Automatic detection of valve events by epicardial accelerometer allows estimation of the left ventricular pressure trace and pressure–displacement loop area

Figures

• 

  Figure 1. Pressure–displacement loops at (a) baseline and (b) after induction of ischemia, in one animal. The arrow indicates the loop direction from the end diastolic (ED) time-point.

• 

  Figure 2. ECG and hemodynamic recordings from a representative canine experiment are shown in the left panel with circumferential acceleration and calculated velocity and displacement. Dashed, vertical lines represent the reference points marking the opening and closing of the mitral and aortic valves. The right panel shows a schematic of the instrumentation. MVC mitral valve closure, AVO aortic valve opening, AVC aortic valve closure, MVO mitral valve opening.

• 

  Figure 3. A flow chart of the algorithm used to detect the different valve events from the acceleration signal.

• 

  Figure 4. Box plot showing the difference between the accelerometer-detected and hemodynamically defined valve events for the three interventions in the canine study. Grey boxes show the total pooled results for all the three interventions and the dotted line represents the median of all the pooled data (2 ms). LBBB left bundle branch block, MVC mitral valve closure, AVO aortic valve opening, AVC aortic valve closure, MVO mitral valve opening.

• 

  Figure 5. (a) Individual left ventricular (LV) pressure traces used to calculate the normalized, averaged pressure trace. Mitral and aortic valve events are indicated. (b) Traces from (a) adjusted along the time and pressure axes. The resulting normalized, averaged LV pressure trace shown with thick black line. (c) Comparison between measured and estimated LV pressure from a representative example. (d) Correlation between the estimated and measured trace shown in (c). (e) Difference between measured and estimated LV pressure over time for all recordings. LVP left ventricular pressure, MVC mitral valve closure, AVO aortic valve opening, AVC aortic valve closure, MVO mitral valve opening.

• 

  Figure 6. Comparison of pressure–displacement loops at baseline and during ischemia in a representative animal using estimated (blue) and measured pressure traces (red). The “bulge” in the lower left corner of the red loop during ischemia is presumably due to a measurement artifact caused by the pressure catheter coming in contact with the ventricular wall.

• 

  Figure 7. (a) Correlation and (b) Bland–Altman plot for pressure–displacement loop areas using measured vs. estimated pressure. Area is shown in mm·mmHg.

• 

  Figure 8. Comparison of the correlation coefficient (r-value) of the estimated pressure (a), and the error in estimated pressure–displacement loop area (b), between open (n = 103) and closed chest (n = 27) animal recordings.