Electrocardiography Diagnosis Research Articles

Real-world validation of smartphone-based photoplethysmography for rate and rhythm monitoring in atrial fibrillation.

Photoplethysmography- (PPG) based smartphone applications facilitate heart rate and rhythm monitoring in patients with paroxysmal and persistent atrial fibrillation (AF). Despite an endorsement from the European Heart Rhythm Association, validation studies in this setting are lacking. Therefore, we evaluated the accuracy of PPG-derived heart rate and rhythm classification in subjects with an established diagnosis of AF in unsupervised real-world conditions. Fifty consecutive patients were enrolled, 4 weeks before undergoing AF ablation. Patients used a handheld single-lead electrocardiography (ECG) device and a fingertip PPG smartphone application to record 3907 heart rhythm measurements twice daily during 8 weeks. The ECG was performed immediately before and after each PPG recording and was given a diagnosis by the majority of three blinded cardiologists. A consistent ECG diagnosis was exhibited along with PPG data of sufficient quality in 3407 measurements. A single measurement exhibited good quality more often with ECG (93.2%) compared to PPG (89.5%; P < 0.001). However, PPG signal quality improved to 96.6% with repeated measurements. Photoplethysmography-based detection of AF demonstrated excellent sensitivity [98.3%; confidence interval (CI): 96.7-99.9%], specificity (99.9%; CI: 99.8-100.0%), positive predictive value (99.6%; CI: 99.1-100.0%), and negative predictive value (99.6%; CI: 99.0-100.0%). Photoplethysmography underestimated the heart rate in AF with 6.6 b.p.m. (95% CI: 5.8 b.p.m. to 7.4 b.p.m.). Bland-Altman analysis revealed increased underestimation in high heart rates. The root mean square error was 11.8 b.p.m. Smartphone applications using PPG can be used to monitor patients with AF in unsupervised real-world conditions. The accuracy of AF detection algorithms in this setting is excellent, but PPG-derived heart rate may tend to underestimate higher heart rates.

Current and Future Use of Artificial Intelligence in Electrocardiography.

Artificial intelligence (AI) is increasingly used in electrocardiography (ECG) to assist in diagnosis, stratification, and management. AI algorithms can help clinicians in the following areas: (1) interpretation and detection of arrhythmias, ST-segment changes, QT prolongation, and other ECG abnormalities; (2) risk prediction integrated with or without clinical variables (to predict arrhythmias, sudden cardiac death, stroke, and other cardiovascular events); (3) monitoring ECG signals from cardiac implantable electronic devices and wearable devices in real time and alerting clinicians or patients when significant changes occur according to timing, duration, and situation; (4) signal processing, improving ECG quality and accuracy by removing noise/artifacts/interference, and extracting features not visible to the human eye (heart rate variability, beat-to-beat intervals, wavelet transforms, sample-level resolution, etc.); (5) therapy guidance, assisting in patient selection, optimizing treatments, improving symptom-to-treatment times, and cost effectiveness (earlier activation of code infarction in patients with ST-segment elevation, predicting the response to antiarrhythmic drugs or cardiac implantable devices therapies, reducing the risk of cardiac toxicity, etc.); (6) facilitating the integration of ECG data with other modalities (imaging, genomics, proteomics, biomarkers, etc.). In the future, AI is expected to play an increasingly important role in ECG diagnosis and management, as more data become available and more sophisticated algorithms are developed.

Expert System and Decision Support System for Electrocardiogram Interpretation and Diagnosis: Review, Challenges and Research Directions

Electrocardiography (ECG) is one of the most widely used recordings in clinical medicine. ECG deals with the recording of electrical activity that is generated by the heart through the surface of the body. The electrical activity generated by the heart is measured using electrodes that are attached to the body surface. The use of ECG in the diagnosis and management of cardiovascular disease (CVD) has been in existence for over a decade, and research in this domain has recently attracted large attention. Along this line, an expert system (ES) and decision support system (DSS) have been developed for ECG interpretation and diagnosis. However, despite the availability of a lot of literature, access to recent and more comprehensive review papers on this subject is still a challenge. This paper presents a comprehensive review of the application of ES and DSS for ECG interpretation and diagnosis. Researchers have proposed a number of features and methods for ES and DSS development that can be used to monitor a patient’s health condition through ECG recordings. In this paper, a taxonomy of the features and methods for ECG interpretation and diagnosis were presented. The significance of the features and methods, as well as their limitations, were analyzed. This review further presents interesting theoretical concepts in this domain, as well as identifies challenges and open research issues on ES and DSS development for ECG interpretation and diagnosis that require substantial research effort. In conclusion, this paper identifies important future research areas with the purpose of advancing the development of ES and DSS for ECG interpretation and diagnosis.

Abstract 13558: Opening the Black Box of Artificial Intelligence: Visualization of Detecting Heart Failure Subtypes Using Electrocardiography

Objectives: To evaluate and visualize how artificial intelligence (AI) recognizes electrocardiography (ECG) characteristics of each heart failure (HF) subtype. We developed a visually explainable and interpretable AI-ECG algorithm for various HF subtypes to meet this need. This study proposes the ShapeExplainer (SE) to interpret models. Methods: From 33,920 patients at two hospitals, we trained a convolutional neural network to identify patients with HF subtypes using ECG and clinical diagnosis. HF subtypes include heart failure with reduced ejection fraction (HFrEF), heart failure with mid-range ejection fraction (HFmrEF, and heart failure with preserved ejection fraction (HFpEF) ( Figure a ). To interprete each HF classifier model, SE, which converts non-HF group data to be recognized as each HF subtype by each classifier, was trained. The converted fake ECG shows what the change is for the classifier to recognize as each HF subtype. SE from the non-HF group of the external validation set shows the result by comparing the shape before and after transformation. Results: When tested on an independent set of 6,784 patients, the network model yielded values for the area under the curves of HFrEF of 0.945, HFmrEF of 0.884, and HFpEF of 0.859, respectively. The results of each classifier were visualized by overlapping the original and newly generated ECGs. Comparing each subtype and its fake ECG, there was no clear trend of electrocardiographic features. However, fake ECG showed a tendency of qrs amplitude as it progressed to non-HF, HFpEF, HFmrEF, and HFrEF, especially in T wave or ST-change ( Figure b ). Conclusions: The proposed fake ECG generating method is a new approach for interpreting the AI-ECG model, helping physicians understand how AI recognizes HF subtypes.

Explainable detection of myocardial infarction using deep learning models with Grad-CAM technique on ECG signals

Myocardial infarction (MI) accounts for a high number of deaths globally. In acute MI, accurate electrocardiography (ECG) is important for timely diagnosis and intervention in the emergency setting. Machine learning is increasingly being explored for automated computer-aided ECG diagnosis of cardiovascular diseases. In this study, we have developed DenseNet and CNN models for the classification of healthy subjects and patients with ten classes of MI based on the location of myocardial involvement. ECG signals from the Physikalisch-Technische Bundesanstalt database were pre-processed, and the ECG beats were extracted using an R peak detection algorithm. The beats were then fed to the two models separately. While both models attained high classification accuracies (more than 95%), DenseNet is the preferred model for the classification task due to its low computational complexity and higher classification accuracy than the CNN model due to feature reusability. An enhanced class activation mapping (CAM) technique called Grad-CAM was subsequently applied to the outputs of both models to enable visualization of the specific ECG leads and portions of ECG waves that were most influential for the predictive decisions made by the models for the 11 classes. It was observed that Lead V4 was the most activated lead in both the DenseNet and CNN models. Furthermore, this study has also established the different leads and parts of the signal that get activated for each class. This is the first study to report features that influenced the classification decisions of deep models for multiclass classification of MI and healthy ECGs. Hence this study is crucial and contributes significantly to the medical field as with some level of visible explainability of the inner workings of the models, the developed DenseNet and CNN models may garner needed clinical acceptance and have the potential to be implemented for ECG triage of MI diagnosis in hospitals and remote out-of-hospital settings.

Application of Artificial Neural Network and Empirical Mode Decomposition with Chaos Theory to Electrocardiography Diagnosis

Application of Artificial Neural Network and Empirical Mode Decomposition with Chaos Theory to Electrocardiography Diagnosis

QRS Complex Detection Using Novel Deep Learning Neural Networks

Objective: Accurate QRS complex detection is essential for electrocardiography (ECG) diagnosis. Many proposed algorithms don't perform satisfactorily on noisy and arrhythmia ECGs. The purpose of this study is to develop a noise resistant and generalizable method to detect QRS complexes accurately. Methods: Two deep learning models based on multi-dilated convolutional blocks are proposed. One model (CNN) is mainly composed of convolutional blocks and Squeeze-and-Excitation networks (SENet). The other model (CRNN) contains a hybrid convolutional and recurrent neural network. With 5-fold cross-validation approach the models are trained and tested on four open-access ECG databases: the China Physiological Signal Challenge (2019) database (CPSCDB), the MIT-BIH Noise Stress Test Database (NSTDB), the MIT-BIH Arrhythmia Database (MITDB) and the QT Database (QTDB). Results: The F1 score of CNN model on CPSCDB, NSTDB, MITDB and QTDB are 0.9929, 0.9892, 0.9994 and 0.9998 respectively. The F1 score of CRNN model on these four databases are 0.9947, 0.9953, 0.9995 and 0.9998 respectively. The ensemble of both models scored the first place in the China Physiological Signal Challenge (2019). Conclusion: The proposed models achieve state-of-the-art performance in QRS complex detection and show good generalization on different databases. This work might help make better ECG diagnosis.

Automatic Multi-Label ECG Diagnosis of Impulse or Conduction Abnormalities in Patients with Deep Learning Algorithm: A Cohort Study

Background: Electrocardiography (ECG) is a fundamental diagnostic tool, frequently used in clinical practice, for testing heart conditions. Recently, computer-aided ECG interpretation has become more widely investigated. However, researches on concurrent ECG diagnosis for multiple heart abnormalities are limited and therefore a market-applicable multi-label automatic diagnosis framework that covers a wide range of arrhythmias, with better-than-human accuracy in diagnosing ECGs with multi-labels, has not yet been developed. We therefore aimed to engineer a deep learning approach for the automated multi-label diagnosis of impulse or conduction abnormalities by realtime ECG analysis. Methods: We utilized a dataset with 70,692 patients, consisting of adults with age ≥18 years. 180112 ECGs (standard 10-second and 12-lead format) with 21 distinct rhythm classes, including various impulse or conduction abnormalities, were used for the diagnosis of arrhythmias in multi-label and single-label level. Base rhythm labels were assessed by trained personnel under the supervision of cardiologists. We ensured that each abnormality had a sample size of at least 4,000 ECGs, except idioventricular rhythm (1,780 ECGs due to the rarity of said disease), for compiling a balanced dataset. We allocated ECGs to a training or a validation dataset for the multi-label diagnostic model training, and evaluated model performance on a test dataset by calculating the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, as well as quantifying accuracy, sensitivity, specificity, and the F1 score. The test dataset was annotated by the consensus of a committee of board-certified actively-practicing cardiologists. Validation of results for the test dataset were compared to diagnoses made by 53 ECG cardiologists who have a wide range of ECG interpretation experience (range=0-12+ years). Findings: We demonstrated that in multi-label level, our convolutional neural network (CNN) approach toward diagnosing heart abnormalities resulted in AUC scores above 0.913 for most of the diseases, with an average score of 0.985 (95% confidence interval (CI) 0.988-0.994). The model obtained an average F1 score of 0.890 (95% CI0.877- 0.903), sensitivity of 0.869 (95% CI 0.851-0.887), specificity of 0.995 (95% CI 0.994- 0.996). The average F1 score computed from our deep learning model exceeded the score from human cardiologists regardless of ECG interpretation experience (0 12 years: 0.829). The CNN outputs also showed identical or increased performance with respect to average sensitivity and accuracy as compared to the cardiologists. In single-label level, the CNN achieved an average F1 of 0.918, higher than the average level of the cardiologists 0.798. Interpretation: Our automatic ECG diagnosis system is better than human cardiologists to classify particular waveforms into clinically defined categories accurately, to distinguish a wide range of distinct arrhythmias, and to apply both single label and multiple label ECGs. Our study is therefore a promising foundation for the deployment of computational decision-support systems in practical clinical applications. Funding Statement: Project supported by the Major International (Regional) Joint Research Project, the Ministry of Management Science, the National Natural Science Foundation of China (Grant No. 71520107003). Declaration of Interests: Dr. Yuan reports grants from National Natural Science Foundation of China, during the conduct of the study; In addition, Dr. Yuan has a patent An automatic ECG classification method, system and equipment based on deep learning algorithm pending to Ye Yuan, Xiaoyun Yang, Hongling Zhu, Yiran Wang, Cheng Cheng, Xingyi Li, Hang Yin, Jingyi Wang. There are no other conflicts of interest. Ethics Approval Statement: To protect patients’ privacy, we used anonymized data as to prevent breaches. Accordingly, a written informed consent was not required for this study, since the ECG samples were appropriately anonymized and de-identified, as according to the Health Insurance Portability and Accountability Act Safe Harbor provision. The study design was evaluated and exempted from full review by the Huazhong University of Science and Technology Institutional Review Board.

Smartphone electrographic monitoring for atrial fibrillation in acute ischemic stroke and transient ischemic attack.

Rationale Paroxysmal atrial fibrillation is a common and preventable cause of devastating strokes. However, currently available monitoring methods, including Holter monitoring, cardiac telemetry and event loop recorders, have drawbacks that restrict their application in the general stroke population. AliveCor™ heart monitor, a novel device that embeds miniaturized electrocardiography (ECG) in a smartphone case coupled with an application to record and diagnose the ECG, has recently been shown to provide an accurate and sensitive single lead ECG diagnosis of atrial fibrillation. This device could be used by nurses to record a 30-s ECG instead of manual pulse taking and automatically provide a diagnosis of atrial fibrillation. Aims To compare the proportion of patients with paroxysmal atrial fibrillation detected by AliveCor™ ECG monitoring with current standard practice. Sample size 296 Patients. Design Consecutive ischemic stroke and transient ischemic attack patients presenting to participating stroke units without known atrial fibrillation will undergo intermittent AliveCor™ ECG monitoring administered by nursing staff at the same frequency as the vital observations of pulse and blood pressure until discharge, in addition to the standard testing paradigm of each participating stroke unit to detect paroxysmal atrial fibrillation. Study outcome Proportion of patients with paroxysmal atrial fibrillation detected by AliveCor™ ECG monitoring compared to 12-lead ECG, 24-h Holter monitoring and cardiac telemetry. Discussion Use of AliveCor™ heart monitor as part of routine stroke unit nursing observation has the potential to be an inexpensive non-invasive method to increase paroxysmal atrial fibrillation detection, leading to improvement in stroke secondary prevention.

Validation of a Novel Digital Tool in Automatic Scoring of an Online ECG Examination at an International Cardiology Meeting.

We have previously developed a novel digital tool capable of automatically recognizing correct electrocardiography (ECG) diagnoses in an online exam and demonstrated a significant improvement in diagnostic accuracy when utilizing an inductive-deductive reasoning strategy over a pattern recognition strategy. In this study, we sought to validate these findings from participants at the International Winter Arrhythmia School meeting, one of the foremost electrophysiology events in Canada. Preregistration to the event was sent by e-mail. The exam was administered on day 1 of the conference. Results and analysis were presented the following morning to participants. Twenty-five attendees completed the exam, providing a total of 500 responses to be marked. The online tool accurately identified 195 of a total of 395 (49%) correct responses (49%). In total, 305 responses required secondary manual review, of which 200 were added to the correct responses pool. The overall accuracy of correct ECG diagnosis for all participants was 69% and 84% when using pattern recognition or inductive-deductive strategies, respectively. Utilization of a novel digital tool to evaluate ECG competency can be set up as a workshop at international meetings or educational events. Results can be presented during the sessions to ensure immediate feedback.

Glibenclamide (glyburide) delay electrocardiographic ST-segment elevation in a diabetic patient with acute myocardial infarction

Acute myocardial infarction (AMI) is the most frequent cause of death worldwide. The timely diagnosis of AMI, especially ST-segment elevation myocardial infarction, is a key to successful management. A 12-lead electrocardiography (ECG) should be obtained and interpreted as soon as possible. The 12 lead ECG diagnosis may be more difficult in patients without diagnostic ST-segment elevation but with persistent typical ischemic symptoms. Here, we present a diabetic patient with AMI where the ST-segment elevation was masked by glibenclamide (glyburide) initially and delayed ECG change was revealed after glibenclamide was eliminated.

Electrical impulse production and propagation abnormalities in nonagerian hearts. Insights from the Ikaria study

Purpose: Electrocardiography (ECG) is a wide spread convenient and low cost tool with diagnostic and prognostic significance. As few studies have examined the incidence of ECG abnormalities in subjects over 90 years of age, we have examined the incidence of electrical impulse production and propagation abnormalities in a population of nonagenarians enrolled in the Ikaria study. Methods: We included 110 subjects aged 60 to 74 years (1st group of age), 114 aged 75 to 89 years (2nd group) and 61 over 90 years old (3rd group). Several demographic and clinical characteristic were recorder in all individuals. ECG was recorded from a 12-lead surface digital recorded. ECG diagnosis and characterization of abnormalities were performed by two independent, blinded to the study physicians. Results: The mean age of the participants in the 1st, 2nd and 3rd age groups was 72±8, 77±9 and 92±8 years respectively. Between the three age groups there was no difference in the presence of male sex (p=0.99), in body mass index (, p=0.47) in the presence of cardiovascular disease (p=0.78), arterial hypertension (p=0.23) and diabetes mellitus (p=0.37). Subjects in the 3rd group had increased heart rate (73±16bpm vs. 66±11bpm vs. 62±10 bpm, p<0.001), lower prevalence of sinus rhythm (54% vs. 74% vs. 94%, p<0.001) and increased prevalence of a paced rhythm (6.6% vs. 3.6% vs. 0.0%, p=0.04), atrial fibrillation (23.0% vs. 15.2% vs. 5.5%, p=0.04), atrial flutter (3.3% vs. 0.0% vs. 0.0%, p=0.026), multifocal atrial rhythms (MARs) with or without tachycardia (14.8% vs. 4.4% vs. 0.9%, p=0.01) as compared to subjects in 1st and 2nd and groups of age. Between 1st, 2nd and 3rd groups there was no difference in the prevalence of right bundle branch block (5.5% vs. 6.2% vs. 9.8%, p=0.53), left bundle branch block (0.0% vs. 1.8% vs. 1.6%, p=0.38), left anterior fascicular block (10% vs. 16% vs. 23%, p=0.07), bifascicular block (3.6% vs. 5.4% vs. 8.2%, p=0.44) and trifascicular block (0.0% vs. 1.7% vs. 1.6%, p=0.38). Interestingly, a J shape pattern was observed in the prevalence of first degree atrioventricular block (15% vs. 30% vs. 18%, p=0.01) and second degree atrioventricular block (0.0% vs. 3.6% vs. 1.6%, p=0.13) for 1st, 2nd and 3rd groups of age respectively. Conclusion: The incidence of common arrhythmias (such as atrial fibrillation) and heart rate increase with aging whereas sinus rhythm is suppressed. Interestingly, in this sample of oldest old subjects the incidence of MARs, which are rarely encountered in the general population, was approximately 15%. The prognostic impact of the latter finding remains to be clarified.

Study on Real-time Monitoring Technique for Cardiac Arrhythmia Based on Smartphone

Electrocardiography (ECG) monitoring is an important method for cardiac disease detection and prevention. With the development of telemedicine, ECG real-time monitoring becomes more convenient. The rise and popularization of smartphone provide a new technical means for ECG monitoring. This paper focuses on the development of software for an ECG monitoring system on a smartphone platform. The system includes a client and a center. The client is developed on smartphone. Its main function is process and diagnose ECG signal. A wavelet method is used to detect the QRS-complex and diagnostic criteria are formulated to diagnose 14 kinds of arrhythmia disease. The tele-monitoring center is developed on a PC with LabVIEW, which receives ECG signal and diagnosis from smartphone by GPRS. Results show that the system can monitor ECG and diagnose arrhythmia by calculating the heart rate in real time. The proposed system is reliable and gives good real-time performance.

Systematic review and meta-analysis of randomized controlled trials comparing Chinese patent medicines Compound Danshen Dripping Pills and Di’ao Xinxuekang in treating angina pectoris

Chinese patent medicines Compound Danshen Dripping Pills (DSP) and Di'ao Xinxuekang (DXK) capsules were both found effective in treating angina pectoris. However, there is no systematic review comparing their efficacy. This systematic review aims to compare the efficacy of DSP and DXK in treating angina pectoris based on randomized controlled trials (RCTs) comparing their efficacy. RCT reports published between 1994 and 2011 were retrieved from databases including China Doctoral Dissertations Full-text Database, Chinese Journal Full-text Database, China Master's Theses Full-text Database, Wanfang Data, Cochrane Library, Excerpts Medica Database, ScienceDirect, MEDLINE (EBSCOhost) and PubMed. The last retrieval was performed on April 7, 2011. RCT reports comparing the effects of DSP and DXK were included, regardless publishing language. Included RCT reports were assessed for their study quality by using the Jadad scale and the Cochrane risk of bias tool. Data including overall effect and electrocardiography (ECG) improvements were extracted from the included RCTs for meta-analysis. The effect sizes based on overall and ECG diagnosis were measured by odds ratio (OR) and 95% confidence interval (CI). Subgroup analysis and sensitivity analysis were also performed. Nine RCT reports with 926 participants were included. Eight were scored 2 and the other one was scored 4 by using the Jadad scale. The OR between DSP and DXK based on overall diagnosis was 2.06 (95% CI: 1.03-4.12; P(overall)=0.04). Six out of the nine included RCTs reported ECG data. The OR between DSP and DXK based on the ECG diagnosis was 1.92 (95% CI: 1.23-3.00; P(ECG)=0.004). The OR results were stable under subgroup analysis and sensitivity analysis. DSP was consistently more effective than DXK according to meta-analysis, which was verified by subgroup analysis and sensitivity analysis. However, more RCTs of higher quality are needed for further confirmation.

A clustering based system for instant detection of cardiac abnormalities from compressed ECG

Compressed Electrocardiography (ECG) is being used in modern telecardiology applications for faster and efficient transmission. However, existing ECG diagnosis algorithms require the compressed ECG packets to be decompressed before diagnosis can be applied. This additional process of decompression before performing diagnosis for every ECG packet introduces undesirable delays, which can have severe impact on the longevity of the patient. In this paper, we first used an attribute selection method that selects only a few features from the compressed ECG. Then we used Expected Maximization (EM) clustering technique to create normal and abnormal ECG clusters. Twenty different segments (13 normal and 7 abnormal) of compressed ECG from a MIT-BIH subject were tested with 100% success using our model. Apart from automatic clustering of normal and abnormal compressed ECG segments, this paper presents an algorithm to identify initiation of abnormality. Therefore, emergency personnel can be contacted for rescue mission, within the earliest possible time. This innovative technique based on data mining of compressed ECGs attributes, enables faster identification of cardiac abnormalities resulting in an efficient telecardiology diagnosis system.

The Realization of Ubiquitous 12-Lead ECG Diagnosis in Emergency Telemedicine

This study presents an e-technology to realize a mobile 12-lead electrocardiography (ECG) information system capable of providing clinicians with ubiquitous 12-lead ECG diagnoses. Until now, there has been limited information on mobile systems that can exchange information with clinically used 12-lead ECG for emergency telemedicine. The development of mobilized 12-lead ECG diagnoses in clinical practice is impeded by the heterogeneous 12-lead ECG file formats and compressed ECG waveform data. In this study, clinically used Standard Communications Protocol ECG and Extensible Markup Language ECG interpreters were developed to extract compressed 12-lead ECG waveform data. A 12-lead ECG database server connecting with a hospital information system was then built to collect 12-lead ECG files and related clinical data. Additionally, a mobile 12-lead ECG database was implemented on mobile phones for remote 12-lead ECG diagnoses. Clinical evaluations confirmed the usefulness of this mobile 12-lead ECG information system, which significantly improves the diagnostic accuracy and the performance of medical treatments in cardio-emergency telemedicine. In summary, this system facilitated ubiquitous 12-lead ECG diagnoses and enhanced the efficiency and service quality of emergency telemedicine.

Computer‐Aided Systematized Approach to Pediatric ECG Analysis

In pediatric electrocardiography (ECG), the correct classification of segments in the medical record is subjected to various components. A conventional analysis based either on the cardiologist's skills or a quick reference to a standard normality table may lead to mistakes and to an incorrect final medical diagnosis. In this study, the evaluation of 12 specific ECG segments (ES) is defined as segmental analysis (SA). We hypothesized that a computer-aided SA can provide better results for a correct classification of pediatric ECGs, compared to the conventional analysis. This study aims to evaluate the accuracy of a computer-aided SA ECG diagnosis of pediatric patients by cardiologists. To validate the software, we selected four normal and one altered ECG as references, all with concordant reports given by two cardiologists using manual planimetry. Nineteen cardiology experts independently examined each of the five standard pediatric ECGs twice. First, the ECG was analyzed in the conventional form. Subsequently, the participants evaluated ECGs with our proposed computer-aided AS, by measuring the 12 specific ES in the grid paper and having their values entered into a custom-made software, in order to compare them with normality standards. The computer-aided SA reduced ECG misreading or ECG misinterpretation errors by 83%. It also showed a more uniform, high-quality analysis of all ES (minimum of 92% correctly reviewed segments from normal pediatric ECGs) by all the participant cardiologists. We consider the computer-aided SA for ECG evaluation in pediatric cardiology an efficient and safe complementary method.

Classification of ST-elevation acute myocardial infarction, acute pericarditis, and benign early repolarization

Reperfusion therapy improves survival of patients with ST elevation early acute myocardial infarction (STEAMI). In prehospital electrocardiography (ECG), a key component of clinical decision support for patients with symptoms suggesting acute coronary syndromes would be an automated quick and accurate diagnosis of STEAMI. Since such a computerized algorithm must consider ST segment elevation criteria, its development is challenged by cardiac conditions other than STEAMI. Confounders, such as acute pericarditis and benign early repolarization, degrade the performance of such algorithms. This study analyzed ST elevation characteristics in three diagnostic categories: STEAMI, acute pericarditis, and benign early repolarization. A statistical tree classification method was used to obtain a satisfactory classification among STEAMI and the other two groups. Twelve-lead resting EGGs were selected from the Providence Medical Center ECG database. Serial ECG characteristics were used to define the patient groups: STEAMI (321 pts)-resolution nor Q wave evolution. ECGs at 500 samples/sec were analyzed using Agilent Technologies' 12-lead resting ECG analysis program. Sensitivity and specificity of the computerized ECG analysis algorithm are: n Sensitivity Specificity STEAMI 321 87% 99% Acute Pericarditis 115 75% 95% Benign Early Repolarization 153 76% 98% When benign early repolarization and acute pericarditis are grouped together as a Non-STEAMI group, the sensitivity and specificity are: n Sensitivity Specificity STEAMI 321 87% 99% Non-STEAMI 268 86% 99% Benign early repolarization and acute pericarditis tend to mimic the ECG diagnosis of STEAMI. However, in STEAMI the ST elevation is localized to a portion of the ECG leads and is often accompanied by ST depression in other leads; in acute pericarditis there is diffuse ST elevation with minimal difference between the maximal and minimal amphitudes; in benign early repolarization ST elevation is also diffuse but higher in amplitudes than acute pericarditis group in all leads except VI and III. Tall T amplitude in all leads but VI and III is another significant feature to distinguish benign early repolarization from acute pericarditis and early acute MI. This new algorithm provides both high sensitivity and specificity for STEAMI diagnosis. Visibly depressed PR segment and the shape of ST elevation were not used but could potentially further improve the distinction between acute pericarditis and benign early repolarization.

Improved diagnostic performance on the severity of left ventricular hypertrophy with body surface mapping.

To improve the diagnostic usefulness of electrocardiography (ECG) in determining the severity of left ventricular hypertrophy (LVH) with body surface mapping, 87 unipolar ECGs were recorded from 57 patients with left ventricular (LV) concentric hypertrophy and 30 with LV dilatation. Body surface ECG features due to LVH were evaluated by increase of QRS voltage and delayed local activation. We measured for each lead R voltage, net area of QRS (AQRS), ventricular activation time (VAT), and departure index (DI) of AQRS and VAT (DI = mean/SD). From these measurements, seven parameters were calculated for each patient: Rmax, the maximal R wave voltage; AQRSmax, the maximal AQRS; AQRS-Dmax, the maximal AQRS DI; AQRS-Darea, the area size where DIs of AQRS are more than 2; VATmax, the maximal VAT; VAT-Dmax, the maximal VAT DI; and VAT-Darea, the area size where DIs of VAT are more than 2. Among these parameters, the most effective for diagnosis of LVH were selected by stepwise multiple regression analysis. In the concentric hypertrophy group, the combination of VAT-Darea and Rmax was determined to be the best for estimating wall thickness. The regression equation determined from them correlated well to wall thickness (r = 0.73). In the LV dilatation hypertrophy group, only AQRSmax was selected for estimating LV dilatation. A good correlation between AQRSmax and LV internal dimension was observed (r = 0.73). With the body surface distribution of VAT prolongation, septal hypertrophy was separated from the other LVH. These were superior to the conventional method of 12-lead ECGs. ECG diagnosis of LVH severity improved by incorporating a mapping study. Also, prolongation of VAT and increase in QRS voltage were shown to be important when determining the severity of LVH.