A Python program was developed for the automated analysis of digitized electrocardiograms (ECGs) with the objective of classifying heart rate, determining the electrical axis, and identifying first-degree atrioventricular block (1AVB) by automatically measuring time intervals and wave amplitudes, thus emulating cardiological diagnostic criteria. The algorithm was evaluated using a sample of 85 ECG recordings sourced from A large scale 12-lead electrocardiogram database for arrhythmia study published on Physionet (115 ECG), all known to exhibit 1AVB. The tool demonstrated an effectiveness of 97% in 1AVB detection, identifying 83 cases. Additionally, the program was able to produce the following measurements of ECGs: Electrical Axis, classifiying them as 48 Normal, 35 Left Axis Deviation, and 2 Right Axis Deviation, and Heart Rate, detecting 40 Bradycardias (38 Mild, 1 Moderate, 1 Severe), 15 Tachycardias, and 28 Normal Rates.

The preliminary risk analysis obtained with the measurements provided by the tool yielded 15 high-risk cases and 70 low-risk cases, with diagnostic implications for conditions such as ventricular hypertrophy or advanced cardiac conduction system disease. The processing time for the whole sample was an average of 163 seconds using a standard computer, resulting in an analysis time per ECG of less than 2 seconds. The computational results showed full concordance with the clinical diagnosis provided by cardiologists from the Desiderio Rosado Carbajal Hospital. The high accuracy, efficient processing time, and concordance with expert diagnosis confirm the potential of this computational program as a reliable and rapid support tool in the field of cardiology, facilitating the screening and preliminary diagnosis of significant ECG abnormalities.