This paper deals with the PNN based detection of QRS-complexes in Electrocardiogram (ECG) using entropy criteria. Recorded Raw ECG signal consist of baseline wander and power line interference. This is known as noise and that can be separated by using digital filtering techniques and this signal is known as filtered signal of ECG. In this entropy criteria is used generate the feature signal and this feature signal is applied to pattern classifier. The QRS-complexes are detected using Probabilistic Neural Networks as pattern classifier. The proposed algorithm is implemented using MATLAB. The performance evaluation of the algorithm is validated using each lead of the 12- lead simultaneously recorded ECGs from the dataset-3 of the CSE multi-lead measurement library. The detection rate of QRS-complexes is 99.34% by using proposed algorithm. The percentage of false negative is 0.66% and false positive is 0.83%. The results obtained by the proposed algorithm in terms of performance, based on the detection rate of QRS-detection that is compared with the other methods as reported in the literature. The algorithm that is proposed in this paper, demonstrate the strength for the QRS-detection field.