Novel tech throws knock-out punch to ECG improving GP referral decisions to cardiology

Abstract

PurposeIn a comparator study, designed with assistance from the Food and Drug Administration, a State-of-the-Art (SOTA) ECG device augmented with automated analysis, the comparator, was compared with a breakthrough technology, Cardio-HART (CHART).MethodsThe referral decision defined by physician reading biosignal-based ECG or CHART report were compared for 550 patients, where its performance is calculated against the ground truth referral decision. The ground truth was established by cardiologist consensus based on all the available measurements and findings including echocardiography (ECHO).ResultsThe results confirmed that CHART analysis was far more effective than ECG only analysis: CHART reduced false negative rates 15.8% and false positive (FP) rates by 5%, when compared with SOTA ECG devices. General physicians (GP’s) using CHART saw their positive diagnosis rate significantly increased, from ~10% to ~26% (260% increase), and the uncertainty rate significantly decreased, from ~31% to ~1.9% (94% decrease). For cardiology, the study showed that in 98% of the cases, the CHART report was found to be a good indicator as to what kind of heart problems can be expected (the ‘start-point’) in the ECHO examination.ConclusionsThe study revealed that GP use of CHART resulted in more accurate referrals for cardiology, resulting in fewer true negative or FP—healthy or mildly abnormal patients not in need of ECHO confirmation. The indirect benefit is the reduction in wait-times and in unnecessary and costly testing in secondary care. Moreover, when used as a start-point, CHART can shorten the echocardiograph examination time.

Unraveling diagnostic co-morbidity makeup of each HF category as characteristically derived by ECG- and ECHO-findings

“…Unlike ECG, bio-signal derived ECHO-findings processed through a purpose designed AI system, can predict the structural and functional abnormalities and measurements, consistent with table 2 above, essential to the detection accuracy for HF, 49 including low LVEF.…”

“…Data used in the research were collection in a prior clinical study, already published with the Declaration of Helsinki state. 49 To analyze prevalence along four HF categories a statistical analysis was performed on patient groupings into a) All – all the patient records, b) Absent HF records, c) HFpEF records, d) HFmrEF records and e) HFrEF records.…”

Detecting Heart Failure using novel bio-signals and a Knowledge Enhanced Neural Network

“…The implications of this breakthrough technology will resonate throughout the healthcare system starting in Primary Care, where this low-cost, non-invasive, immediate use on initial patient presentation will, 1) drastically reduce time to diagnosis from months, to one day; 2) enable early detection of HF, by phenotype; which in turn 3) provide access to treatments earlier, 4) reduce the damage caused by delayed diagnosis; 5) reduce failures to diagnose, especially in women patients [50]; and, 6) clinically proven to reduce unnecessary referrals to cardiology [51]. The potential costs savings to healthcare are significant.…”

“…The impact also accrues to cardiology; they will get a better patient. As a direct substitute for ECG, prior to conducting a full echo examination, the implications are [51]: 1) screening for true negative patients that don't need an echo; 2) screening for the presence of HF, valve disease, and myocardial infarctions, when the incoming referral fails to indicate these conditions; and critically, 3) providing a "big picture" of a patient's overall cardiac status and revealing potential comorbidities that might otherwise be missed.…”

Will Artificial Intelligence Provide Answers to Current Gaps and Needs in Chronic Heart Failure?

“…Cardio-HART HF prediction reaches significantly higher overall performance compared to the best ECG criteria, with a sensitivity of 83 per cent, specificity of 87 per cent and a positive predictive value of 70 per cent. CHART has the potential to enable effective widespread screening of patients for the early detection of CVD onset and resolve many 'inconclusive ECG' results, thereby reducing time in referral decisions [66]. Along the same line, in a recent retrospective study, the authors developed novel machine learning (ML) models to predict HF-related mortality, incorporating social determinants of health.…”