The electrocardiogram (ECG) is characterized by a recurrent wave sequence of P, QRS and T wave associated with each beat. This paper presents a new method for the detection of P and T waves in 12-lead ECG using the vector support machine (SVM). Digital filtering techniques are used to eliminate interference from the power line and baseline wandering present in the ECG signal. SVM is used as a classifier for the detection of P and T-waves. The performance of the algorithm is evaluated using 50 original 12-lead ECG records recorded simultaneously from the CSE's standard ECG database. A significant detection rate is achieved. The method successfully detects monophasic and biphasic waves. The electrocardiogram (ECG) is a very important tool to know the functional status of the heart. The ECG pattern consists of a recurrent wave sequence of P, QRS and T wave associated with each beat. Automatic ECG wave detection is important for the diagnosis of heart disease. In a clinical setting, such as intensive care units, it is essential that automated systems accurately detect and classify ECG waveform components. The correct functioning of these systems depends on several important factors, including the ECG signal quality, the applied classification rule, the set of learning data and test used.