Electrocardiography (ECG) consists of pivotal points named P-Q-R-S-T, and accurately identifying these points is crucial for diagnosing heart conditions. In detecting ST elevation myocardial infarction and non-ST elevation myocardial infarction, the segment between the S and T peaks is evaluated, for this reason it’s important to detect those peaks. This study proposes an automatic method based on derivatives for the high-accuracy detection of the S and T peaks. The suggested method was tested on ECG signals recorded from a total of 100 individuals, comprising 51 males and 49 females. With the proposed method, the S and T peaks were detected with an average success rate of $\mathbf{9 7 . 6 5} \%$, achieving $\mathbf{1 2 . 5 \%}$ greater accuracy than the most recent study presented in the literature. This method is anticipated to contribute to the detection of heart conditions such as myocardial infarction, which rely on the examination of the ST segment.