In-hospital cardiac arrest (I-HCA) is a significant public health problem because it is associated with high mortality and low survival. Electrocardiography (ECG) is an important diagnostic tool to prevent and treat I-HCA. ECG recordings may provide clues for the presence of complications and show life threatening heart rhythms that require resuscitation therapy. The ECG from the continuous monitoring systems at hospitals cannot be saved digitally. Therefore, establishing ECG predictors by analyzing different segments of ECG prior to I-HCA requires manual measurements which are time consuming and difficult. The objective of this project was to develop a method for semi-automated analysis of paper ECG tracings using digital image processing, in order to correlate changes in ECG parameters with I-HCA.