scholarly journals Reliable Deep Learning–Based Detection of Misplaced Chest Electrodes During Electrocardiogram Recording: Algorithm Development and Validation

10.2196/25347 ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. e25347
Author(s):  
Khaled Rjoob ◽  
Raymond Bond ◽  
Dewar Finlay ◽  
Victoria McGilligan ◽  
Stephen J Leslie ◽  
...  
Keyword(s):  
Deep Learning ◽  
Ventricular Hypertrophy ◽  
Left Ventricular ◽  
Intercostal Space ◽  
The Past ◽  
Body Surface Potential Maps ◽  
Body Surface Potential ◽  
Using Data ◽  
Development And Validation ◽  
Electrocardiogram Ecg

Background A 12-lead electrocardiogram (ECG) is the most commonly used method to diagnose patients with cardiovascular diseases. However, there are a number of possible misinterpretations of the ECG that can be caused by several different factors, such as the misplacement of chest electrodes. Objective The aim of this study is to build advanced algorithms to detect precordial (chest) electrode misplacement. Methods In this study, we used traditional machine learning (ML) and deep learning (DL) to autodetect the misplacement of electrodes V1 and V2 using features from the resultant ECG. The algorithms were trained using data extracted from high-resolution body surface potential maps of patients who were diagnosed with myocardial infarction, diagnosed with left ventricular hypertrophy, or a normal ECG. Results DL achieved the highest accuracy in this study for detecting V1 and V2 electrode misplacement, with an accuracy of 93.0% (95% CI 91.46-94.53) for misplacement in the second intercostal space. The performance of DL in the second intercostal space was benchmarked with physicians (n=11 and age 47.3 years, SD 15.5) who were experienced in reading ECGs (mean number of ECGs read in the past year 436.54, SD 397.9). Physicians were poor at recognizing chest electrode misplacement on the ECG and achieved a mean accuracy of 60% (95% CI 56.09-63.90), which was significantly poorer than that of DL (P<.001). Conclusions DL provides the best performance for detecting chest electrode misplacement when compared with the ability of experienced physicians. DL and ML could be used to help flag ECGs that have been incorrectly recorded and flag that the data may be flawed, which could reduce the number of erroneous diagnoses.

scholarly journals Deep learning consistently detects misplaced chest electrodes when recording the electrocardiogram: An error that is commonly undetected by physicians (Preprint)

2020 ◽  
Author(s):  
Khaled Rjoob ◽  
Raymond Bond ◽  
Dewar Finlay ◽  
Victoria McGilligan ◽  
Stephen J Leslie ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Deep Learning ◽  
Left Ventricular ◽  
Intercostal Space ◽  
Influential Factor ◽  
Body Surface Potential Maps ◽  
Erroneous Diagnosis ◽  
Body Surface Potential ◽  
Using Data ◽  
Electrocardiogram Ecg

BACKGROUND A 12-lead electrocardiogram (ECG) is the most common method to diagnose cardiovascular diseases such as acute myocardial infarction. However, there are a number of misinterpretations of the ECG caused by several different factors. One influential factor can take place during ECG acquisition where chest electrodes are misplaced. OBJECTIVE This research is the first experiment to build advanced algorithms to detect precordial (chest) electrode misplacement. METHODS in this article we used traditional machine learning (ML) and deep learning (DL) to auto-detect the misplacement of electrodes V1 and V2 using features from the resultant ECG. The algorithms were trained using data extracted from high resolution body surface potential maps consisting of patients who were diagnosed with myocardial infarction, left ventricular hypertrophy or normal. RESULTS DL achieved the highest accuracy in this study for detecting V1 and V2 electrode misplacement with an accuracy of 93.0% [95%CI=91.46,94.53] for misplacement in the second intercostal space. DL performance in the second intercostal space was benchmarked with physicians (n=11 and age=47.3±15.5) who are experienced in reading ECGs (mean number of ECGs read in the past year = 436.54±397.9). Physicians were poor at recognising chest electrode misplacement on the ECG and achieved a mean accuracy of 60% [95%CI=56.09,63.90] which was significantly poorer when compared to DL (P<.001). CONCLUSIONS DL provides the best performance for detecting chest electrode misplacement when compared to the ability of experienced physicians. Clinical Impact: DL and ML could be used to help flag ECGs that have been incorrectly recorded and that the data maybe be flawed, which could reduce an erroneous diagnosis.

scholarly journals The Opposite Polarity of the PQ Segment Compared to the P Wave Isointegral Maps

2011 ◽  
pp. 777-784
Author(s):  
K. KOZLÍKOVÁ ◽  
J. MARTINKA ◽  
J. MURÍN ◽  
J. BULAS
Keyword(s):  
Myocardial Infarction ◽  
Body Surface ◽  
Ventricular Hypertrophy ◽  
Opposite Polarity ◽  
Left Ventricular ◽  
P Wave ◽  
Healthy Controls ◽  
Body Surface Potential Maps ◽  
Body Surface Potential ◽  
The Right

The aim of our work was to study the opposite polarity of the PQ segment to the P wave body surface potential maps in different groups of patients. We constructed isointegral maps (IIM) in 26 healthy controls (C), 16 hypertensives (HT), 26 patients with arterial hypertension and left ventricular hypertrophy (LVH) and 15 patients with myocardial infarction (MI). We analyzed values and positions of map extrema and compared the polarity of maps using the correlation coefficient. The IIM P maxima appeared mainly over the precordium, the minima mainly in the right subclavicular area. The highest maxima were in the MI group, being significantly higher than in the HT and LVH groups. No differences concerning any values of other extrema were significant. The IIM PQ maxima were distributed over the upper half of the chest; the minima mainly over the middle sternum. A statistically significant opposite polarity between the IIM P and IIM PQ was found in 80 % of cases. The opposite polarity of the P wave and the PQ segment was proved in isointegral body surface maps. The extrema occurred in areas not examined by the standard chest leads. This has to be considered for diagnostic purposes.

Body surface potential map patterns in left ventricular hypertrophy during PRST

1990 ◽  
Vol 23 (3) ◽  
pp. 281
Author(s):  
Fred Kornreich ◽  
Terrence J. Montague ◽  
Pentti M. Rautaharju ◽  
Mikhail Kavadias ◽  
Milan B. Horacek ◽  
...  
Keyword(s):  
Left Ventricular Hypertrophy ◽  
Surface Potential ◽  
Body Surface ◽  
Ventricular Hypertrophy ◽  
Left Ventricular ◽  
Potential Map ◽  
Body Surface Potential

scholarly journals Data Driven Computer Simulation to Analyse an ECG Limb Lead System Used in Connected Health Environments

2016 ◽  
Vol 55 (03) ◽  
pp. 258-265 ◽  
Author(s):  
Dewar Finlay ◽  
Daniel Guldenring ◽  
Cathal Breen ◽  
Raymond Bond
Keyword(s):  
Computer Simulation ◽  
Lead System ◽  
Ecg Signals ◽  
Connected Health ◽  
Body Surface Potential Maps ◽  
Body Surface Potential ◽  
The Difference ◽  
Electrocardiogram Ecg ◽  
Monitoring Devices ◽  
Computerised Analysis

SummaryBackground: Recently under the Connected Health initiative, researchers and small-medium engineering companies have developed Electrocardiogram (ECG) monitoring devices that incorporate non-standard limb electrode positions, which we have named the Central Einthoven (CE) configuration.Objectives: The main objective of this study is to compare ECG signals recorded from the CE configuration with those recorded from the recommended Mason-Likar (ML) configuration.Methods: This study involved extracting two different sets of ECG limb leads from each patient to compare the difference in the signals. This was done using computer simulation that is driven by body surface potential maps. This simulator was developed to facilitate this experiment but it can also be used to test similar hypotheses. This study included, (a) 176 ECGs derived using the ML electrode positions and (b) the 176 corresponding ECGs derived using the CE electrode positions. The signals from these ECGs were compared using root mean square error (RMSE), Pearson product-moment correlation coefficient (r) and similarity coefficient (SC). We also investigated whether the CE configuration influences the calculated mean cardiac axis. The top 10 cases where the ECGs were significantly different between the two configurations were visually compared by an ECG interpreter.Results: We found that the leads aVL, III and aVF are most affected when using the CE configuration. The absolute mean difference between the QRS axes from both configurations was 28° (SD = 37°). In addition, we found that in 82% of the QRS axes calculated from the CE configuration was more rightward in comparison to the QRS axes derived from the ML configuration. Also, we found that there is an 18% chance that a misleading axis will be located in the inferior right quadrant when using the CE approach. Thus, the CE configuration can emulate right axis deviation. The clinician visually identified 6 out of 10 cases where the CE based ECG yielded clinical differences that could result in false positives.Conclusions: The CE configuration will not yield the same diagnostic accuracy for diagnosing pathologies that rely on current amplitude criteria. Conversely, rhythm lead II was not significantly affected, which supports the use of the CE approach for assessing cardiac rhythm only. Any computerised analysis of the CE based ECG will need to take these findings into consideration.

INTELLIGENT DATA ANALYSIS FOR THE CLASSIFICATION OF BODY SURFACE POTENTIAL MAPS

2008 ◽  
Vol 07 (03) ◽  
pp. 249-263 ◽  
Author(s):  
MARK P. DONNELLY ◽  
CHRIS D. NUGENT ◽  
DEWAR D. FINLAY ◽  
NORMAN D. BLACK
Keyword(s):  
Surface Potential ◽  
Body Surface ◽  
Left Ventricular ◽  
Control Group ◽  
Final Decision ◽  
Data Sets ◽  
Classification Models ◽  
Body Surface Potential Maps ◽  
Body Surface Potential

Body surface potential maps were investigated to identify a set of optimal recording sites required to discriminate between several diseases. Specifically, recordings captured from subjects exhibiting myocardial infarction or left ventricular hypertrophy, as well as a control group consisting of healthy subjects, were investigated. Owing to the fact that multi-class problems are inherently difficult to solve we divided the problem into several two-class scenarios. Six data sets were generated from the available 744 records, each viewing the available data differently, to form several two-class problems. A data-driven selection algorithm was applied to each of the generated data sets to produce six classification models, each utilizing as features those recording sites offering most to the discrimination task being investigated. Subsequently, a framework was introduced to facilitate the combination of outputs from each classifier. Essentially, the framework used the outputs from half of the classification models to determine which of the remaining models would be employed to form a final decision. A benchmark, in the form of a multi-group classifier, was introduced to evaluate the perceived benefits of the proposed approach. An improvement of approximately 10% upon the benchmark was observed resulting in an overall accuracy of 79.19%.

scholarly journals Using deep learning method to identify left ventricular hypertrophy on echocardiography

2021 ◽  
Author(s):  
Xiang Yu ◽  
Xinxia Yao ◽  
Bifeng Wu ◽  
Hong Zhou ◽  
Shudong Xia ◽  
...  
Keyword(s):  
Deep Learning ◽  
Left Ventricular Hypertrophy ◽  
Sensitivity And Specificity ◽  
Ventricular Hypertrophy ◽  
Early Stage ◽  
Hypertensive Heart Disease ◽  
Complete Classification ◽  
Left Ventricular ◽  
Classification Model ◽  
Detection Model

Abstract Background Left ventricular hypertrophy (LVH) is an independent prognostic factor for cardiovascular events and it can be detected by echocardiography in the early stage. In this study, we aim to develop a semi-automatic diagnostic network based on deep learning algorithms to detect LVH. Methods We retrospectively collected 1610 transthoracic echocardiograms, included 724 patients [189 hypertensive heart disease (HHD), 218 hypertrophic cardiomyopathy (HCM), and 58 cardiac amyloidosis (CA), along with 259 controls]. The diagnosis of LVH was defined by two experienced clinicians. For the deep learning architecture, we introduced ResNet and U-net++ to complete classification and segmentation tasks respectively. The models were trained and validated independently. Then, we connected the best-performing models to form the final framework and tested its capabilities. Results In terms of individual networks, the view classification model produced AUC = 1.0. The AUC of the LVH detection model was 0.98 (95% CI 0.94–0.99), with corresponding sensitivity and specificity of 94.0% (95% CI 85.3–98.7%) and 91.6% (95% CI 84.6–96.1%) respectively. For etiology identification, the independent model yielded good results with AUC = 0.90 (95% CI 0.82–0.95) for HCM, AUC = 0.94 (95% CI 0.88–0.98) for CA, and AUC = 0.88 (95% CI 0.80–0.93) for HHD. Finally, our final integrated framework automatically classified four conditions (Normal, HCM, CA, and HHD), which achieved an average of AUC 0.91, with an average sensitivity and specificity of 83.7% and 90.0%. Conclusion Deep learning architecture has the ability to detect LVH and even distinguish the latent etiology of LVH.

Arrhythmia Classification Using Deep Learning: A Review

2021 ◽  
Vol 18 ◽  
pp. 96-105
Author(s):  
Deepali Koppad
Keyword(s):  
Deep Learning ◽  
Human Error ◽  
Traditional Approach ◽  
Learning Technology ◽  
Ecg Signals ◽  
The Past ◽  
Medical Disorders ◽  
Medical Reports ◽  
Diagnosis Method ◽  
Electrocardiogram Ecg

In most hospitals, the diagnosis of medical disorders involves the traditional approach of doctors manually analyzing the medical reports of the patient. This method is not only time consuming and strenuous, but is also highly prone to human error. With the advent of deep learning technology, an efficient autonomous diagnosis method holds the possibility of replacing the existing tedious approach. This in turn results in the reduction of human error which is of major concern in the medical industry today. Through this paper, we aim to put forth an articulate review of the different deep learning methodologies, observed in the past four years, to classify arrhythmia using electrocardiogram (ECG) signals.

Sign in / Sign up

Export Citation Format

Share Document