Portable ECG Monitor on Edge Device using TensorFlow Lite

초록

In recent years, the healthcare and wellness industry has grown dramatically due to the increased focus on health. Early detection of cardiovascular disease through electrocardiogram monitoring is becoming increasingly important. Recent advances in artificial intelligence have made great strides in the early detection of cardiovascular disease through electrocardiogram. This paper presents a method to monitor electrocardiogram in real time by implementing artificial intelligence technology on edge device using IoT technology, unlike the existing server-based artificial intelligence technology. Since the edge device used in this paper has limited resources, the TensorFlow Lite that is specially designed to support real-time machine learning (ML) inference on low-power and resource-constrained edge devices was used. The performance of the pre-trained model and the lightweight model was compared through TensorFlow Lite to confirm whether the application was suitable.