Related Experiment Videos
Quantification of left ventricular function with contrast-enhanced harmonic colour Doppler and a semiautomated
Aims:
Patients with poor quality echocardiograms impede the application of available automatic boundary detection technologies. Tissue harmonic imaging and contrast media can allow optimal differentiation of left ventricular blood pool and tissue, making it possible to utilize automatic boundary detection software for automatic non-operator-dependent computation of left ventricular volumes. We integrated contrast-enhanced harmonic colour Doppler with a semiautomated boundary detection algorithm to explore the feasibility, accuracy and inter-observer variability for left ventricular volume assessment in technically difficult patients.
Methods And Results:
Twenty-six patients with more than two segments not clearly visualized in tissue harmonic imaging were studied with contrast-enhanced harmonic colour Doppler using Levovist. Twenty patients (77%) achieved full left ventricular contrast filling without apparent blooming artefacts. Contrast-enhanced harmonic colour Doppler-automatic boundary detection was successfully implemented in these 20 patients, despite three (15%) in which it was not possible to acquire more than three cardiac cycles' values. Contrast-enhanced harmonic colour Doppler-automatic boundary detection measurements agreed closely with the manually drawn data. Among the three independent readers in the three techniques, the best correlation, lowest SEE, smallest limits of agreement and inter-observer variability were obtained in contrast-enhanced harmonic colour Doppler-automatic boundary detection.
Conclusion:
Contrast-enhanced harmonic colour Doppler-automatic boundary detection was feasible and accurate in estimation of left ventricular volume and function in patients with poor acoustic windows. This technique significantly reduced inter-observer variability, thus improving reliability and confidence of investigators in left ventricular function assessment. Contrast-enhanced harmonic colour Doppler-automatic boundary detection may have great potential in clinical evaluation of left ventricular volume and function, especially when on-line software is available.