Ken Gin

Diastolic filling in human severe sepsis: an echocardiographic study.

OBJECTIVE To determine if nonsurvivors have a more abnormal pattern of left ventricular relaxation than survivors with severe sepsis. DESIGN Prospective, observational, cohort study. SETTING Intensive care unit in a university-affiliated tertiary care hospital. PATIENTS Twenty-four adults with severe sepsis. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Baseline clinical and hemodynamic variables, Acute Physiology and Chronic Health Evaluation (APACHE) II scores and Doppler echocardiographic mitral inflow pattern (analyzed for normalized peak early filling rate [E/VTI, systolic volumes/sec], deceleration time [msec], and early to atrial filling velocity ratio [E/A]). There were seven deaths. The patients did not differ in baseline demographics, inotropic infusions, hemodynamic measurements or ventilatory settings or variables. Nonsurvivors had a more abnormal pattern of left ventricular relaxation (E/VTI, 4.7 [range 3.8 to 5.8] vs. 5.8 [range 3.8 to 8.9], p= .04; deceleration time, 235 [range 209 to 367] vs. 182 [range 155 to 255], p = .002). E/A showed a nonsignificant trend in the same direction (0.9 [range 0.8 to 1.6] vs. 1.2 [range 0.7 to 1.9], p = .12). In a multivariate analysis, deceleration time (p< .004) and APACHE II score (p < .02) were the only independent predictors of mortality. CONCLUSION Severe sepsis nonsurvivors have a more abnormal echocardiographic pattern of left ventricular relaxation than survivors.

Automatic Detection of Patients with a High Risk of Systolic Cardiac Failure in Echocardiography

Heart disease is the global leading cause of death. A key predictor of heart failure and the most commonly measured cardiac parameter is left ventricular ejection fraction (LVEF). Despite available segmentation technologies, experienced cardiologists often rely on visual estimation of LVEF for a swift assessment. In this paper, we present a direct dual-channel LVEF estimation approach that mimics cardiologists’ visual assessment for detecting patients with high risk of systolic heart failure. The proposed framework consists of various layers for extracting spatial and temporal features from echocardiography (echo) cines. A data set of 1,186 apical two-chamber (A2C) and four-chamber (A4C) echo cines were used in this study. LVEF labels were assigned based on risk of heart failure: high-risk for \(\text {LVEF}\le 40\%\) and low-risk for \(40\%<\text {LVEF}\le 75\%\). We validated the proposed framework on 237 clinical exams and achieved a success rate of 83.1% for risk-based LVEF classification. Our experiments suggests the fusion of the two apical views improves the performance, compared to single-view networks, especially A2C. The proposed solution is promising for segmentation-free detection of high-risk LVEF. Direct LVEF estimation eliminates ventricle segmentation, and can hence be a useful tool for formal echo and point-of-care cardiac ultrasound.