Cardiac safety pharmacology requires in-vitro testing of all drug candidates before clinical trials in order to ensure they are screened for cardio-toxic effects which may result in severe arrhythmias. Micro-electrode arrays (MEA) serve as a complement to current in-vitro methods for drug safety testing. However, MEA recordings produce huge volumes of data and manual analysis forms a bottleneck for high-throughput screening. To overcome this issue, we have developed an offline, semi-automatic data analysis software, 'Cardiomyocyte MEA Data Analysis (CardioMDA)', equipped with correlation analysis and ensemble averaging techniques to improve the accuracy, reliability and throughput rate of analysing human pluripotent stem cell derived cardiomyocyte (CM) field potentials. With the program, true field potential and arrhythmogenic complexes can be distinguished from one another. The averaged field potential complexes, analysed using our software to determine the field potential duration, were compared with the analogous values obtained from manual analysis. The reliability of the correlation analysis algorithm, evaluated using various arrhythmogenic and morphology changing signals, revealed a mean sensitivity and specificity of 99.27% and 94.49% respectively, in determining true field potential complexes. The field potential duration of the averaged waveforms corresponded well to the manually analysed data, thus demonstrating the reliability of the software. The software has also the capability to create overlay plots for signals recorded under different drug concentrations in order to visualize and compare the magnitude of response on different ion channels as a result of drug treatment. Our novel field potential analysis platform will facilitate the analysis of CM MEA signals in semi-automated way and provide a reliable means of efficient and swift analysis for cardiomyocyte drug or disease model studies.