1,720,972 research outputs found
Exercise‐induced Repolarization Alternans Heterogeneity in Patients with an Implanted Cardiac Defibrillator
No full textRepolarization alternans (RA), generally recognized as a promising noninvasive index for risk stratification, is often measured under exercise conditions, since RA increases its amplitude with heart rate. Instead, the effect of exercise on the RA location along the JT interval is still unknown. Aim of the present study was to evaluate exercise-induced RA heterogeneity in terms of both amplitude and location. To this aim, we analyzed the ECG precordial leads of 36 patients with an implanted cardiac defibrillator (ICD) who underwent a bicycle ergometer test during which the working load was increased from zero (NoWL) to the patient's maximum capacity (MaxWL). RA was analyzed using our heart-rate adaptive match filter method, which provides an RA parameterization in terms of its amplitude (RAA, mu V) and location, the latter measured as time-delay (RAD, ms) with respect to the T-wave apex, so that positive values of RAD indicate RA occurring in the T-wave right side and vice versa. According to our results, during MaxWL, RAA was higher than during NoWL (34 +/- 21 mu V vs. 16 +/- 10 mu V; P< 10(-5)), whereas RAD was shorter (26 +/- 29 ms vs. 69 +/- 45 ms; P< 10(-4)). Thus, in ICD patients, exercise not only induced a significant increment of RA amplitude, but also caused RA location to move from very late in the repolarization segment toward the T-wave apex
Comparison of standard versus orthogonal ECG leads for T-wave alternans identification
No full textT-wave alternans (TWA), an electrophysiologic phenomenon associated with ventricular arrhythmias, is usually detected from selected ECG leads. TWA amplitude measured in the 12-standard and the 3-orthogonal (vectorcardiographic) leads were compared here to identify which lead system yields a more adequate detection of TWA as a noninvasive marker for cardiac ulnerability to ventricular arrhythmias. Our adaptive match filter (AMF) was applied to exercise ECG tracings from 58 patients with an implanted cardiac defibrillator, 29 of which had ventricular tachycardia or fibrillation during follow-up (cases), while the remaining 29 were used as controls. Two kinds of TWA indexes were considered, the single-lead indexes, defined as the mean TWA amplitude over each lead (MTWAA), and lead-system indexes, defined as the mean and the maximum MTWAA values over the standard leads and over the orthogonal leads. Significantly (P < 0.05) higher TWA in the cases versus controls
was identified only occasionally by the single-lead indexes (odds ratio: 1.0–9.9, sensitivity: 24–76%, specificity: 76–86%), and consistently by the lead-system indexes (odds ratio: 4.5–8.3, sensitivity: 57–72%, specificity: 76%). The latter indexes also showed a significant correlation (0.65–0.83) between standard and orthogonal leads. Hence, when using the AMF, TWA should be detected in all leads of a system to compute the lead-system indexes, which provide a more reliable TWA identification than
single-lead indexes, and a better discrimination of patients at increased risk of cardiac instability. The standard and the orthogonal leads can be considered equivalent for TWA identification, so that TWA analysis can be limited to one-lead system
Enhanced adaptive matched filter for automated identification and measurement of electrocardiographic alternans
Get PDFElectrocardiographic alternans, consisting of P-wave alternans (PWA), QRS-complex alternans (QRSA) and T-wave alternans (TWA), is an index of cardiac risk. However, only automated TWA measurement methods have been proposed so far. Here, we presented the enhanced adaptive matched filter (EAMF) method and tested its reliability in both simulated and experimental conditions. Our methodological novelty consists in the introduction of a signal enhancement procedure according to which all sections of the electrocardiogram (ECG) but the wave of interest are set to baseline, and in the extraction of the alternans area (AAr) in addition to the standard alternans amplitude (AAm). Simulated data consisted of 27 simulated ECGs representing all combinations of PWA, QRSA and TWA of low (10 μV) and high (100 μV) amplitude. Experimental data consisted of exercise 12-lead ECGs from 266 heart failure patients with an implanted cardioverter defibrillator for primary prevention. EAMF was able to accurately identify and measure all kinds of simulated alternans (absolute maximum error equal to 2%). Moreover, different alternans kinds were simultaneously present in the experimental data and EAMF was able to identify and measure all of them (AAr: 545 μV × ms, 762 μV × ms and 1382 μV × ms; AAm: 5 μV, 9 μV and 7 μV; for PWA, QRSA and TWA, respectively) and to discriminate TWA as the prevalent one (with the highest AAr). EAMF accurately identifies and measures all kinds of electrocardiographic alternans. EAMF may support determination of incremental clinical utility of PWA and QRSA with respect to TWA only
Repeated Structuring & Learning Procedure for Detection of Myocardial Ischemia: a Robustness Analysis
No full textMyocardial ischemia, consisting in a reduction of blood flow to the heart, may cause sudden cardiac death by myocardial infarction or trigger serious abnormal rhythms. Thus, its timely identification is crucial. The Repeated Structuring and Learning Procedure (RS&LP), an innovative constructive algorithm able to dynamically create neural networks (NN) alternating structuring and learning phases, was previously found potentially useful for myocardial ischemia detection. However, performance of created NN depends on three parameters, the values of which need to be set a priori by the user: maximal number of layers (NL), maximal number of initializations (NI) and maximal number of confirmations (NC). A robustness analysis of RS&LP to varying values of NL, NI and NC is fundamental for clinical applications concerning myocardial ischemia detection but was never performed before; thus, it was the aim the present study. Thirteen serial ECG features were extracted by pairs of ECGs belonging to 84 cases (patients with induced myocardial ischemia) and 398 controls (patients with no myocardial ischemia) and used as inputs to learn (50% of population) and test (50% of population) NNs with varying values of NL (1,2,3,4,10), NI (50,250,500,1000,1500) and NC (2,5,10,20,50). Performance of obtained NNs was compared in terms of area under the curve (AUC) of the receiver operating characteristics. Overall, 13 NNs were considered; 12 (92%) were characterized by AUC≥80% and 4 (31%) by AUC≥85%. Thus, RS&LP proved to be robust when creating NNs for detecting of myocardial ischemia
Artificial neural network for atrial fibrillation identification in portable devices
Get PDFAtrial fibrillation (AF) is a common cardiac disorder that can cause severe complications. AF diagnosis is typically based on the electrocardiogram (ECG) evaluation in hospitals or in clinical facilities. The aim of the present work is to propose a new artificial neural network for reliable AF identification in ECGs acquired through portable devices. A supervised fully connected artificial neural network (RSL_ANN), receiving 19 ECG features (11 morphological, 4 on F waves and 4 on heart-rate variability (HRV)) in input and discriminating between AF and non-AF classes in output, was created using the repeated structuring and learning (RSL) procedure. RSL_ANN was created and tested on 8028 (training: 4493; validation: 1125; testing: 2410) annotated ECGs belonging to the “AF Classification from a Short Single Lead ECG Recording” database and acquired with the portable KARDIA device by AliveCor. RSL_ANN performance was evaluated in terms of area under the curve (AUC) and confidence intervals (CIs) of the received operating characteristic. RSL_ANN performance was very good and very similar in training, validation and testing datasets. AUC was 91.1% (CI: 89.1%–93.0%), 90.2% (CI: 86.2%–94.3%) and 90.8% (CI: 88.1%–93.5%) for the training, validation and testing datasets, respectively. Thus, RSL_ANN is a promising tool for reliable identification of AF in ECGs acquired by portable devices
Advanced repeated structuring and learning procedure to detect acute myocardial ischemia in serial 12-lead ECGs
Get PDFObjectives. Acute myocardial ischemia in the setting of acute coronary syndrome (ACS) may lead to myocardial infarction. Therefore, timely decisions, already in the pre-hospital phase, are crucial to preserving cardiac function as much as possible. Serial electrocardiography, a comparison of the acute electrocardiogram with a previously recorded (reference) ECG of the same patient, aids in identifying ischemia-induced electrocardiographic changes by correcting for interindividual ECG variability. Recently, the combination of deep learning and serial electrocardiography provided promising results in detecting emerging cardiac diseases; thus, the aim of our current study is the application of our novel Advanced Repeated Structuring and Learning Procedure (AdvRS&LP), specifically designed for acute myocardial ischemia detection in the pre-hospital phase by using serial ECG features.Approach. Data belong to the SUBTRACT study, which includes 1425 ECG pairs, 194 (14%) ACS patients, and 1035 (73%) controls. Each ECG pair was characterized by 28 serial features that, with sex and age, constituted the inputs of the AdvRS&LP, an automatic constructive procedure for creating supervised neural networks (NN). We created 100 NNs to compensate for statistical fluctuations due to random data divisions of a limited dataset. We compared the performance of the obtained NNs to a logistic regression (LR) procedure and the Glasgow program (Uni-G) in terms of area-under-the-curve (AUC) of the receiver-operating-characteristic curve, sensitivity (SE), and specificity (SP).Main Results. NNs (median AUC = 83%, median SE = 77%, and median SP = 89%) presented a statistically (Pvalue lower than 0.05) higher testing performance than those presented by LR (median AUC = 80%, median SE = 67%, and median SP = 81%) and by the Uni-G algorithm (median SE = 72% and median SP = 82%).Significance. In conclusion, the positive results underscore the value of serial ECG comparison in ischemia detection, and NNs created by AdvRS&LP seem to be reliable tools in terms of generalization and clinical applicability
Heart-rate dependency of T-wave alternans during the exercise and recovery
No full textMicrovolt T-wave alternans (TWA), a subtle every-other-beat fluctuation of the electrocardiographic (ECG) T-wave amplitude, is a risk index for the occurrence of malignant ventricular arrhythmias. It is well-known that TWA increases with HR so that TWA is often analyzed during exercise. Since TWA during recovery is usually not evaluated, it is not clear if there is a HR-TWA one-to-one correspondence, or if TWA depends on how such HR was reached. Thus, the aim of the present study was to investigate TWA during both HR-increasing exercise and HR-decreasing recovery. To this aim, our HR adaptive match filter (AMF) for automatic TWA identification was applied to ECG recordings of 266 patients with implanted cardio-defibrillator (ICD). HR varied from 70 bpm to 125 bpm during exercise and vice versa during recovery. Our results indicate that TWA increased (from 17 μV to 51 μV) with increasing HR during exercise, and decreased (from 28 μV to 17 μV) with decreasing HR during recovery. However, such TWA changes did not occur symmetrically, but TWA was significantly lower during exercise than during recovery at low HRs (17-18 μV vs. 20-25 μV for HR<100 bpm; P100 bpm; P<10-6). Thus, there is not a HR-TWA one-to-one correspondence but the value of TWA at a specific HR depends on how such HR was reached
Feature Contributions to ECG-based Heart-Failure Detection: Deep Learning vs. Statistical Analysis
No full textAssessing feature contributions to a specific diagnosis is commonly done by statistical analysis. In the context of heart failure (HF) diagnosis from the electrocardiogram (ECG), this work compares feature contributions assessed by deep learning with those obtained by statistical analysis. Data consists of ECG pairs (baseline and follow-up) from patients with a history of myocardial infarction. When the follow-up ECG was made, controls patients had remained stable, while cases patients had developed HF. The 42 features that characterized each ECG served as inputs of a deep-learning neural network (NN) created by our Repeated Structuring & Learning Procedure. Subject-specific feature ranking was obtained from the local-interpretable model-agnostic explanatory algorithm and processed to obtain feature relevances (FR). Additionally, 42 areas under the curve (AUC) by univariate statistical analysis were obtained. FR and AUC were compared by Pearson's correlation coefficient (p). After training, the NN had a 99% classification performance. FR ranged from 0.32 to 4.47; AUC ranged from 23% to 82%. Correlation analysis yielded no significant association between AUC and FR (ρ=0.18, P-value =0.25). Deep-learning and statistical-analysis feature contributions to HF detection were discordant. Further studies will investigate which of the two approaches better reflects clinical interpretation
Automatic diagnosis of newly emerged heart failure from serial electrocardiography by repeated structuring & learning procedure
Get PDFHeart failure (HF) diagnosis, typically visually performed by serial electrocardiography, may be supported by machine-learning approaches. Repeated structuring & learning procedure (RS&LP) is a constructive algorithm able to automatically create artificial neural networks (ANN); it relies on three parameters, namely maximal number of hidden layers (MNL), initializations (MNI) and confirmations (MNC), arbitrarily set by the user. The aim of this study is to evaluate RS&LP robustness to varying values of parameters and to identify an optimized combination of parameter values for HF diagnosis. To this aim, the Leiden University Medical Center HF database was used. The database is constituted by 129 serial ECG pairs acquired in patients who experienced myocardial infarction; 48 patients developed HF at follow-up (cases), while 81 remained clinically stable (controls). Overall, 15 ANNs were created by considering 13 serial ECG features as inputs (extracted from each serial ECG pair), 2 classes as outputs (cases/controls), and varying values of MNL (1, 2, 3, 4 and 10), MNI (50, 250, 500, 1000 and 1500) and MNC (2, 5, 10, 20 and 50). The area under the curve (AUC) of the receiver operating characteristic did not significantly vary with varying parameter values (P ≥ 0.09). The optimized combination of parameter values, identified as the one showing the highest AUC, was obtained for MNL = 3, MNI = 500 and MNC = 50 (AUC = 86 %; ANN structure: 3 hidden layers of 14, 14 and 13 neurons, respectively). Thus, RS&LP is robust, and the optimized ANN represents a potentially useful clinical tool for a reliable automatic HF diagnosis
- …
