Device agnostic AI-based analysis of ambulatory ECG recordings.

Deep Convolutional Neural Networks (DCNNs) have been shown to provide improved performance over traditional heuristic algorithms for the detection of arrhythmias from ambulatory ECG recordings. However, these DCNNs have primarily been trained and tested on device-specific databases with standardized electrode positions and uniform sampling frequencies. This work explores the possibility of training a DCNN for Atrial Fibrillation (AF) detection on a database of single­lead ECG rhythm strips extracted from resting 12­lead ECGs. We then test the performance of the DCNN on recordings from ambulatory ECG devices with different recording leads and sampling frequencies. We developed an extensive proprietary resting 12­lead ECG dataset of 549,211 patients. This dataset was randomly split into a training set of 494,289 patients and a testing set of the remaining 54,922 patients. We trained a 34-layer convolutional DCNN to detect AF and other arrhythmias on this dataset. The DCNN was then validated on two Physionet databases commonly used to benchmark automated ECG algorithms (1) MIT-BIH Arrhythmia Database and (2) MIT-BIH Atrial Fibrillation Database. Validation was performed following the EC57 guidelines, with performance assessed by gross episode and duration sensitivity and positive predictive value (PPV). Finally, validation was also performed on a selection of rhythm strips from an ambulatory ECG patch that a committee of board-certified cardiologists annotated. On MIT-BIH, The DCNN achieved a sensitivity of 100% and 84% PPV in detecting episodes of AF. and 100% sensitivity and 94% PPV in quantifying AF episode duration. On AFDB, The DCNN achieved a sensitivity of 94% and PPV of 98% in detecting episodes of AF, and 98% sensitivity and 100% PPV in quantifying AF episode duration. On the patch database, the DCNN demonstrated performance that was closely comparable to that of a cardiologist. The results indicate that DCNN models can learn features that generalize between resting 12­lead and ambulatory ECG recordings, allowing DCNNs to be device agnostic for detecting arrhythmias from single­lead ECG recordings and enabling a range of clinical applications.

Overview of featurization techniques used in traditional versus emerging deep learning-based algorithms for automated interpretation of the 12-lead ECG.

Automated interpretation of the 12-lead ECG has remained an underpinning interest in decades of research that has seen a diversity of computing applications in cardiology. The application of computers in cardiology began in the 1960s with early research focusing on the conversion of analogue ECG signals (voltages) to digital samples. Alongside this, software techniques that automated the extraction of wave measurements and provided basic diagnostic statements, began to emerge. In the years since then there have been many significant milestones which include the widespread commercialisation of 12-lead ECG interpretation software, associated clinical utility and the development of the related regulatory frameworks to promote standardised development. In the past few years, the research community has seen a significant rejuvenation in the development of ECG interpretation programs. This is evident in the research literature where a large number of studies have emerged tackling a variety of automated ECG interpretation problems. This is largely due to two factors. Specifically, the technical advances, both software and hardware, that have facilitated the broad adoption of modern artificial intelligence (AI) techniques, and, the increasing availability of large datasets that support modern AI approaches. In this article we provide a very high-level overview of the operation of and approach to the development of early 12-lead ECG interpretation programs and we contrast this to the approaches that are now seen in emerging AI approaches. Our overview is mainly focused on highlighting differences in how input data are handled prior to generation of the diagnostic statement.

Machine learning techniques for detecting electrode misplacement and interchanges when recording ECGs: A systematic review and meta-analysis.

INTRODUCTION: Electrode misplacement and interchange errors are known problems when recording the 12­lead electrocardiogram (ECG). Automatic detection of these errors could play an important role for improving clinical decision making and outcomes in cardiac care. The objectives of this systematic review and meta-analysis is to 1) study the impact of electrode misplacement on ECG signals and ECG interpretation, 2) to determine the most challenging electrode misplacements to detect using machine learning (ML), 3) to analyse the ML performance of algorithms that detect electrode misplacement or interchange according to sensitivity and specificity and 4) to identify the most commonly used ML technique for detecting electrode misplacement/interchange. This review analysed the current literature regarding electrode misplacement/interchange recognition accuracy using machine learning techniques. METHOD: A search of three online databases including IEEE, PubMed and ScienceDirect identified 228 articles, while 3 articles were included from additional sources from co-authors. According to the eligibility criteria, 14 articles were selected. The selected articles were considered for qualitative analysis and meta-analysis. RESULTS: The articles showed the effect of lead interchange on ECG morphology and as a consequence on patient diagnoses. Statistical analysis of the included articles found that machine learning performance is high in detecting electrode misplacement/interchange except left arm/left leg interchange. CONCLUSION: This review emphasises the importance of detecting electrode misplacement detection in ECG diagnosis and the effects on decision making. Machine learning shows promise in detecting lead misplacement/interchange and highlights an opportunity for developing and operationalising deep learning algorithms such as convolutional neural network (CNN) to detect electrode misplacement/interchange.

Novel hybrid method for interpolating missing information in body surface potential maps.

BACKGROUND: Body surface potential mapping (BSPM) provides additional electrophysiological information that can be useful for the detection of cardiac diseases. Moreover, BSPMs are currently utilized in electrocardiographic imaging (ECGI) systems within clinical practice. Missing information due to noisy recordings, poor electrode contact is inevitable. In this study, we present an interpolation method that combines Laplacian minimization and principal component analysis (PCA) techniques for interpolating this missing information. METHOD: The dataset used consisted of 117 lead BSPMs recorded from 744 subjects (a training set of 384 subjects, and a test set of 360). This dataset is a mixture of normal, old myocardial infarction, and left ventricular hypertrophy subjects. The missing data was simulated by ignoring data recorded from 7 regions: the first region represents three rows of five electrodes on the anterior torso surface (high potential gradient region), and the other six regions were realistic patterns that have been drawn from clinical data and represent the most likely regions of broken electrodes. Three interpolation methods including PCA based interpolation, Laplacian interpolation, and hybrid Laplacian-PCA interpolation methods were used to interpolate the missing data from the remaining electrodes. In the simulated region of missing data, the calculated potentials from each interpolation method were compared with the measured potentials using relative error (RE) and correlation coefficient (CC) over time. In the hybrid Laplacian-PCA interpolation method, the missing data are firstly interpolated using Laplacian interpolation, then the resulting BSPM of 117 potentials was multiplied by the (117 × 117) coefficient matrix calculated using the training set to get the principal components. Out of 117 principal components (PCs), the first 15 PCs were utilized for the second stage of interpolation. The best performance of interpolation was the reason for choosing the first 15 PCs. RESULTS: The differences in the median of relative error (RE) between Laplacian and Hybrid method ranged from 0.01 to 0.35 (p < 0.001), while the differences in the median of correlation between them ranged from 0.0006 to 0.034 (p < 0.001). PCA-interpolation method performed badly especially in some scenarios where the number of missing electrodes was up to 12 or higher causing a high region of missing data. The figures of median of RE for PCA-method were between 0.05 and 0.6 lower than that for Hybrid method (p < 0.001). However, the median of correlation was between 0.0002 and 0.26 lower than the figure for the Hybrid method (p < 0.001). CONCLUSION: Comparison between the three methods of interpolation (Laplacian, PCA, Hybrid) in reconstructing missing data in BSPM showed that the Hybrid method was always better than the other methods in all scenarios; whether the number of missed electrodes is high or low, and irrespective of the location of these missed electrodes.

Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12­lead electrocardiogram.

BACKGROUND: Electrocardiogram (ECG) lead misplacement can adversely affect ECG diagnosis and subsequent clinical decisions. V1 and V2 are commonly placed superior of their correct position. The aim of the current study was to use machine learning approaches to detect V1 and V2 lead misplacement to enhance ECG data quality. METHOD: ECGs for 453 patients, (normal n = 151, Left Ventricular Hypertrophy (LVH) n = 151, Myocardial Infarction n = 151) were extracted from body surface potential maps. These were used to extract both the correct and incorrectly placed V1 and V2 leads. The prevalence for correct and incorrect leads were 50%. Sixteen features were extracted in three different domains: time-based, statistical and time-frequency features using a wavelet transform. A hybrid feature selection approach was applied to select an optimal set of features. To ensure optimal model selection, five classifiers were used and compared. The aforementioned feature selection approach and classifiers were applied for V1 and V2 misplacement in three different positions: first, second and third intercostal spaces (ICS). RESULTS: The accuracy for V1 misplacement detection was 93.9%, 89.3%, 72.8% in the first, second and third ICS respectively. In V2, the accuracy was 93.6%, 86.6% and 68.1% in the first, second and third ICS respectively. There is a noticeable decline in accuracy when detecting misplacement in the third ICS which is expected.

Automation bias in medicine: The influence of automated diagnoses on interpreter accuracy and uncertainty when reading electrocardiograms.

INTRODUCTION: Interpretation of the 12­lead Electrocardiogram (ECG) is normally assisted with an automated diagnosis (AD), which can facilitate an 'automation bias' where interpreters can be anchored. In this paper, we studied, 1) the effect of an incorrect AD on interpretation accuracy and interpreter confidence (a proxy for uncertainty), and 2) whether confidence and other interpreter features can predict interpretation accuracy using machine learning. METHODS: This study analysed 9000 ECG interpretations from cardiology and non-cardiology fellows (CFs and non-CFs). One third of the ECGs involved no ADs, one third with ADs (half as incorrect) and one third had multiple ADs. Interpretations were scored and interpreter confidence was recorded for each interpretation and subsequently standardised using sigma scaling. Spearman coefficients were used for correlation analysis and C5.0 decision trees were used for predicting interpretation accuracy using basic interpreter features such as confidence, age, experience and designation. RESULTS: Interpretation accuracies achieved by CFs and non-CFs dropped by 43.20% and 58.95% respectively when an incorrect AD was presented (p < 0.001). Overall correlation between scaled confidence and interpretation accuracy was higher amongst CFs. However, correlation between confidence and interpretation accuracy decreased for both groups when an incorrect AD was presented. We found that an incorrect AD disturbs the reliability of interpreter confidence in predicting accuracy. An incorrect AD has a greater effect on the confidence of non-CFs (although this is not statistically significant it is close to the threshold, p = 0.065). The best C5.0 decision tree achieved an accuracy rate of 64.67% (p < 0.001), however this is only 6.56% greater than the no-information-rate. CONCLUSION: Incorrect ADs reduce the interpreter's diagnostic accuracy indicating an automation bias. Non-CFs tend to agree more with the ADs in comparison to CFs, hence less expert physicians are more effected by automation bias. Incorrect ADs reduce the interpreter's confidence and also reduces the predictive power of confidence for predicting accuracy (even more so for non-CFs). Whilst a statistically significant model was developed, it is difficult to predict interpretation accuracy using machine learning on basic features such as interpreter confidence, age, reader experience and designation.

A decision support system and rule-based algorithm to augment the human interpretation of the 12-lead electrocardiogram.

BACKGROUND: The 12-lead Electrocardiogram (ECG) has been used to detect cardiac abnormalities in the same format for more than 70years. However, due to the complex nature of 12-lead ECG interpretation, there is a significant cognitive workload required from the interpreter. This complexity in ECG interpretation often leads to errors in diagnosis and subsequent treatment. We have previously reported on the development of an ECG interpretation support system designed to augment the human interpretation process. This computerised decision support system has been named 'Interactive Progressive based Interpretation' (IPI). In this study, a decision support algorithm was built into the IPI system to suggest potential diagnoses based on the interpreter's annotations of the 12-lead ECG. We hypothesise semi-automatic interpretation using a digital assistant can be an optimal man-machine model for ECG interpretation. OBJECTIVES: To improve interpretation accuracy and reduce missed co-abnormalities. METHODS: The Differential Diagnoses Algorithm (DDA) was developed using web technologies where diagnostic ECG criteria are defined in an open storage format, Javascript Object Notation (JSON), which is queried using a rule-based reasoning algorithm to suggest diagnoses. To test our hypothesis, a counterbalanced trial was designed where subjects interpreted ECGs using the conventional approach and using the IPI+DDA approach. RESULTS: A total of 375 interpretations were collected. The IPI+DDA approach was shown to improve diagnostic accuracy by 8.7% (although not statistically significant, p-value=0.1852), the IPI+DDA suggested the correct interpretation more often than the human interpreter in 7/10 cases (varying statistical significance). Human interpretation accuracy increased to 70% when seven suggestions were generated. CONCLUSION: Although results were not found to be statistically significant, we found; 1) our decision support tool increased the number of correct interpretations, 2) the DDA algorithm suggested the correct interpretation more often than humans, and 3) as many as 7 computerised diagnostic suggestions augmented human decision making in ECG interpretation. Statistical significance may be achieved by expanding sample size.

PDF-ECG in clinical practice: A model for long-term preservation of digital 12-lead ECG data.

BACKGROUND: In clinical practice, data archiving of resting 12-lead electrocardiograms (ECGs) is mainly achieved by storing a PDF report in the hospital electronic health record (EHR). When available, digital ECG source data (raw samples) are only retained within the ECG management system. OBJECTIVE: The widespread availability of the ECG source data would undoubtedly permit successive analysis and facilitate longitudinal studies, with both scientific and diagnostic benefits. METHODS & RESULTS: PDF-ECG is a hybrid archival format which allows to store in the same file both the standard graphical report of an ECG together with its source ECG data (waveforms). Using PDF-ECG as a model to address the challenge of ECG data portability, long-term archiving and documentation, a real-world proof-of-concept test was conducted in a northern Italy hospital. A set of volunteers undertook a basic ECG using routine hospital equipment and the source data captured. Using dedicated web services, PDF-ECG documents were then generated and seamlessly uploaded in the hospital EHR, replacing the standard PDF reports automatically generated at the time of acquisition. Finally, the PDF-ECG files could be successfully retrieved and re-analyzed. CONCLUSION: Adding PDF-ECG to an existing EHR had a minimal impact on the hospital's workflow, while preserving the ECG digital data.

A computer-human interaction model to improve the diagnostic accuracy and clinical decision-making during 12-lead electrocardiogram interpretation.

INTRODUCTION: The 12-lead Electrocardiogram (ECG) presents a plethora of information and demands extensive knowledge and a high cognitive workload to interpret. Whilst the ECG is an important clinical tool, it is frequently incorrectly interpreted. Even expert clinicians are known to impulsively provide a diagnosis based on their first impression and often miss co-abnormalities. Given it is widely reported that there is a lack of competency in ECG interpretation, it is imperative to optimise the interpretation process. Predominantly the ECG interpretation process remains a paper based approach and whilst computer algorithms are used to assist interpreters by providing printed computerised diagnoses, there are a lack of interactive human-computer interfaces to guide and assist the interpreter. METHODS: An interactive computing system was developed to guide the decision making process of a clinician when interpreting the ECG. The system decomposes the interpretation process into a series of interactive sub-tasks and encourages the clinician to systematically interpret the ECG. We have named this model 'Interactive Progressive based Interpretation' (IPI) as the user cannot 'progress' unless they complete each sub-task. Using this model, the ECG is segmented into five parts and presented over five user interfaces (1: Rhythm interpretation, 2: Interpretation of the P-wave morphology, 3: Limb lead interpretation, 4: QRS morphology interpretation with chest lead and rhythm strip presentation and 5: Final review of 12-lead ECG). The IPI model was implemented using emerging web technologies (i.e. HTML5, CSS3, AJAX, PHP and MySQL). It was hypothesised that this system would reduce the number of interpretation errors and increase diagnostic accuracy in ECG interpreters. To test this, we compared the diagnostic accuracy of clinicians when they used the standard approach (control cohort) with clinicians who interpreted the same ECGs using the IPI approach (IPI cohort). RESULTS: For the control cohort, the (mean; standard deviation; confidence interval) of the ECG interpretation accuracy was (45.45%; SD=18.1%; CI=42.07, 48.83). The mean ECG interpretation accuracy rate for the IPI cohort was 58.85% (SD=42.4%; CI=49.12, 68.58), which indicates a positive mean difference of 13.4%. (CI=4.45, 22.35) An N-1 Chi-square test of independence indicated a 92% chance that the IPI cohort will have a higher accuracy rate. Interpreter self-rated confidence also increased between cohorts from a mean of 4.9/10 in the control cohort to 6.8/10 in the IPI cohort (p=0.06). Whilst the IPI cohort had greater diagnostic accuracy, the duration of ECG interpretation was six times longer when compared to the control cohort. CONCLUSIONS: We have developed a system that segments and presents the ECG across five graphical user interfaces. Results indicate that this approach improves diagnostic accuracy but with the expense of time, which is a valuable resource in medical practice.

Automated detection of atrial fibrillation using R-R intervals and multivariate-based classification.

Automated detection of AF from the electrocardiogram (ECG) still remains a challenge. In this study, we investigated two multivariate-based classification techniques, Random Forests (RF) and k-nearest neighbor (k-nn), for improved automated detection of AF from the ECG. We have compiled a new database from ECG data taken from existing sources. R-R intervals were then analyzed using four previously described R-R irregularity measurements: (1) the coefficient of sample entropy (CoSEn), (2) the coefficient of variance (CV), (3) root mean square of the successive differences (RMSSD), and (4) median absolute deviation (MAD). Using outputs from all four R-R irregularity measurements, RF and k-nn models were trained. RF classification improved AF detection over CoSEn with overall specificity of 80.1% vs. 98.3% and positive predictive value of 51.8% vs. 92.1% with a reduction in sensitivity, 97.6% vs. 92.8%. k-nn also improved specificity and PPV over CoSEn; however, the sensitivity of this approach was considerably reduced (68.0%).

Computing the spatial QRS-T angle using reduced electrocardiographic lead sets.

The 'spatial QRS-T angle' (SA) is frequently determined using linear lead transformation matrices that require the entire 12-lead electrocardiogram (ECG). While this approach is adequate when using 12-lead ECG data that is recorded in the resting supine position, it is not optimal in monitoring applications. This is because maintaining a good quality recording of the complete 12-lead ECG in monitoring applications is difficult. In this research, we assessed the differences between the 'gold standard' SA as determined using the Frank VGG and the SA as determined using different reduced lead systems (RLSs). The random error component (span of the Bland-Altman 95% limits of agreement) of the differences between the 'gold standard' SA and the SA values based upon the different RLSs was quantified. This was performed for all 62 RLSs that can be constructed from Mason-Likar (ML) limb leads I, II and all possible precordial lead subsets that contain between one and five of the precordial leads V1 to V6. The RLS with the smallest lead set size that produced SA estimates of a quality similar to what is achieved using the ML 12-lead ECG was based upon ML limb leads I, II and precordial leads V1, V3 and V6. The random error component (mean [95% confidence interval]) associated with this RLS and the ML 12-lead ECG were found to be 40.74° [35.56°-49.29°] and 39.57° [33.78°-45.70°], respectively. Our findings suggest that a RLS that is based upon the ML limb leads I and II and the three best precordial leads can yield SA estimates of a quality similar to what is achieved when using the complete ML 12-lead ECG.

Human factors analysis of the CardioQuick Patch®: A novel engineering solution to the problem of electrode misplacement during 12-lead electrocardiogram acquisition.

INTRODUCTION: The CardioQuick Patch® (CQP) has been developed to assist operators in accurately positioning precordial electrodes during 12-lead electrocardiogram (ECG) acquisition. This study describes the CQP design and assesses the device in comparison to conventional electrode application. METHODS: Twenty ECG technicians were recruited and a total of 60 ECG acquisitions were performed on the same patient model over four phases: (1) all participants applied single electrodes to the patient; (2) all participants were then re-trained on electrode placement and on how to use the CQP; (3) participants were randomly divided into two groups, the standard group applied single electrodes and the CQP group used the CQP; (4) after a one day interval, the same participants returned to carry out the same procedure on the same patient (measuring intra-practitioner variability). Accuracy was measured with reference to pre-marked correct locations using ultra violet ink. NASA-TLK was used to measure cognitive workload and the Systematic Usability Scale (SUS) was used to quantify the usability of the CQP. RESULTS: There was a large difference between the minimum time taken to complete each approach (CQP=38.58s vs. 65.96s). The standard group exhibited significant levels of electrode placement error (V1=25.35mm±29.33, V2=18.1mm±24.49, V3=38.65mm±15.57, V4=37.73mm±12.14, V5=35.75mm±15.61, V6=44.15mm±14.32). The CQP group had statistically greater accuracy when placing five of the six electrodes (V1=6.68mm±8.53 [p<0.001], V2=8.8mm±9.64 [p=0.122], V3=6.83mm±8.99 [p<0.001], V4=14.90mm±11.76 [p<0.001], V5=8.63mm±10.70 [p<0.001], V6=18.13mm±14.37 [p<0.001]). There was less intra-practitioner variability when using the CQP on the same patient model. NASA TLX revealed that the CQP did increase the cognitive workload (CQP group=16.51%±8.11 vs. 12.22%±8.07 [p=0.251]). The CQP also achieved a high SUS score of 91±7.28. CONCLUSION: The CQP significantly improved the reproducibility and accuracy of placing precordial electrodes V1, V3-V6 with little additional cognitive effort, and with a high degree of usability.

Human factors approach to evaluate the user interface of physiologic monitoring.

BACKGROUND: As technology infiltrates more of our personal and professional lives, user expectations for intuitive design have driven many consumer products, while medical equipment continues to have high training requirements. Not much is known about the usability and user experience associated with hospital monitoring equipment. This pilot project aimed to better understand and describe the user interface interaction and user experience with physiologic monitoring technology. DESIGN: This was a prospective, descriptive, mixed-methods quality improvement project to analyze perceptions and task analyses of physiologic monitors. METHODS: Following a survey of practice patterns and perceived abilities to accomplish key tasks, 10 voluntary experienced physician and nurse subjects were asked to perform a series of tasks in 7 domains of monitor operations on GE Monitoring equipment in a single institution. For each task analysis, data were collected on time to complete the task, the number of button pushes or clicks required to accomplish the task, economy of motion, and observed errors. RESULTS: Although 60% of the participants reported incorporating monitoring data into patient care, 80% of participants preferred to receive monitoring data at the point of care (bedside). Average perceived central station usability is 5.3 out of 10 (ten is easiest). CONCLUSIONS: High variability exists in monitoring station interaction performance among those participating in this project. Alarms were almost universally silenced without cognitive recognition of the alarm state. Education related to monitoring operations appeared largely absent in this sample. Most users perceived the interface to not be intuitive, complaining of multiple layers and steps for data retrieval. These clinicians report real-time monitoring helpful for abrupt changes in condition like arrhythmias; however, reviewing alarms is not prioritized as valuable due to frequent false alarms. Participants requested exporting monitoring data to electronic medical records. Much research is needed to develop best practices for display of real-time information, organization and filtering of meaningful data, and simplified ways to find information.

Using computerised interactive response technology to assess electrocardiographers and for aggregating diagnoses.

The 12-lead electrocardiogram (ECG) is a crucial diagnostic tool. However, the ideal method to assess competency in ECG interpretation remains unclear. We sought to evaluate whether keypad response technology provides a rapid, interactive way to assess ECG knowledge. 75 participants were enrolled [32 (43%) Primary Care Physicians, 24 (32%) Hospital Medical Staff and 19 (25%) Nurse Practitioners]. Nineteen ECGs with 4 possible answers were interpreted. Out of 1425 possible decisions 1054 (73.9%) responses were made. Only 570/1425 (40%) of the responses were correct. Diagnostic accuracy varied (0% to 78%, mean 42%±21%) across the entire cohort. Participation was high, (median 83%, IQR 50%-100%). Hospital Medical Staff had significantly higher diagnostic accuracy than nurse practitioners (50±20% vs. 38±19%, p=0.04) and Primary Care Physicians (50±20% vs. 40±21%, p=0.07) although not significant. Interactive voting systems can be rapidly and successfully used to assess ECG interpretation. Further education is necessary to improve diagnostic accuracy.

Improved recording of atrial activity by modified bipolar leads derived from the 12-lead electrocardiogram.

This study investigates the use of multivariate linear regression to estimate three bipolar ECG leads from the 12-lead ECG in order to improve P-wave signal strength. The study population consisted of body surface potential maps recorded from 229 healthy subjects. P-waves were then isolated and population based transformation weights developed. A derived P-lead (measured between the right sternoclavicular joint and midway along the costal margin in line with the seventh intercostal space) demonstrated significant improvement in median P-wave root mean square (RMS) signal strength when compared to lead II (94µV vs. 76µV, p<0.001). A derived ES lead (from the EASI lead system) also showed small but significant improvement in median P-wave RMS (79µV vs. 76µV, p=0.0054). Finally, a derived modified Lewis lead did not improve median P-wave RMS when compared to lead II. However, this derived lead improved atrioventricular RMS ratio. P-wave leads derived from the 12-lead ECG can improve signal-to-noise ratio of the P-wave; this may improve the performance of detection algorithms that rely on P-wave analysis.

The derivation of the spatial QRS-T angle and the spatial ventricular gradient using the Mason-Likar 12-lead electrocardiogram.

Research has shown that the 'spatial QRS-T angle' (SA) and the 'spatial ventricular gradient' (SVG) have clinical value in a number of different applications. The determination of the SA and the SVG requires vectorcardiographic data. Such data is seldom recorded in clinical practice. The SA and the SVG are therefore frequently derived from 12-lead electrocardiogram (ECG) data using linear lead transformation matrices. This research compares the performance of two previously published linear lead transformation matrices (Kors and ML2VCG) in deriving the SA and the SVG from Mason-Likar (ML) 12-lead ECG data. This comparison was performed through an analysis of the estimation errors that are made when deriving the SA and the SVG for all 181 subjects in the study population. The estimation errors were quantified as the systematic error (mean difference) and the random error (span of the Bland-Altman 95% limits of agreement). The random error was found to be the dominating error component for both the Kors and the ML2VCG matrix. The random error [ML2VCG; Kors; result of the paired, two-sided Pitman-Morgan test for statistical significance of differences in the error variance between ML2VCG and Kors] for the vectorcardiographic parameters SA, magnitude of the SVG, elevation of the SVG and azimuth of the SVG were found to be [37.33°; 50.52°; p<0.001], [30.17mVms; 39.09mVms; p<0.001], [36.77°; 47.62°; p=0.001] and [63.45°; 80.32°; p<0.001] respectively. The findings of this research indicate that in comparison to the Kors matrix the ML2VCG provides greater precision for estimating the SA and SVG from ML 12-lead ECG data.

Data analysis of diagnostic accuracies in 12-lead electrocardiogram interpretation by junior medical fellows.

BACKGROUND: The electrocardiogram (ECG) is the most commonly used diagnostic procedure for assessing the cardiovascular system. The aim of this study was to compare ECG diagnostic skill among fellows of cardiology and of other internal medicine specialties (non-cardiology fellows). METHODS: A total of 2900 ECG interpretations were collected. A set of 100 clinical 12-lead ECG tracings were selected and classified into 12 diagnostic categories. The ECGs were evaluated by 15 cardiology fellows and of 14 non-cardiology fellows. Diagnostic interpretations were classified as (1) correct, (2) almost correct, (3) incorrect, and (4) dangerously incorrect. Multivariate logistic regression was used to assess confounding factors and to determine the odds ratios for the months of experience, age, sex, and the distinction between cardiology and non-cardiology fellows. RESULTS: The mean rate of correct diagnoses by cardiology vs. non-cardiology fellows was 48.9±8.9% vs. 35.9±8.0% (p=0.001; 70.1% vs. 55.0% for the aggregate of 'correct' and 'almost correct' diagnoses). There were 10.2±5.6% of interpretations classified as 'dangerously incorrect' by cardiology fellows vs. 16.3±5.0% by non-cardiology fellows (p=0.008). The cardiology fellows achieved statistically significantly greater diagnostic accuracy in 7 out of the 12 diagnostic classes. In multivariable logistic regression, the distinction between cardiology and non-cardiology fellows was the only independent statistically significant (p<0.001) predictor of whether the reader is likely correct or incorrect. Being a non-cardiology fellow reduced the probability of correct classification by 42% (odds ratio [95% confidence interval]: 0.58 [0.50; 0.68]). CONCLUSIONS: Although cardiology fellows out-performed the others, skills in ECG interpretation were found not adequately proficient. A comprehensive approach to ECG education is necessary. Further studies are needed to evaluate proper methods of training, testing, and continuous medical education in ECG interpretation.

Methods for presenting and visualising electrocardiographic data: From temporal signals to spatial imaging.

INTRODUCTION: The electrocardiogram (ECG) is a recording of the electrical activity of the heart. It is commonly used to non-invasively assess the cardiac activity of a patient. Since 1938, ECG data has been visualised as 12 scalar traces (known as the standard 12-lead ECG). Although this is known as the standard approach, there has been a myriad of alternative methods proposed to visualise ECG data. The purpose of this paper is to provide an overview of these methods and to introduce the field of ECG visualisation to early stage researchers. A scientific purpose is to consider the future of ECG visualisation within routine clinical practice. METHODS: This paper structures the different ECG visualisation methods using four categories, i.e. temporal, vectorial, spatial and interactive. Temporal methods present the data with respect to time, vectorial methods present data with respect to direction and magnitude, spatial methods present data in 2D or 3D space and interactive methods utilise interactive computing to facilitate efficient interrogation of ECG data at different levels of detail. CONCLUSION: Spatial visualisation has been around since its introduction by Waller and vector based visualisation has been around since the 1920s. Given these approaches have already been given the 'test of time', they are unlikely to be replaced as the standard in the near future. Instead of being replaced, the standard is more likely to be 'supplemented'. However, the design and presentation of these ECG visualisation supplements need to be universally standardised. Subsequent to the development of 'standardised supplements', as a requirement, they could then be integrated into all ECG machines. We recognise that without intuitive software and interactivity on mobile devices (e.g. tablet PCs), it is impractical to integrate the more advanced ECG visualisation methods into routine practice (i.e. epicardial mapping using an inverse solution).

Estimation performance of a reduced lead system during continuous 12-lead ECG ST-segment monitoring.

BACKGROUND: Reduced lead systems utilizing patient-specific transformation weights have been reported to achieve superior estimates than those utilizing population-based transformation weights. We report upon the effects of ischemic-type electrocardiographic changes on the estimation performance of a reduced lead system when utilizing patient-specific transformation weights and population-based transformation weights. METHOD: A reduced lead system that used leads I, II, V2 and V5 to estimate leads V1, V3, V4, and V6 was investigated. Patient-specific transformation weights were developed on electrocardiograms containing no ischemic-type changes. Patient-specific and population-based transformations weights were assessed on 45 electrocardiograms with ischemic-type changes and 59 electrocardiograms without ischemic-type changes. RESULTS: For patient-specific transformation weights the estimation performance measured as median root mean squared error values (no ischemic-type changes vs. ischemic-type changes) was found to be (V1, 27.5 µV vs. 95.8 µV, P<.001; V3, 33.9 µV vs. 65.2 µV, P<.001; V4, 24.8 µV vs. 62.0 µV, P<.001; V6, 11.7 µV vs. 51.5 µV, P<.001). The median magnitude of ST-amplitude difference 60 ms after the J-point between patient-specific estimated leads and actual recorded leads (no ischemic-type changes vs. ischemic-type changes) was found to be (V1, 18.9 µV vs. 61.4 µV, P<.001; V3, 14.3 µV vs. 61.1 µV, P<.001; V4, 9.7 µV vs. 61.3 µV, P<.001; V6, 5.9 µV vs. 46.0 µV, P<.001). CONCLUSION: The estimation performance of patient-specific transformations weights can deteriorate when ischemic-type changes develop. Performance assessment of patient-specific transformation weights should be performed using electrocardiographic data that represent the monitoring situation for which the reduced lead system is targeted.

Machine learning and the electrocardiogram over two decades: Time series and meta-analysis of the algorithms, evaluation metrics and applications.

BACKGROUND: The application of artificial intelligence to interpret the electrocardiogram (ECG) has predominantly included the use of knowledge engineered rule-based algorithms which have become widely used today in clinical practice. However, over recent decades, there has been a steady increase in the number of research studies that are using machine learning (ML) to read or interrogate ECG data. OBJECTIVE: The aim of this study is to review the use of ML with ECG data using a time series approach. METHODS: Papers that address the subject of ML and the ECG were identified by systematically searching databases that archive papers from January 1995 to October 2019. Time series analysis was used to study the changing popularity of the different types of ML algorithms that have been used with ECG data over the past two decades. Finally, a meta-analysis of how various ML techniques performed for various diagnostic classifications was also undertaken. RESULTS: A total of 757 papers was identified. Based on results, the use of ML with ECG data started to increase sharply (p < 0.001) from 2012. Healthcare applications, especially in heart abnormality classification, were the most common application of ML when using ECG data (p < 0.001). However, many new emerging applications include using ML and the ECG for biometrics and driver drowsiness. The support vector machine was the technique of choice for a decade. However, since 2018, deep learning has been trending upwards and is likely to be the leading technique in the coming few years. Despite the accuracy paradox, accuracy was the most frequently used metric in the studies reviewed, followed by sensitivity, specificity, F1 score and then AUC. CONCLUSION: Applying ML using ECG data has shown promise. Data scientists and physicians should collaborate to ensure that clinical knowledge is being applied appropriately and is informing the design of ML algorithms. Data scientists also need to consider knowledge guided feature engineering and the explicability of the ML algorithm as well as being transparent in the algorithm's performance to appropriately calibrate human-AI trust. Future work is required to enhance ML performance in ECG classification.