB). For the fragmentation dataset, the best results were 0.94 sensitivity, 0.88 specificity, 0.89 positive predictive value, 0.93 negative predictive value and 0.91 accuracy when using SVM with Gaussian kernel. For the fibrosis databases, more limited accuracy was reached, with 0.47 sensitivity, 0.91 specificity, 0.82 predictive positive value, 0.66 negative predictive value and 0.70 accuracy when using Gaussian NB. Nevertheless, this is the first time that fibrosis detection is attempted automatically from ECG postprocessing, paving the way towards improved algorithms and methods for it. Therefore, we can conclude that the proposed techniques could offer a valuable tool to clinicians for both fragmentation and fibrosis diagnoses support.

scholarly journals Anticipating Atrial Fibrillation Signal Using Efficient Algorithm

2021 ◽  
Vol 17 (02) ◽  
pp. 106
Author(s):  
Mohand Lokman Ahmad Al-dabag ◽  
Haider Th. Salim ALRikabi ◽  
Raid Rafi Omar Al-Nima
Keyword(s):  
Atrial Fibrillation ◽  
Computational Cost ◽  
Extraction Methods ◽  
Support Vector ◽  
Ecg Signal ◽  
Heart Problem ◽  
Ecg Signals ◽  
The Right ◽  
Electrocardiogram Ecg ◽  
Low Computational Cost

One of the common types of arrhythmia is Atrial Fibrillation (AF), it may cause death to patients. Correct diagnosing of heart problem through examining the Electrocardiogram (ECG) signal will lead to prescribe the right treatment for a patient. This study proposes a system that distinguishes between the normal and AF ECG signals. First, this work provides a novel algorithm for segmenting the ECG signal for extracting a single heartbeat. The algorithm utilizes low computational cost techniques to segment the ECG signal. Then, useful pre-processing and feature extraction methods are suggested. Two classifiers, Support Vector Machine (SVM) and Multilayer Perceptron (MLP), are separately used to evaluate the two proposed algorithms. The performance of the last proposed method with the two classifiers (SVM and MLP) show an improvement of about (19% and 17%, respectively) after using the proposed segmentation method so it became 96.2% and 97.5%, respectively.

scholarly journals A machine learning approach to predict ethnicity using personal name and census location in Canada

PLoS ONE ◽  
2020 ◽  
Vol 15 (11) ◽  
pp. e0241239
Author(s):  
Kai On Wong ◽  
Osmar R. Zaïane ◽  
Faith G. Davis ◽  
Yutaka Yasui
Keyword(s):  
Machine Learning ◽  
First Nations ◽  
Predictive Value ◽  
Large Scale ◽  
Performance Metrics ◽  
Characteristic Curve ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Learning Approach ◽  
Machine Learning Approach

Background Canada is an ethnically-diverse country, yet its lack of ethnicity information in many large databases impedes effective population research and interventions. Automated ethnicity classification using machine learning has shown potential to address this data gap but its performance in Canada is largely unknown. This study conducted a large-scale machine learning framework to predict ethnicity using a novel set of name and census location features. Methods Using census 1901, the multiclass and binary class classification machine learning pipelines were developed. The 13 ethnic categories examined were Aboriginal (First Nations, Métis, Inuit, and all-combined)), Chinese, English, French, Irish, Italian, Japanese, Russian, Scottish, and others. Machine learning algorithms included regularized logistic regression, C-support vector, and naïve Bayes classifiers. Name features consisted of the entire name string, substrings, double-metaphones, and various name-entity patterns, while location features consisted of the entire location string and substrings of province, district, and subdistrict. Predictive performance metrics included sensitivity, specificity, positive predictive value, negative predictive value, F1, Area Under the Curve for Receiver Operating Characteristic curve, and accuracy. Results The census had 4,812,958 unique individuals. For multiclass classification, the highest performance achieved was 76% F1 and 91% accuracy. For binary classifications for Chinese, French, Italian, Japanese, Russian, and others, the F1 ranged 68–95% (median 87%). The lower performance for English, Irish, and Scottish (F1 ranged 63–67%) was likely due to their shared cultural and linguistic heritage. Adding census location features to the name-based models strongly improved the prediction in Aboriginal classification (F1 increased from 50% to 84%). Conclusions The automated machine learning approach using only name and census location features can predict the ethnicity of Canadians with varying performance by specific ethnic categories.

Deep Learning for Detecting Sleep Apnea from ECG Signals

2020 ◽  
Vol 10 (6) ◽  
pp. 1265-1273
Author(s):  
Lili Chen ◽  
Huoyao Xu
Keyword(s):  
Neural Network ◽  
Sleep Apnea ◽  
Deep Neural Network ◽  
Experimental Results ◽  
Support Vector ◽  
Ecg Signals ◽  
Novel Approach ◽  
Learning Machine ◽  
High Level ◽  
Electrocardiogram Ecg

Sleep apnea (SA) is a common sleep disorders affecting the sleep quality. Therefore the automatic SA detection has far-reaching implications for patients and physicians. In this paper, a novel approach is developed based on deep neural network (DNN) for automatic diagnosis SA. To this end, five features are extracted from electrocardiogram (ECG) signals through wavelet decomposition and sample entropy. The deep neural network is constructed by two-layer stacked sparse autoencoder (SSAE) network and one softmax layer. The softmax layer is added at the top of the SSAE network for diagnosing SA. Afterwards, the SSAE network can get more effective high-level features from raw features. The experimental results reveal that the performance of deep neural network can accomplish an accuracy of 96.66%, a sensitivity of 96.25%, and a specificity of 97%. In addition, the performance of deep neural network outperforms the comparison models including support vector machine (SVM), random forest (RF), and extreme learning machine (ELM). Finally, the experimental results reveal that the proposed method can be valid applied to automatic SA event detection.

scholarly journals An algorithm to safely manage oral food challenge in an office-based setting for children with multiple food allergies

2021 ◽  
Vol 5 (1) ◽  
pp. 030-037
Author(s):  
Cottel Nathalie ◽  
Dieme Aïcha ◽  
Orcel Véronique ◽  
Chantran Yannick ◽  
Bourgoin-Heck Mélisande ◽  
...  
Keyword(s):  
Negative Predictive Value ◽  
Predictive Value ◽  
Validation Cohort ◽  
Hospital Setting ◽  
Food Challenge ◽  
Oral Food Challenge ◽  
Low Risk ◽  
Food Allergies ◽  
Machine Learning Approach ◽  
High Positive Predictive Value

Background: In France, from 30% to 35% of children suffer from multiple food allergies (MFA). The gold standard to diagnosis a food allergy is the oral food challenge (OFC) which is conducted in a hospital setting due to risk of anaphylaxis. The aim of this study was to evaluate an algorithm to predict OFCs at low risk of anaphylaxis that could safely be performed in an office-based setting. Methods: Children with MFA and at least one open OFC reactive or non-reactive to other allergens were included. The algorithm was based on multiple clinical and biological parameters related to food allergens, and designed mainly to predict “low-risk” OFCs i.e., practicable in an office-based setting. The algorithm was secondarily tested in a validation cohort. Results: Ninety-one children (median age 9 years) were included; 94% had at least one allergic comorbidity with an average of three OFCs per child. Of the 261 OFCs analyzed, most (192/261, 74%) were non-reactive. The algorithm failed to correctly predict 32 OFCs with a potentially detrimental consequence but among these only three children had severe symptoms. One hundred eighty-four of the 212 “low-risk” OFCs, (88%) were correctly predicted with a high positive predictive value (87%) and low negative predictive value (44%). These results were confirmed with a validation cohort giving a specificity of 98% and negative predictive value of 100%. Conclusion: This study suggests that the algorithm we present here can predict “low-risk” OFCs in children with MFA which could be safely conducted in an office-based setting. Our results must be confirmed with an algorithm-based machine-learning approach.

scholarly journals Classification of Heart Rate Time Series Using Machine Learning Algorithms

2021 ◽  
Vol 2 (1) ◽  
Keyword(s):  
Detection System ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Ecg Signal ◽  
Ecg Signals ◽  
Heart Patients ◽  
Extraction Step ◽  
Ecg Arrhythmia ◽  
Electrocardiogram Ecg ◽  
The Given

An important diagnostic method for diagnosing abnormalities in the human heart is the electrocardiogram (ECG). A large number of heart patients increase the assignment of physicians. To reduce their assignment, an automatic computer detection system is needed. In this study, a computer system for classifying ECG signals is presented. The MIT-BIH, ECG arrhythmia database is used for analysis. After the ECG signal is noisy in the preprocessing stage, the data feature is extracted. In the feature extraction step, the decision tree is used and the support vector machine (SVM) is constructed to classify the ECG signal into two categories. It is normal or abnormal. The results show that the system classifies the given ECG signal with 90% sensitivity.

scholarly journals On the classification of arrhythmia using supplementary features from Tetrolet transforms

2019 ◽  
Vol 9 (6) ◽  
pp. 5006
Author(s):  
G. Jayagopi ◽  
S. Pushpa
Keyword(s):  
Heart Diseases ◽  
Classification Performance ◽  
Kernel Functions ◽  
Support Vector ◽  
Ecg Signals ◽  
The People ◽  
Vector Machines ◽  
Rbf Kernel ◽  
Electrocardiogram Ecg

<span>Heart diseases had been molded as potential threats to human lives, especially to elderly people in recent days due to the dynamically varying food habits among the people. However, these diseases could be easily caught by proper analysis of Electrocardiogram (ECG) signals acquired from individuals. This paper proposes a better method to detect and classify the arrhythmia using 15 features which include 4 R-R interval features, 3 statistical and 6 chaotic features estimated from ECG signals. Additionally, Entropy and Energy features had been gained after converting one dimensional ECG signals to two dimensional data and applied Tetrolet transforms on that.  Total numbers of 15 features had been utilized to classify the heart beats from the benchmark MIT-Arrhythmia database using Support Vector Machines (SVM). The classification performance was analyzed under various kernel functions and different Tetrolet decomposition levels. It is found that Radial Basis Function (RBF) kernel could perform better than linear and polynomial kernels. This research attempt yielded an accuracy of 99.35 % against the existing works. Moreover, addition of two more features had introduced a negligible overhead of time. Hence, this method is better suitable to detect and classify the Arrhythmia in both online and offline.</span>

scholarly journals Predicting Hospital Readmission in Heart Failure Patients in Iran: A Comparison of Various Machine Learning Methods

2021 ◽  
Vol 27 (4) ◽  
pp. 307-314
Author(s):  
Roya Najafi-Vosough ◽  
Javad Faradmal ◽  
Seyed Kianoosh Hosseini ◽  
Abbas Moghimbeigi ◽  
Hossein Mahjub
Keyword(s):  
Machine Learning ◽  
Heart Failure ◽  
Predictive Value ◽  
Hospital Readmission ◽  
Class Imbalance ◽  
Least Square ◽  
Support Vector ◽  
Common Disease ◽  
Hospital Readmission Rate ◽  
Sensitivity Specificity

Objectives: Heart failure (HF) is a common disease with a high hospital readmission rate. This study considered class imbalance and missing data, which are two common issues in medical data. The current study’s main goal was to compare the performance of six machine learning (ML) methods for predicting hospital readmission in HF patients.Methods: In this retrospective cohort study, information of 1,856 HF patients was analyzed. These patients were hospitalized in Farshchian Heart Center in Hamadan Province in Western Iran, from October 2015 to July 2019. The support vector machine (SVM), least-square SVM (LS-SVM), bagging, random forest (RF), AdaBoost, and naïve Bayes (NB) methods were used to predict hospital readmission. These methods’ performance was evaluated using sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Two imputation methods were also used to deal with missing data.Results: Of the 1,856 HF patients, 29.9% had at least one hospital readmission. Among the ML methods, LS-SVM performed the worst, with accuracy in the range of 0.57–0.60, while RF performed the best, with the highest accuracy (range, 0.90–0.91). Other ML methods showed relatively good performance, with accuracy exceeding 0.84 in the test datasets. Furthermore, the performance of the SVM and LS-SVM methods in terms of accuracy was higher with the multiple imputation method than with the median imputation method.Conclusions: This study showed that RF performed better, in terms of accuracy, than other methods for predicting hospital readmission in HF patients.

scholarly journals Detection of Atrial Fibrillation Based on Optimized Electrocardiogram (Ecg) Recordings

2019 ◽  
Vol 9 (2S3) ◽  
pp. 198-203
Keyword(s):  
Atrial Fibrillation ◽  
Genetic Algorithm ◽  
Support Vector ◽  
Svm Classifier ◽  
Ecg Signal ◽  
Ecg Signals ◽  
Public Dataset ◽  
Electrocardiogram Ecg ◽  
Heart Arrhythmias ◽  
Typical Process

Atrial fibrillation (AF) is one among the foremost common heart arrhythmias. It is terribly tough to discover unless a precise arrhythmia episode happens throughout the exploration. If the diagnosis and the treatment is delayed the Atrial fibrillation can lead to heart strokes and causes death, therefore automatic detection of AF is an urgent need. The analysis of ECG recordings is considered as one of the typical process of detecting AF. The ECG signals analysed by considering normal rhythm (N), other arrhythmias (O) and Atrial Fibrillation(A) and noises. In this paper the proposed technique is validated by considering open accessible public dataset. In the proposed method initially pre-processing of ECG signal is performed, next extraction of features, optimizing the features using genetic algorithm (GA) and finally classifying using support vector machine (SVM) classifier. The proposed algorithm achieves overall accuracy of 95.8% and by considering top 10 features the rate of accuracy is 96.8% which is better compared to the existing algorithm with an SNR of dB. The experimental results are performed using MATLAB and uggest that by availing the short ECG recording also the detection of AF is obtained accurately.

scholarly journals From ECG signals to images: a transformation based approach for deep learning

2021 ◽  
Vol 7 ◽  
pp. e386
Author(s):  
Mahwish Naz ◽  
Jamal Hussain Shah ◽  
Muhammad Attique Khan ◽  
Muhammad Sharif ◽  
Mudassar Raza ◽  
...  
Keyword(s):  
Deep Learning ◽  
Ventricular Arrhythmias ◽  
Ventricular Tachyarrhythmia ◽  
Heart Rhythm ◽  
Support Vector ◽  
Learning Approach ◽  
Ecg Signals ◽  
Different Types ◽  
Electrical Impulses ◽  
Electrocardiogram Ecg

Provocative heart disease is related to ventricular arrhythmias (VA). Ventricular tachyarrhythmia is an irregular and fast heart rhythm that emerges from inappropriate electrical impulses in the ventricles of the heart. Different types of arrhythmias are associated with different patterns, which can be identified. An electrocardiogram (ECG) is the major analytical tool used to interpret and record ECG signals. ECG signals are nonlinear and difficult to interpret and analyze. We propose a new deep learning approach for the detection of VA. Initially, the ECG signals are transformed into images that have not been done before. Later, these images are normalized and utilized to train the AlexNet, VGG-16 and Inception-v3 deep learning models. Transfer learning is performed to train a model and extract the deep features from different output layers. After that, the features are fused by a concatenation approach, and the best features are selected using a heuristic entropy calculation approach. Finally, supervised learning classifiers are utilized for final feature classification. The results are evaluated on the MIT-BIH dataset and achieved an accuracy of 97.6% (using Cubic Support Vector Machine as a final stage classifier).

A NOVEL TWO-LEAD ARRHYTHMIA CLASSIFICATION SYSTEM BASED ON CNN AND LSTM

2019 ◽  
Vol 19 (03) ◽  
pp. 1950004 ◽  
Author(s):  
JINGHUI CHU ◽  
HONG WANG ◽  
WEI LU
Keyword(s):  
Disease Diagnosis ◽  
Medical Instrumentation ◽  
Support Vector ◽  
Svm Classifier ◽  
Binary Particle Swarm Optimization ◽  
Ecg Signals ◽  
Deep Learning Features ◽  
Heart Disease Diagnosis ◽  
Electrocardiogram Ecg ◽  
Multi Classification

Arrhythmia classification is useful during heart disease diagnosis. Although well-established for intra-patient diagnoses, inter-patient arrhythmia classification remains difficult. Most previous work has focused on the intra-patient condition and has not followed the Association for the Advancement of Medical Instrumentation (AAMI) standards. Here, we propose a novel system for arrhythmia classification based on multi-lead electrocardiogram (ECG) signals. The core of the design is that we fuse two types of deep learning features with some common traditional features and select discriminating features using a binary particle swarm optimization algorithm (BPSO). Then, the feature vector is classified using a weighted support vector machine (SVM) classifier. For a better generalization of the model and to draw fair comparisons, we carried out inter-patient experiments and followed the AAMI standards. We found that, when using common metrics aimed at multi-classification either macro- or micro-averaging, our system outperforms most other state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document