Telemedicine technology is increasingly needed for remote patient health care. Especially for patients with a high risk/severe disease such as heart disease. Due to its important role and benefits, a lot of research has been carried out to improve this technology so that it's relevant with current state-of-the-art technologies. One of the latest technologies that have been adapted on telemedicine research is cloud computing. With this combination, CPU intensive task can be transferred to the cloud. However, a new problems arose. The resulting system now will always depend on the internet, while not all areas are covered or have a good internet connection. Therefore, this researcher proposes a QRS detection system with parallel processing on the edge device using the Pan-Tompkins algorithm and Renderscript framework. This system is an Android-based application that receives ECG data and then performs the QRS detection process by utilizing Renderscript to perform parallel processing. Execution time on the detection process will be measured to determine the difference in performance with serial processing. The results of the detection process will be compared to determine the level of accuracy of the system. A profiler application will also be used to find out the CPU Utilization level. In testing, the system obtained 3 times smaller execution time with an average processing speed of 301,340 row/second. The system also obtained a sensitivity value of 99.64% and a positive predictivity value of 99.47%. The average CPU utilization on the system is at 82%, 32% higher than serial counterpart.