According to 2016 data from the Centers for Disease Control and Prevention, more than 41 million ECGs are performed annually in the United States. The enormous amount of patient data yielded can provide a wealth of information regarding cardiac diagnoses, but clinicians may balk at the work required to sift through such massive data sets, especially if the data comes from multiple healthcare organizations.
Fortunately, a combination of comprehensive cardiology data management systems and artificial intelligence (AI) can help streamline this process and provide pertinent insights, allowing cardiologists to better identify risk factors for significant cardiovascular events.
Benefits of Comprehensive Cardiology Data Management Systems
Several new technologies provide data storage platforms that can streamline communication, keep patient data secure, and integrate with other systems (such as electronic medical records) to grant clinicians access to all pertinent patient data. And because many of these systems seamlessly connect to diagnostic testing devices such as ECG, healthcare providers can retrieve near real-time data, which can prove invaluable for arriving at a correct diagnosis.
With the assistance of these systems, cardiologists can readily view patient information and clinical results from any location, which can provide greater workflow flexibility and lead to better care for the patient. In addition, collecting and examining ECGs retrospectively can help reduce future diagnostic inaccuracies and guide better clinical decision-making.
Data Mining and ECG Retrospective Analysis
Several studies have illustrated how large-scale ECG analysis can uncover useful insights—such as the identification of inconsistent methods of measurement, overlooked abnormalities that may suggest underlying disease, or the impact of making a change in how ECGs are processed across an institution.
A recently published study in Cureus compared Bazett's QT interval formula (QTcB) with Fridericia's formula (QTcFri) to determine the extent of difference in those identified as "prolonged QT" in more than 44,000 ECGs. The impact was significant. There were 57% fewer ECGs identified as having a QTc > 500ms when using Fridericia.
The threshold of 500ms is key here since it is the value selected for identifying a patient at risk of drug-induced torsade de pointes (TdP) in a scientific statement from the American Heart Association (AHA) and the American College of Cardiology Foundation.
Hence, a change of 57% impacting thousands of ECGs in this hospital-based study has important clinical consequences, including the administration or withholding of medication.
However, the AHA also specified that an increase of 60ms after administering a drug known to be associated with TdP is as risky as a patient with an ECG that has a QTc > 500ms. Consequently, when comparing ECGs over time, it is vital that a consistent method be used, including the correction formula for correcting QT for heart rate.
The study published in Cureus covers in significant detail the reasons why the Bazett and Fridericia formulas result in different values for a given QT and heart rate, with Bazett generating longer QTc values at higher heart rates. Regardless, the question of which is the best correction formula for QT is still a matter of debate, and evidence suggests the relationship between the QT and RR intervals depends on the characteristics of the person being tested.
Therefore, it is advisable that whenever possible, a resting ECG should be acquired when the heart rate is steady and preferably between 60 and 100 beats per minute when assessing a patient for prolonged QT.
Retrospective analysis of ECGs may be useful for determining cost-effectiveness and optimized workflow. A 2015 study in The Journal of Pediatrics examined clinical data from over 8,600 ECGs performed as part of athletic pre-participation evaluations (PPE) to assess children and young adults' readiness to participate in sports. From 2005 to 2010, pediatricians selectively performed ECG during the PPE based on suspicious cardiac symptoms, such as dizziness or syncope. Only 0.5% of individuals receiving PPE alone were referred to cardiologists based on clinical suspicion, but 13% of individuals receiving PPE and an ECG were referred to cardiologists for further evaluation.
Evaluation of the study's data years later further supports the use of ECG for diagnostic purposes, as PPE alone identified cardiac disease with a sensitivity of only 44%. Within one year of the PPE, cardiac disease was diagnosed in 0.5% of people receiving the test. However, incorporating the ECG into a PPE helped identify underlying cardiac disease in 18% of participants, allowing more patients to receive therapeutic procedures.
To learn more about the power of the ECG in today's clinical landscape, browse our Diagnostic ECG Clinical Insights Center.
AI's Role in Evaluating Data
The expanding use of AI can also benefit clinicians seeking to analyze previously recorded ECGs. The machine learning techniques involved go beyond evaluating only a small portion of the information contained in an ECG readout; rather, researchers are aiming to develop algorithmic frameworks that support large-scale ECG analysis.
Researchers created such an algorithm and published their findings in Circulation: Cardiovascular Quality and Outcomes with the ultimate goal of developing an automated, scalable, and interpretable method of characterizing cardiac structure and diastolic function, as well as detecting and tracking disease via patient-specific ECG profiles.
Using standard 12-lead ECGs collected from 2010 to 2017, the model was restricted to ECGs presenting normal sinus rhythm. Using continuous metrics, the AI approach enabled estimation of the severity of structural abnormalities, which were verified against reference echocardiographic measurements from GE's MUSE Cardiology Information System. The algorithm also helped to classify four example diseases (pulmonary arterial hypertension, hypertrophic cardiomyopathy, cardiac amyloidosis, and mitral valve prolapse), even identifying new ECG predictors for each disease.
Other research supports the use of machine learning techniques for interpreting ECG findings. Research in Nature Communications examined the use of deep neural networks (DNNs) in classifying six types of abnormalities (first degree AV block, right bundle branch block, left bundle branch block, sinus bradycardia, atrial fibrillation, and sinus tachycardia) by employing data from over two million standard 12-lead ECGs. The data was split into a training set to develop the DNN, and a validation set to confirm the results.
When compared to interpretations from two fourth-year cardiology residents, two third-year emergency residents, and two fifth-year medical students, the DNN matched or outperformed all people in identifying the six cardiac abnormalities. Other work published in Nature Medicine supports these conclusions, showing that another DNN exceeded average cardiologist sensitivity for all rhythm classes in a sample of over 91,000 single-lead ECGs.
The authors of the report hypothesize that future availability of large-scale ECG tracing databases, combined with DNNs performing automatic ECG analysis, could save clinicians time and help to prevent incorrect diagnoses.
ECG Storage in the COVID-19 Era
ECG storage is especially important as healthcare providers grapple with the COVID-19 pandemic because the health consequences of the infection are not yet fully understood. Current research, such as that in the Journal of Cardiac Failure, shows that several ECG abnormalities may be prognostic, specifically for death from COVID-19.
The authors of the study found that the presence of one or more atrial premature contractions, a right bundle branch block, an ischemic T-wave inversion, or nonspecific repolarization increased the odds of death. While this study was completed using ECG interpretations from two electrocardiographers, training AI to recognize these conduction problems could provide more accurate prognostic information for patients with COVID-19.
Insights gained from analysis of ECG may prove indispensable in future diagnostic and prognostic cases. Clinicians should continue to analyze ECGs using best practices and consider the potential of comprehensive data management systems and machine learning techniques to supplement current understanding and knowledge. Ultimately, any technique that provides more accurate data about a patient's cardiac condition is valuable and could help to direct an appropriate course of treatment.