Artificial intelligence and imaging: Opportunities in cardio-oncology.

Cardiovascular disease is a leading cause of death in cancer survivors. It is critical to apply new predictive and early diagnostic methods in this population, as this can potentially inform cardiovascular treatment and surveillance decision-making. We discuss the application of artificial intelligence (AI) technologies to cardiovascular imaging in cardio-oncology, with a particular emphasis on prevention and targeted treatment of a variety of cardiovascular conditions in cancer patients. Recently, the use of AI-augmented cardiac imaging in cardio-oncology is gaining traction. A large proportion of cardio-oncology patients are screened and followed using left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS), currently obtained using echocardiography. This use will continue to increase with new cardiotoxic cancer treatments. AI is being tested to increase precision, throughput, and accuracy of LVEF and GLS, guide point-of-care image acquisition, and integrate imaging and clinical data to optimize the prediction and detection of cardiac dysfunction. The application of AI to cardiovascular magnetic resonance imaging (CMR), computed tomography (CT; especially coronary artery calcium or CAC scans), single proton emission computed tomography (SPECT) and positron emission tomography (PET) imaging acquisition is also in early stages of analysis for prediction and assessment of cardiac tumors and cardiovascular adverse events in patients treated for childhood or adult cancer. The opportunities for application of AI in cardio-oncology imaging are promising, and if availed, will improve clinical practice and benefit patient care.

Bedside identification of patients at risk for PVC-induced cardiomyopathy: Is ECG useful?

BACKGROUND: Premature ventricular complexes (PVCs) are an underrecognized cause of cardiomyopathy. Standard 12-lead electrocardiogram (ECG) has potential to direct attention toward at-risk patients. METHODS: We performed a single-center, retrospective chart review of 1,240 patients who completed ECG and Holter monitoring at Oregon Health and Science University Hospital between January 1, 2011 and December 31, 2013 to investigate the relationship of PVC frequency on ECG with burden on Holter. Primary outcome measures included PVC quantity on ECG, mean PVC quantity on Holter, and percentage of total beats on Holter recorded as PVCs. High PVC burden was defined as ≥10% of total beats. RESULTS: Weighted mean percentages of total beats on Holter monitor recorded as PVCs were calculated for 0, 1, 2, and ≥3 PVCs on ECG and found to be 1.4% (n = 1,128), 3.5% (n = 32), 4.3% (n = 25), and 16.6% (n = 55), respectively, which represent statistically significant differences (P < 0.001). The positive predictive value of at least three PVCs on ECG for ≥10% PVC Holter burden was 58%. Negative predictive value for 0 PVCs on ECG was 98%. The sensitivity and specificity of ECG to identify high PVC burden on Holter was 72% and 93.6%, respectively, when utilizing a positive ECG result as one PVC or more, and 44% and 98.9%, respectively, with ≥3 PVCs on ECG. The positive likelihood ratio corresponding to ≥3 PVCs on ECG was 40. CONCLUSION: These findings demonstrate that the number of PVCs on ECG can be utilized for quick bedside estimation of high PVC burden.

Measuring health-related quality of life in population-based studies of coronary heart disease: comparing six generic indexes and a disease-specific proxy score.

PURPOSE: To compare HRQoL differences with CHD in generic indexes and a proxy CVD-specific score in a nationally representative sample of U.S. adults. METHODS: The National Health Measurement Study, a cross-sectional random-digit-dialed telephone survey of adults aged 35-89, administered the EQ-5D, QWB-SA, HUI2, HUI3, SF-36v2 (yielding PCS, MCS, and SF-6D), and HALex. Analyses compared 3,350 without CHD (group 1), 265 with CHD not taking chest pain medication (group 2), and 218 with CHD currently taking chest pain medication (group 3), with and without adjustment for demographic variables and comorbidities. Data on 154 patients from heart failure clinics were used to construct a proxy score utilizing generic items probing CVD symptoms. RESULTS: Mean scores differed between CHD groups for all indexes with and without adjustment (P < 0.0001 for all except MCS P = 0.018). Unadjusted group 3 versus 1 differences were about three times larger than for group 2 versus 1. Standardized differences for the proxy score were similar to those for generic indexes, and were about 1.0 for all except MCS for group 3 versus 1. CONCLUSIONS: Generic indexes capture differences in HRQoL in population-based studies of CHD similarly to a score constructed from questions probing CVD-specific symptoms.