A novel compression framework using energy-sensitive QRS complex detection method for a mobile ECG

Dear editor, Novel wearable applications provide improved data compression for reduced power consumption [1, 2]; however, real-time monitoring of a single source electrocardiogram (ECG) signal leads to extended data usage of 2.77 GB per day. The Q wave, R wave, and S wave (QRS) complex seen on an ECG is the basis for the automatic determination of heart rate and an entry point for the classification schemes of the cardiac cycle [3]. Therefore, it is necessary that the compressed data should retain maximum QRS area information, which is the origin of the concept of areas of interests in compressed sensing [4]. Currently, most researches concentrate on developing methods for efficient extraction of QRS waves without redundant calculations from the complex and noisy ECG signals and compression frameworks. This study aims to propose a novel framework that includes an energy-sensitive QRS complex detection algorithm based on simplified empirical mode decomposition and Hilbert transform (EMD-HT) method and a multi-compression ratio CS strategy. The proposed framework encompasses three advantages: (a) In comparison with a previous study [4], the proposed method uses percentage root-mean-square difference (PRD) and improved reduction quality under the same compression ratio (CR); (b) it can accurately locate the interested area of the QRS cluster, which solves the interference problem of stationary noise and; (c) it is indicated that EMD-based compression results in a better CR and PRD than the other methods [2]. Considering the specific conditions for the wearable devices, we employ a simplified EMD algorithm whose operation for detecting interested area for ECG reconstruction is characterized by sufficient accuracy. Using the EMD-HT method, the proposed framework can overcome the limitations associated with stationary noise interference and thus, can achieve precise positioning.