Automatic detection of adult cardiomyocyte for high throughput measurements of calcium and contractility.

Abstract

Simultaneous calcium and contractility measurements on isolated adult cardiomyocytes have been the gold standard for the last decades to study cardiac (patho)physiology. However, the throughput of this system is low which limits the number of compounds that can be tested per animal. We developed instrumentation and software that can automatically find adult cardiomyocytes. Cells are detected based on the cell boundary using a Sobel-filter to find the edge information in the field of view. Separately, we detected motion by calculating the variance of intensity for each pixel in the frame through time. Additionally, it detects the best region for calcium and contractility measurements. A sensitivity of 0.66 ± 0.08 and a precision of 0.82 ± 0.03 was reached using our cell finding algorithm. The percentage of cells that were found and had good contractility measurements was 90 ± 10%. In addition, the average time between 2 cardiomyocyte calcium and contractility measurements decreased from 93.5 ± 80.2 to 15.6 ± 8.0 seconds using our software and microscope. This drastically increases throughput and provides a higher statistical reliability when performing adult cardiomyocyte functional experiments.