Unsupervised machine learning identifies symptoms of indigestion as a predictor of acute decompensation and adverse cardiac events in patients with heart failure presenting to the emergency department

BACKGROUND

Patients with known heart failure (HF) present to emergency departments (ED) with a plethora of symptoms. Although symptom clusters have been suggested as prognostic features, accurately triaging HF patients is a longstanding challenge.

OBJECTIVES

We sought to use machine learning to identify subtle phenotypes of patient symptoms and evaluate their diagnostic and prognostic value among HF patients seeking emergency care.

METHODS

This was a secondary analysis of a prospective cohort study of consecutive patients seen in the ED for chest pain or equivalent symptoms. Independent reviewers extracted clinical data from charts, including nine categories of subjective symptoms reported during initial evaluation. The diagnostic outcome was acute HF exacerbation and prognostic outcome was 30-day major adverse cardiac events (MACE). Outcomes were adjudicated by two independent reviewers. K-means clustering was used to derive latent patient symptom clusters, and their associations with outcomes were assessed using multivariate logistic regression.

RESULTS

Sample included 438 patients (age 65±14 years; 45% female, 49% Black, 18% HF exacerbation, 32% MACE). K-means clustering identified three presentation phenotypes: patients with dyspnea only (Cluster A, 40%); patients with indigestion, with or without dyspnea (Cluster B, 23%); patients with neither dyspnea nor indigestion (Cluster C, 37%). Compared to Cluster C, indigestion was a significant predictor of acute HF exacerbation (OR=1.8, 95%CI=1.0-3.4) and 30-day MACE (OR=1.8, 95%CI=1.0-3.1), independent of age, sex, race, and other comorbidities.

CONCLUSION

Indigestion symptoms in patients with known HF signify excess risk of adverse events, suggesting that these patients should be triaged as high-risk during initial ED evaluation.

Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction

Patients with occlusion myocardial infarction (OMI) and no ST-elevation on presenting electrocardiogram (ECG) are increasing in numbers. These patients have a poor prognosis and would benefit from immediate reperfusion therapy, but, currently, there are no accurate tools to identify them during initial triage. Here we report, to our knowledge, the first observational cohort study to develop machine learning models for the ECG diagnosis of OMI. Using 7,313 consecutive patients from multiple clinical sites, we derived and externally validated an intelligent model that outperformed practicing clinicians and other widely used commercial interpretation systems, substantially boosting both precision and sensitivity. Our derived OMI risk score provided enhanced rule-in and rule-out accuracy relevant to routine care, and, when combined with the clinical judgment of trained emergency personnel, it helped correctly reclassify one in three patients with chest pain. ECG features driving our models were validated by clinical experts, providing plausible mechanistic links to myocardial injury.

Machine Learning for the ECG Diagnosis and Risk Stratification of Occlusion Myocardial Infarction at First Medical Contact

Patients with occlusion myocardial infarction (OMI) and no ST-elevation on presenting ECG are increasing in numbers. These patients have a poor prognosis and would benefit from immediate reperfusion therapy, but we currently have no accurate tools to identify them during initial triage. Herein, we report the first observational cohort study to develop machine learning models for the ECG diagnosis of OMI. Using 7,313 consecutive patients from multiple clinical sites, we derived and externally validated an intelligent model that outperformed practicing clinicians and other widely used commercial interpretation systems, significantly boosting both precision and sensitivity. Our derived OMI risk score provided superior rule-in and rule-out accuracy compared to routine care, and when combined with the clinical judgment of trained emergency personnel, this score helped correctly reclassify one in three patients with chest pain. ECG features driving our models were validated by clinical experts, providing plausible mechanistic links to myocardial injury.

Incorporation of Serial 12-Lead Electrocardiogram With Machine Learning to Augment the Out-of-Hospital Diagnosis of Non-ST Elevation Acute Coronary Syndrome

STUDY OBJECTIVE

Ischemic electrocardiogram (ECG) changes are subtle and transient in patients with suspected non-ST-segment elevation (NSTE)-acute coronary syndrome. However, the out-of-hospital ECG is not routinely used during subsequent evaluation at the emergency department. Therefore, we sought to compare the diagnostic performance of out-of-hospital and ED ECG and evaluate the incremental gain of artificial intelligence-augmented ECG analysis.

METHODS

This prospective observational cohort study recruited patients with out-of-hospital chest pain. We retrieved out-of-hospital-ECG obtained by paramedics in the field and the first ED ECG obtained by nurses during inhospital evaluation. Two independent and blinded reviewers interpreted ECG dyads in mixed order per practice recommendations. Using 179 morphological ECG features, we trained, cross-validated, and tested a random forest classifier to augment non ST-elevation acute coronary syndrome (NSTE-ACS) diagnosis.

RESULTS

Our sample included 2,122 patients (age 59 [16]; 53% women; 44% Black, 13.5% confirmed acute coronary syndrome). The rate of diagnostic ST elevation and ST depression were 5.9% and 16.2% on out-of-hospital-ECG and 6.1% and 12.4% on ED ECG, with ∼40% of changes seen on out-of-hospital-ECG persisting and ∼60% resolving. Using expert interpretation of out-of-hospital-ECG alone gave poor baseline performance with area under the receiver operating characteristic (AUC), sensitivity, and negative predictive values of 0.69, 0.50, and 0.92. Using expert interpretation of serial ECG changes enhanced this performance (AUC 0.80, sensitivity 0.61, and specificity 0.93). Interestingly, augmenting the out-of-hospital-ECG alone with artificial intelligence algorithms boosted its performance (AUC 0.83, sensitivity 0.75, and specificity 0.95), yielding a net reclassification improvement of 29.5% against expert ECG interpretation.

CONCLUSION

In this study, 60% of diagnostic ST changes resolved prior to hospital arrival, making the ED ECG suboptimal for the inhospital evaluation of NSTE-ACS. Using serial ECG changes or incorporating artificial intelligence-augmented analyses would allow correctly reclassifying one in 4 patients with suspected NSTE-ACS.

Remote and Wearable ECG Devices with Diagnostic Abilities in Adults: A State-of-the-Science Scoping Review

The electrocardiogram (ECG) records the electrical activity in the heart in real-time, providing an important opportunity to detecting various cardiac pathologies. The 12-lead ECG currently serves as the "standard" ECG acquisition technique for diagnostic purposes for many cardiac pathologies other than arrhythmias. However, the technical aspects of acquiring a 12-lead ECG are not easy and its usage is currently restricted to trained medical personnel, limiting the scope of its usefulness. Remote and wearable ECG devices have attempted to bridge this gap by enabling patients to take their own ECG using a simplified method at the expense of a reduced number of leads, usually a single-lead ECG. In this review article, we summarize the studies which investigate the use of remote ECG devices and their clinical utility in diagnosing cardiac pathologies. Eligible studies discussed FDA-cleared, commercially available devices that were validated on an adult population. We summarize technical logistics of signal quality and device reliability, dimensional and functional features, and diagnostic value. In summary, our synthesis shows that reduced-set ECG wearables have huge potential for long-term monitoring, particularly if paired with real-time notification techniques. Such capabilities make them primarily useful for abnormal rhythm detection and there is sufficient evidence that a remote ECG device can be more superior to traditional 12-lead ECG in diagnosing specific arrhythmias such as atrial fibrillation. However, this review identifies important challenges faced by this technology, highlighting the limited availability of clinical research examining their usefulness.

Your neighborhood matters: A machine-learning approach to the geospatial and social determinants of health in 9-1-1 activated chest pain

Healthcare disparities in the initial management of patients with acute coronary syndrome (ACS) exist. Yet, the complexity of interactions between demographic, social, economic, and geospatial determinants of health hinders incorporating such predictors in existing risk stratification models. We sought to explore a machine-learning-based approach to study the complex interactions between the geospatial and social determinants of health to explain disparities in ACS likelihood in an urban community. This study identified consecutive patients transported by Pittsburgh emergency medical service for a chief complaint of chest pain or ACS-equivalent symptoms. We extracted demographics, clinical data, and location coordinates from electronic health records. Median income was based on US census data by zip code. A random forest (RF) classifier and a regularized logistic regression model were used to identify the most important predictors of ACS likelihood. Our final sample included 2400 patients (age 59 ± 17 years, 47% Females, 41% Blacks, 15.8% adjudicated ACS). In our RF model (area under the receiver operating characteristic curve of 0.71 ± 0.03) age, prior revascularization, income, distance from hospital, and residential neighborhood were the most important predictors of ACS likelihood. In regularized regression (akaike information criterion = 1843, bayesian information criterion = 1912, χ²  = 193, df = 10, p < 0.001), residential neighborhood remained a significant and independent predictor of ACS likelihood. Findings from our study suggest that residential neighborhood constitutes an upstream factor to explain the observed healthcare disparity in ACS risk prediction, independent from known demographic, social, and economic determinants of health, which can inform future work on ACS prevention, in-hospital care, and patient discharge.

Novel ECG features and machine learning to optimize culprit lesion detection in patients with suspected acute coronary syndrome

BACKGROUND

Novel temporal-spatial features of the 12‑lead ECG can conceptually optimize culprit lesions' detection beyond that of classical ST amplitude measurements. We sought to develop a data-driven approach for ECG feature selection to build a clinically relevant algorithm for real-time detection of culprit lesion.

METHODS

This was a prospective observational cohort study of chest pain patients transported by emergency medical services to three tertiary care hospitals in the US. We obtained raw 10-s, 12‑lead ECGs (500 s/s, HeartStart MRx, Philips Healthcare) during prehospital transport and followed patients 30 days after the encounter to adjudicate clinical outcomes. A total of 557 global and lead-specific features of P-QRS-T waveform were harvested from the representative average beats. We used Recursive Feature Elimination and LASSO to identify 35/557, 29/557, and 51/557 most recurrent and important features for LAD, LCX, and RCA culprits, respectively. Using the union of these features, we built a random forest classifier with 10-fold cross-validation to predict the presence or absence of culprit lesions. We compared this model to the performance of a rule-based commercial proprietary software (Philips DXL ECG Algorithm).

RESULTS

Our sample included 2400 patients (age 59 ± 16, 47% female, 41% Black, 10.7% culprit lesions). The area under the ROC curves of our random forest classifier was 0.85 ± 0.03 with sensitivity, specificity, and negative predictive value of 71.1%, 84.7%, and 96.1%. This outperformed the accuracy of the automated interpretation software of 37.2%, 95.6%, and 92.7%, respectively, and corresponded to a net reclassification improvement index of 23.6%. Metrics of ST80; Tpeak-Tend; spatial angle between QRS and T vectors; PCA ratio of STT waveform; T axis; and QRS waveform characteristics played a significant role in this incremental gain in performance.

CONCLUSIONS

Novel computational features of the 12‑lead ECG can be used to build clinically relevant machine learning-based classifiers to detect culprit lesions, which has important clinical implications.

In Search of an Optimal Subset of ECG Features to Augment the Diagnosis of Acute Coronary Syndrome at the Emergency Department

Background Classical ST-T waveform changes on standard 12-lead ECG have limited sensitivity in detecting acute coronary syndrome (ACS) in the emergency department. Numerous novel ECG features have been previously proposed to augment clinicians' decision during patient evaluation, yet their clinical utility remains unclear. Methods and Results This was an observational study of consecutive patients evaluated for suspected ACS (Cohort 1 n=745, age 59±17, 42% female, 15% ACS; Cohort 2 n=499, age 59±16, 49% female, 18% ACS). Out of 554 temporal-spatial ECG waveform features, we used domain knowledge to select a subset of 65 physiology-driven features that are mechanistically linked to myocardial ischemia and compared their performance to a subset of 229 data-driven features selected by multiple machine learning algorithms. We then used random forest to select a final subset of 73 most important ECG features that had both data- and physiology-driven basis to ACS prediction and compared their performance to clinical experts. On testing set, a regularized logistic regression classifier based on the 73 hybrid features yielded a stable model that outperformed clinical experts in predicting ACS, with 10% to 29% of cases reclassified correctly. Metrics of nondipolar electrical dispersion (ie, circumferential ischemia), ventricular activation time (ie, transmural conduction delays), QRS and T axes and angles (ie, global remodeling), and principal component analysis ratio of ECG waveforms (ie, regional heterogeneity) played an important role in the improved reclassification performance. Conclusions We identified a subset of novel ECG features predictive of ACS with a fully interpretable model highly adaptable to clinical decision support applications. Registration URL: https://www.clinicaltrials.gov; Unique Identifier: NCT04237688.

Machine learning-based prediction of acute coronary syndrome using only the pre-hospital 12-lead electrocardiogram

Prompt identification of acute coronary syndrome is a challenge in clinical practice. The 12-lead electrocardiogram (ECG) is readily available during initial patient evaluation, but current rule-based interpretation approaches lack sufficient accuracy. Here we report machine learning-based methods for the prediction of underlying acute myocardial ischemia in patients with chest pain. Using 554 temporal-spatial features of the 12-lead ECG, we train and test multiple classifiers on two independent prospective patient cohorts (n = 1244). While maintaining higher negative predictive value, our final fusion model achieves 52% gain in sensitivity compared to commercial interpretation software and 37% gain in sensitivity compared to experienced clinicians. Such an ultra-early, ECG-based clinical decision support tool, when combined with the judgment of trained emergency personnel, would help to improve clinical outcomes and reduce unnecessary costs in patients with chest pain.