Introduction: COVID-19 has become a global pandemic that even the World Health Organization declared in February 2020. Artificial Intelligence (AI) based techniques such as Machine learning (ML) and Deep Learning (DL) techniques may play an essential role in determining patient populations who have COVID-19. The machine learning technique allows the users to imitate artificial intellect as well as to absorb massive amounts of data to detect trends and observations quickly. Deep learning constructs networks across layers that build an artificial neural network intelligent enough to understand and making rational decisions concerning COVID-19 diagnosis. Objective: The main objective of the research is to accurately detect as well as diagnose the covid-19 patients based on chest images.Covid-19 pandemic is spreading rapidly and the testing kits, labs, hospitals, health workers as well as diagnosis are limited. Methods: The DL and ML approach helps in explaining the COVID-19 pandemic and resolving it. This research mainly utilizes the chest X-ray data set (collected from the USA) for the prediction of Covid-19 patients (confirm, recover, death). Results: The experimental results are showing the accuracy result of CNN method 92.4 % and Random forest method 87.9 %. Conclusion: Based on the experimental results we can say the CNN (deep learning) method performs outstanding over Ran dom forest (machine learning) in terms of detection rate % and accuracy %.