Given the current generational trends, stress is becoming an increasingly important component of people's life nowadays. Nobody on the planet is immune to the negative consequences of stress and despair. We proposed in this study that the electrocardiogram can be utilised to predict and detect differences in human heart rate, rhythm, and electrical activity caused by psychological issues using a machine learning approach. Additionally, particular ECG patterns are utilised to indicate mental health issues such as panic disorder, bipolar disease, and weaknesses. Electrocardiogram (ECG) data records the heart’s electrical activity and is usually used to diagnose and treat a variety of heart problems as well as to detect psychological illnesses. Mental stress has a significant impact on the human body, particularly the cardiovascular system. In this paper we have used the Hidden Markov Model (HMM) to train ECG dataset. The dataset was compiled from multiple health care facilities and hospitals, and it included 876 people of various ages as well as an ECG pattern report. We employed HMM classifiers to classify the accuracy, precision, and recall properties. The accuracy of HMM is (91.75%).