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

Machine learning based hybrid anomaly detection technique for automatic diagnosis of cardiovascular diseases using cardiac sympathetic nerve activity and electrocardiogram

OBJECTIVES

Coronary artery diseases (CADs) are the leading cause of death worldwide and early diagnosis is crucial for timely treatment. To address this, our study presents a novel automated Artificial Intelligence (AI)-based Hybrid Anomaly Detection (AIHAD) technique that combines various signal processing, feature extraction, supervised, and unsupervised machine learning methods. By jointly and simultaneously analyzing 12-lead cardiac sympathetic nerve activity (CSNA) and electrocardiogram (ECG) data, the automated AIHAD technique performs fast, early, and accurate diagnosis of CADs.

METHODS

In order to develop and evaluate the proposed automated AIHAD technique, we utilized the fully labeled STAFF III and PTBD databases, which contain the 12-lead wideband raw recordings non-invasively acquired from 260 subjects. Using these wideband raw recordings, we developed a signal processing technique that simultaneously detects the 12-lead CSNA and ECG signals of all subjects. Using the pre-processed 12-lead CSNA and ECG signals, we developed a time-domain feature extraction technique that extracts the statistical CSNA and ECG features critical for the reliable diagnosis of CADs. Using the extracted discriminative features, we developed a supervised classification technique based on Artificial Neural Networks (ANNs) that simultaneously detects anomalies in the 12-lead CSNA and ECG data. Furthermore, we developed an unsupervised clustering technique based on Gaussian mixture models (GMMs) and Neyman-Pearson criterion, which robustly detects outliers corresponding to CADs.

RESULTS

Using the automated AIHAD technique, we have, for the first time, demonstrated a significant association between the increase in CSNA signals and anomalies in ECG signals during CADs. The AIHAD technique achieved highly reliable detection of CADs with a sensitivity of 98.48 %, specificity of 97.73 %, accuracy of 98.11 %, positive predictive value of 97.74 %, negative predictive value of 98.47 %, and F1-score of 98.11 %. Hence, the automated AIHAD technique demonstrates superior performance compared to the gold standard diagnostic test ECG in the diagnosis of CADs. Additionally, it outperforms other techniques developed in this study that separately utilize either only CSNA data or only ECG data. Therefore, it significantly increases the detection performance of CADs by taking advantage of the diversity in different data types and leveraging their strengths. Furthermore, its performance is comparatively better than that of most previously proposed machine and deep learning methods that exclusively used ECG data to diagnose or classify CADs. Additionally, it has a very low implementation time, which is highly desirable for real-time detection of CADs.

CONCLUSIONS

The proposed automated AIHAD technique may serve as an efficient decision-support system to increase physicians' success in fast, early, and accurate diagnosis of CADs. It may be highly beneficial and valuable, particularly for asymptomatic patients, for whom the diagnostic information provided by ECG alone is not sufficient to reliably diagnose the disease. Hence, it may significantly improve patient outcomes by enabling timely treatments and considerably reducing the mortality of cardiovascular diseases (CVDs).

Exploring decision making 'noise' when interpreting the electrocardiogram in the context of cardiac cath lab activation

In this commentary paper, we discuss the use of the electrocardiogram to help clinicians make diagnostic and patient referral decisions in acute care settings. The paper discusses the factors that are likely to contribute to the variability and noise in the clinical decision making process for catheterization lab activation. These factors include the variable competence in reading ECGs, the intra/inter rater reliability, the lack of standard ECG training, the various ECG machine and filter settings, cognitive biases (such as automation bias which is the tendency to agree with the computer-aided diagnosis or AI diagnosis), the order of the information being received, tiredness or decision fatigue as well as ECG artefacts such as the signal noise or lead misplacement. We also discuss potential research questions and tools that could be used to mitigate this 'noise' and improve the quality of ECG based decision making.

Machine learning with electrocardiograms: A call for guidelines and best practices for 'stress testing' algorithms

This paper provides a brief description of how computer programs are used to automatically interpret electrocardiograms (ECGs), and also provides a discussion regarding new opportunities. The algorithms that are typically used today in hospitals are knowledge engineered where a computer programmer manually writes computer code and logical statements which are then used to deduce a possible diagnosis. The computer programmer's code represents the criteria and knowledge that is used by clinicians when reading ECGs. This is in contrast to supervised machine learning (ML) approaches which use large, labelled ECG datasets to induct their own 'rules' to automatically classify ECGs. Although there are many ML techniques, deep neural networks are being increasingly explored as ECG classification algorithms when trained on large ECG datasets. Whilst this paper presents some of the pros and cons of each of these approaches, perhaps there are opportunities to develop hybridised algorithms that combine both knowledge and data driven techniques. In this paper, it is pointed out that open ECG data can dramatically influence what international ECG ML researchers focus on and that, ideally, open datasets could align with real world clinical challenges. In addition, some of the pitfalls and opportunities for ML with ECGs are outlined. A potential opportunity for the ECG community is to provide guidelines to researchers to help guide ECG ML practices. For example, whilst general ML guidelines exist, there is perhaps a need to recommend approaches for 'stress testing' and evaluating ML algorithms for ECG analysis, e.g. testing the algorithm with noisy ECGs and ECGs acquired using common lead and electrode misplacements. This paper provides a primer on ECG ML and discusses some of the key challenges and opportunities.

SPICED-ACS: Study of the potential impact of a computer-generated ECG diagnostic algorithmic certainty index in STEMI diagnosis: Towards transparent AI

BACKGROUND

Computerised electrocardiogram (ECG) interpretation diagnostic algorithms have been developed to guide clinical decisions like with ST segment elevation myocardial infarction (STEMI) where time in decision making is critical. These computer-generated diagnoses have been proven to strongly influence the final ECG diagnosis by the clinician; often called automation bias. However, the computerised diagnosis may be inaccurate and could result in a wrong or delayed treatment harm to the patient. We hypothesise that an algorithmic certainty index alongside a computer-generated diagnosis might mitigate automation bias. The impact of reporting a certainty index on the final diagnosis is not known.

PURPOSE

To ascertain whether knowledge of the computer-generated ECG algorithmic certainty index influences operator diagnostic accuracy.

METHODOLOGY

Clinicians who regularly analyse ECGs such as cardiology or acute care doctors, cardiac nurses and ambulance staff were invited to complete an online anonymous survey between March and April 2019. The survey had 36 ECGs with a clinical vignette of a typical chest pain and which were either a STEMI, normal, or borderline (but do not fit the STEMI criteria) along with an artificially created certainty index that was either high, medium, low or none. Participants were asked whether the ECG showed a STEMI and their confidence in the diagnosis. The primary outcomes were whether a computer-generated certainty index influenced interpreter's diagnostic decisions and improved their diagnostic accuracy. Secondary outcomes were influence of certainty index between different types of clinicians and influence of certainty index on user's own-diagnostic confidence.

RESULTS

A total of 91 participants undertook the survey and submitted 3262 ECG interpretations of which 75% of ECG interpretations were correct. Presence of a certainty index significantly increased the odds ratio of a correct ECG interpretation (OR 1.063, 95% CI 1.022-1.106, p = 0.004) but there was no significant difference between correct certainty index and incorrect certainty index (OR 1.028, 95% CI 0.923-1.145, p = 0.615). There was a trend for low certainty index to increase odds ratio compared to no certainty index (OR 1.153, 95% CI 0.898-1.482, p = 0.264) but a high certainty index significantly decreased the odds ratio of a correct ECG interpretation (OR 0.492, 95% CI 0.391-0.619, p < 0.001). There was no impact of presence of a certainty index (p = 0.528) or correct certainty index (p = 0.812) on interpreters' confidence in their ECG interpretation.

CONCLUSIONS

Our results show that the presence of an ECG certainty index improves the users ECG interpretation accuracy. This effect is not seen with differing levels of confidence within a certainty index, with reduced ECG interpretation success with a high certainty index compared with a trend for increased success with a low certainty index. This suggests that a certainty index improves interpretation when there is an increased element of doubt, possibly forcing the ECG user to spend more time and effort analysing the ECG. Further research is needed looking at time spent analysing differing certainty indices with alternate ECG diagnoses.

Impact of Mobile Device-Based Clinical Decision Support Tool on Guideline Adherence and Mental Workload

OBJECTIVE

To evaluate the individual-level impact of an electronic clinical decision support (ECDS) tool, PedsGuide, on febrile infant clinical decision making and cognitive load.

METHODS

A counterbalanced, prospective, crossover simulation study was performed among attending and trainee physicians. Participants performed simulated febrile infant cases with use of PedsGuide and with standard reference text. Cognitive load was assessed using the NASA-Task Load Index (NASA-TLX), which determines mental, physical, temporal demand, effort, frustration, and performance. Usability was assessed with the System Usability Scale (SUS). Scores on cases and NASA-TLX scores were compared between condition states.

RESULTS

A total of 32 participants completed the study. Scores on febrile infant cases using PedsGuide were greater compared with standard reference text (89% vs 72%, P = .001). NASA-TLX scores were lower (ie, more optimal) with use of PedsGuide versus control (mental 6.34 vs 11.8, P < .001; physical 2.6 vs 6.1, P = .001; temporal demand 4.6 vs 8.0, P = .003; performance 4.5 vs 8.3, P < .001; effort 5.8 vs 10.7, P < .001; frustration 3.9 vs 10, P < .001). The SUS had an overall score of 88 of 100 with rating of acceptable on the acceptability scale.

CONCLUSIONS

Use of PedsGuide led to increased adherence to guidelines and decreased cognitive load in febrile infant management when compared with the use of a standard reference tool. This study employs a rarely used method of assessing ECDS tools using a multifaceted approach (medical decision-making, assessing usability, and cognitive workload,) that may be used to assess other ECDS tools in the future.