Muhammad Adam

Automated detection of coronary artery disease using different durations of ECG segments with convolutional neural network

Coronary artery disease (CAD) is caused due by the blockage of inner walls of coronary arteries by plaque. This constriction reduces the blood flow to the heart muscles resulting in myocardial infarction (MI). The electrocardiogram (ECG) is commonly used to screen the cardiac health. The ECG signals are nonstationary and nonlinear in nature whereby the transient disease indicators may appear randomly on the time scale. Therefore, the procedure to diagnose the abnormal beat is arduous, time consuming and prone to human errors. The automated diagnosis system overcomes these problems. In this study, convolutional neural network (CNN) structures comprising of four convolutional layers, four max pooling layers and three fully connected layers are proposed for the diagnosis of CAD using two and five seconds durations of ECG signal segments. Deep CNN is able to differentiate between normal and abnormal ECG with an accuracy of 94.95%, sensitivity of 93.72%, and specificity of 95.18% for Net 1 (two seconds) and accuracy of 95.11%, sensitivity of 91.13% and specificity of 95.88% for Net 2 (5 s). The proposed system can help the clinicians in their accurate and reliable decision making of CAD using ECG signals.

A deep convolutional neural network model to classify heartbeats

The electrocardiogram (ECG) is a standard test used to monitor the activity of the heart. Many cardiac abnormalities will be manifested in the ECG including arrhythmia which is a general term that refers to an abnormal heart rhythm. The basis of arrhythmia diagnosis is the identification of normal versus abnormal individual heart beats, and their correct classification into different diagnoses, based on ECG morphology. Heartbeats can be sub-divided into five categories namely non-ectopic, supraventricular ectopic, ventricular ectopic, fusion, and unknown beats. It is challenging and time-consuming to distinguish these heartbeats on ECG as these signals are typically corrupted by noise. We developed a 9-layer deep convolutional neural network (CNN) to automatically identify 5 different categories of heartbeats in ECG signals. Our experiment was conducted in original and noise attenuated sets of ECG signals derived from a publicly available database. This set was artificially augmented to even out the number of instances the 5 classes of heartbeats and filtered to remove high-frequency noise. The CNN was trained using the augmented data and achieved an accuracy of 94.03% and 93.47% in the diagnostic classification of heartbeats in original and noise free ECGs, respectively. When the CNN was trained with highly imbalanced data (original dataset), the accuracy of the CNN reduced to 89.07%% and 89.3% in noisy and noise-free ECGs. When properly trained, the proposed CNN model can serve as a tool for screening of ECG to quickly identify different types and frequency of arrhythmic heartbeats.

Automated detection and localization of myocardial infarction using electrocardiogram

Identification and timely interpretation of changes occurring in the 12 electrocardiogram (ECG) leads is crucial to identify the types of myocardial infarction (MI). However, manual annotation of this complex nonlinear ECG signal is not only cumbersome and time consuming but also inaccurate. Hence, there is a need of computer aided techniques to be applied for the ECG signal analysis process. Going further, there is a need for incorporating this computerized software into the ECG equipment, so as to enable automated detection of MIs in clinics. Therefore, this paper proposes a novel method of automated detection and localization of MI by using ECG signal analysis. In our study, a total of 200 twelve lead ECG subjects (52 normal and 148 with MI) involving 611,405 beats (125,652 normal beats and 485,753 beats of MI ECG) are segmented from the 12 lead ECG signals. Firstly, ECG signal obtained from 12 ECG leads are subjected to discrete wavelet transform (DWT) up to four levels of decomposition. Then, 12 nonlinear features namely, approximate entropy ( E a x ), signal energy (?x), fuzzy entropy ( E f x ), Kolmogorov-Sinai entropy ( E k s x ), permutation entropy ( E p x ), Renyi entropy ( E r x ), Shannon entropy ( E s h x ), Tsallis entropy ( E t s x ), wavelet entropy ( E w x ), fractal dimension ( F D x ), Kolmogorov complexity ( C k x ), and largest Lyapunov exponent ( E L L E x ) are extracted from these DWT coefficients. The extracted features are then ranked based on the t value. Then these features are fed into the k-nearest neighbor (KNN) classifier one by one to get the highest classification performance by using minimum number of features. Our proposed method has achieved the highest average accuracy of 98.80%, sensitivity of 99.45% and specificity of 96.27% in classifying normal and MI ECG (two classes), by using 47 features obtained from lead 11 (V5). We have also obtained the highest average accuracy of 98.74%, sensitivity of 99.55% and specificity of 99.16% in differentiating the 10 types of MI and normal ECG beats (11 class), by using 25 features obtained from lead 9 (V3). In addition, our study results achieved an accuracy of 99.97% in locating inferior posterior infarction by using only lead 9 (V3) ECG signal. Our proposed method can be used as an automated diagnostic tool for (i) the detection of different (10 types of) MI by using 12 lead ECG signal, and also (ii) to locate the MI by analyzing only one lead without the need to analyze other leads. Thus, our proposed algorithm and computerized system software (incorporated into the ECG equipment) can aid the physicians and clinicians in accurate and faster location of MIs, and thereby providing adequate time available for the requisite treatment decision.

Application of stacked convolutional and long short-term memory network for accurate identification of CAD ECG signals

Coronary artery disease (CAD) is the most common cause of heart disease globally. This is because there is no symptom exhibited in its initial phase until the disease progresses to an advanced stage. The electrocardiogram (ECG) is a widely accessible diagnostic tool to diagnose CAD that captures abnormal activity of the heart. However, it lacks diagnostic sensitivity. One reason is that, it is very challenging to visually interpret the ECG signal due to its very low amplitude. Hence, identification of abnormal ECG morphology by clinicians may be prone to error. Thus, it is essential to develop a software which can provide an automated and objective interpretation of the ECG signal. This paper proposes the implementation of long short-term memory (LSTM) network with convolutional neural network (CNN) to automatically diagnose CAD ECG signals accurately. Our proposed deep learning model is able to detect CAD ECG signals with a diagnostic accuracy of 99.85% with blindfold strategy. The developed prototype model is ready to be tested with an appropriate huge database before the clinical usage.

Automated diagnosis of congestive heart failure using dual tree complex wavelet transform and statistical features extracted from 2s of ECG signals

Identification of alarming features in the electrocardiogram (ECG) signal is extremely significant for the prediction of congestive heart failure (CHF). ECG signal analysis carried out using computer-aided techniques can speed up the diagnosis process and aid in the proper management of CHF patients. Therefore, in this work, dual tree complex wavelets transform (DTCWT)-based methodology is proposed for an automated identification of ECG signals exhibiting CHF from normal. In the experiment, we have performed a DTCWT on ECG segments of 2s duration up to six levels to obtain the coefficients. From these DTCWT coefficients, statistical features are extracted and ranked using Bhattacharyya, entropy, minimum redundancy maximum relevance (mRMR), receiver-operating characteristics (ROC), Wilcoxon, t-test and reliefF methods. Ranked features are subjected to k-nearest neighbor (KNN) and decision tree (DT) classifiers for automated differentiation of CHF and normal ECG signals. We have achieved 99.86% accuracy, 99.78% sensitivity and 99.94% specificity in the identification of CHF affected ECG signals using 45 features. The proposed method is able to detect CHF patients accurately using only 2s of ECG signal length and hence providing sufficient time for the clinicians to further investigate on the severity of CHF and treatments.

Computer aided diagnosis of diabetic foot using infrared thermography

Diabetes mellitus (DM) is a chronic metabolic disorder that requires regular medical care to prevent severe complications. The elevated blood glucose level affects the eyes, blood vessels, nerves, heart, and kidneys after the onset. The affected blood vessels (usually due to atherosclerosis) may lead to insufficient blood circulation particularly in the lower extremities and nerve damage (neuropathy), which can result in serious foot complications. Hence, an early detection and treatment can prevent foot complications such as ulcerations and amputations. Clinicians often assess the diabetic foot for sensory deficits with clinical tools, and the resulting foot severity is often manually evaluated. The infrared thermography is a fast, nonintrusive and non-contact method which allows the visualization of foot plantar temperature distribution. Several studies have proposed infrared thermography-based computer aided diagnosis (CAD) methods for diabetic foot. Among them, the asymmetric temperature analysis method is more superior, as it is easy to implement, and yielded satisfactory results in most of the studies. In this paper, the diabetic foot, its pathophysiology, conventional assessments methods, infrared thermography and the different infrared thermography-based CAD analysis methods are reviewed.

Characterization of Cardiovascular Diseases Using Wavelet Packet Decomposition and Nonlinear Measures of Electrocardiogram Signal

Cardiovascular diseases (CVDs) remain as the primary causes of disability and mortality worldwide and are predicted to continue rise in the future due to inadequate preventive actions. Electrocardiogram (ECG) signal contains vital clinical information that assists significantly in the diagnosis of CVDs. Assessment of subtle ECG parameters that indicate the presence of CVDs are extremely difficult and requires long hours of manual examination for accurate diagnosis. Hence, automated computer-aided diagnosis systems might help in overcoming these limitations. In this study, a novel algorithm is proposed based on the combination of wavelet packet decomposition (WPD) and nonlinear features. The proposed method achieved classification results of 97.98% accuracy, 99.61% sensitivity and 94.84% specificity with 8 reliefF ranked features. The proposed methodology is highly efficient in helping clinical staff to detect cardiac abnormalities using a single algorithm.

Automated characterization of arrhythmias using nonlinear features from tachycardia ECG beats

Arrhythmias are abnormal heartbeat rhythms, categorized as either harmless or life-threatening. Commonly, elderly people are more vulnerable to life-threatening arrhythmias, namely Atrial Fibrillation (A-Fib), Atrial Flutter (AFL) and Ventricular Fibrillation (V-Fib). Electrocardiogram (ECG) is the primary diagnostic tool that can be used to detect and diagnose cardiac abnormalities including serious arrhythmias. Therefore, using ECG signal beats, we have proposed a computer-aided diagnosis (CAD) system for automated diagnosis of serious arrhythmias. The ECG beats are analyzed using thirteen nonlinear features namely, Shannon entropy, Fuzzy entropy, Tsallis entropy, approximate entropy, Permutation entropy, Modified Multi Scale entropy, Wavelet entropy, Sample entropy, Renyi entropy, Signal Energy, Fractal Dimension, Kolmogorov Sinai entropy and Largest Lyapunov Exponent. Subsequently, the extracted features are ranked using ANOVA and subjected to automated classification using the K-Nearest Neighbor (KNN) and Decision Tree (DT) classifiers. In addition, the extracted features are trained and tested with ten-fold cross validation analysis. DT classifier yielded 96.3% accuracy, 99.3% sensitivity and 84.1% specificity with 14 features and the KNN classifier yielded 93.3% accuracy, 97.5% sensitivity and 75.3% specificity using 12 features. Positively, the proposed CAD system is able to assist the clinical staff in the arrhythmias diagnosis by making the process faster and simpler. Consequently, the necessary treatments can be given expeditiously.

Automated Characterization of Cardiovascular Diseases Using Relative Wavelet Nonlinear Features Extracted from ECG Signals

Cardiovascular diseases (CVDs) are the leading cause of deaths worldwide. The rising mortality rate can be reduced by early detection and treatment interventions. Clinically, electrocardiogram (ECG) signal provides useful information about the cardiac abnormalities and hence employed as a diagnostic modality for the detection of various CVDs. However, subtle changes in these time series indicate a particular disease. Therefore, it may be monotonous, time-consuming and stressful to inspect these ECG beats manually. In order to overcome this limitation of manual ECG signal analysis, this paper uses a novel discrete wavelet transform (DWT) method combined with nonlinear features for automated characterization of CVDs. ECG signals of normal, and dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM) and myocardial infarction (MI) are subjected to five levels of DWT. Relative wavelet of four nonlinear features such as fuzzy entropy, sample entropy, fractal dimension and signal energy are extracted from the DWT coefficients. These features are fed to sequential forward selection (SFS) technique and then ranked using ReliefF method. Our proposed methodology achieved maximum classification accuracy (acc) of 99.27%, sensitivity (sen) of 99.74%, and specificity (spec) of 98.08% with K-nearest neighbor (kNN) classifier using 15 features ranked by the ReliefF method. Our proposed methodology can be used by clinical staff to make faster and accurate diagnosis of CVDs. Thus, the chances of survival can be significantly increased by early detection and treatment of CVDs.

AUTOMATED DIAGNOSIS OF DIABETES USING ENTROPIES AND DIABETIC INDEX

Diabetes Mellitus (DM) is a chronic metabolic disorder that hampers the body’s energy absorption capacity from the food. It is either caused by pancreatic malfunctioning or the body cells being inactive to insulin production. Prolonged diabetes results in severe complications, such as retinopathy, neuropathy, cardiomyopathy and cardiovascular diseases. DM is an incurable disorder. Thus, diagnosis and monitoring of diabetes is essential to prevent the body organs from severe damage. Heart Rate Variability (HRV) signal processing can be used as one of the methods for the diagnosis of DM. Our paper introduces a noninvasive technique of automated diabetic diagnosis using HRV signals. The R-R interval signals are decomposed using Shearlet transforms integrated with Continuous Wavelet Transform (CWT), and their characteristic features are extracted by using Shannon’s, Renyi’s, Kapur entropies, energy and Higher Order Spectra (HOS). Then, Locality Sensitive Discriminant Analysis (LSDA) is employed to remove insignificant features and reduce the number of employed features. These redundant features are eliminated by using six feature selection algorithms: Student’s t-test, Receiver Operating Characteristic Curve (ROC), Wilcoxon signed-rank test, Bhattacharyya distance, Information entropy and Fuzzy Max-Relevance and Min-Redundancy (MRMR). This step is followed by classification of normal and diabetic signals using different classifiers, such as discriminant classifiers, Decision Tree (DT), Support Vector Machine (SVM), Probabilistic Neural Network (PNN), Naive Bayes (NB), Fuzzy Sugeno (FSC), Gaussian Mixture Model (GMM), AdaBoost and k-Nearest Neighbor (k-NN) classifier. In these classifiers, the selected features are employed to distinguish diabetic signals from normal signals. These classifiers are trained and then tested to validate their accuracy to make accurate diagnosis. The FSC classifier is shown to have the highest (100%) accuracy. Nevertheless, we go one step further in formulating another novel classifier in the form of the diabetic index, and have shown how distinctly it is able to separate diabetic signals from normal signals.