In this article, we present an energy-efficient real-time electrocardiogram (ECG) acquisition system that utilizes near-data processing for prediagnosis and controls the frequency of wireless communication to reduce power consumption. The system consists of a wearable ECG monitoring device, an Android-based smartphone application, and a database server, which mitigate the inconvenience of the periodical visit to the hospital. To maximize the battery lifetime, the prototype wearable device adopted algorithms with near-data processing to enable prediagnosis, which was verified using the MIT-BIH arrhythmia database. Also, the real-time acquisition is available by the data transmission between the wearable device and the smartphone utilizing Bluetooth low energy (BLE) when the device starts collecting the patient’s ECG data. The wearable device cannot only acquire the raw ECG waveform data but also provide an energy-saving option to send only feature data, which is processed within the device for near-data prediagnosis. The overall power consumption of the device was reduced because the power consumption by communication was greatly reduced. Measured results verified that the feature data transmission with near-data processing can increase the battery lifetime by 15.9% compared to the conventional raw ECG data transmission.