A Remote Diagnosis Service Platform for Wearable ECG Monitors

By Jun Dong, Jia-wei Zhang, Hong-hai Zhu, Li-ping Wang, Xia Liu, Zhen-jiang Li

NOTE: This is an overview of the entire article, which appeared in the November/December 2012 issue of the IEEE Intelligent Systems magazine.
Click here to read the entire article.

To lessen increased pressures on hospitals and improve service quality for outpatients, China aims to provide a Basic Community Medical Insurance System (BCMIS) before 2020. By reducing face-to-face consultations and shortening hospitalization, the BCMIS could minimize strain on healthcare resources, while maintaining and improving service quality.

Cardiovascular disease is among the major subjects the BCMIS is addressing. hospitals. The ability to continuously monitor the heart health status of patients who suffer from cardiovascular problems is vital. To meet the need for large-scale monitoring of patients with cardiac arrythmias, for example, the authors describe a patient location- independent and continuous ECG monitoring and diagnosis system they are developing. Their ECG recognition method is based on signal morphology and physicians’ knowledge incorporated in the Chinese Cardiovascular Diseases Database, an open database which they have developed. (The article provides a link to this database).

Figure 1. Platform architecture. The electrocardiogram (ECG) online service provides outpatients with remote ECG monitoring services. The knowledge base contains the Chinese Cardiovascular Diseases (CCD) database and experts' experiences.

Figure 1. Platform architecture. The electrocardiogram (ECG) online service provides outpatients with remote ECG monitoring services. The knowledge base contains the Chinese Cardiovascular Diseases (CCD) database and experts’ experiences.

The platform has three main aspects: an ECG online service (ECG-OS), a knowledge base (KB), and a computer-aided diagnosis approach (CADA). The ECG-OS provides the remote ECG checking services to outpatients through which the patient and physician are connected. CADA provides a stable classification model after the model is evaluated in the CCDD.

The authors state that the primary difference between the CCDD and other databases is that its data content is dynamic – that is, the data is enhanced continuously on the basis of feedback from other parts of the platform, thus providing better support for physicians and better service for patients who are far away.

Figure 2. The remote diagnosis service system. Its basic components are a belted detector, a smartphone, a data server, a diagnosis terminal, and short- and long- distance wireless modules.

Figure 2. The remote diagnosis service system. Its basic components are a belted detector, a smartphone, a data server, a diagnosis terminal, and short- and long- distance wireless modules.

The basic components of the ECG-OS are a belted detector, a smartphone, a data server, a diagnosis terminal, and short- and long-distance wireless modules. Figure 2 illustrates this system. The ECG algorithm in the ECG-OS analyzes the ECG (especially a QRS complex) and signals an alarm when needed. The algorithm uses the altitude, slope, and interval parameters of adjacent QRS complexes.

The article discusses in detail the features of the ECG signal that the algorithm utilizes. The approach taken in this project aims to distinguish between normal and abnormal ECGs automatically and let physicians diagnose only the abnormal ECG data.

As part of this project, the detection algorithms were tested extensively against existing databases, for three quality measures, sensitivity, selectivity, and general correctness rate. They report values of over 98.5%, for instance, in general correctness rate, which consists of the sum of True Positive and True Negative results as a percent of all results obtained.

Finally, the authors report on plans for further investigation, such as incorporating other physiological signals (e.g., blood pressure) in the detector.

ABOUT THE AUTHORS

Jun Dong (jdong2010@sinano.ac.cn) is a professor in the interdisciplinary division at the Suzhou Institute of Nano- Tech and Nano-Bionics, Chinese Academy of Sciences. His research focuses on intelligence simulation, thinking patterns, and cognition modeling, including computer-aided medicine diagnosis methods, relative wearable terminals, and art creation simulation. Dong has a PhD in computer simulation from Zhejiang University, China.

Jia-wei Zhang (zwei.jiawei@gmail.com) is a CT R&D software engineer at Siemens Shanghai Medical Equipment. His research focuses on ECG features recognition. Zhang has a PhD in computer application from East China Normal University.

Hong-hai Zhu (hhzhu2010@sinano.ac.cn) is a PhD candidate in the interdisciplinary division at the Suzhou Institute of Nano-Tech and Nano-Bionics, Chinese Academy of Sciences. His research interests include wireless medical terminal and pattern recognition. Zhu has an MS in software engineering from East China Normal University.

Li-ping Wang (lipingwang@sei.ecnu.edu.cn) is a lecturer and PhD candidate in the school of software engineering at East China Normal University. Her research focuses on ECG diseases classification. Wang has an MS in computer software and theory from Northwest University, China.

Xia Liu (liuxia9110@yahoo.com.cn) is Xia Liu is chief physician at Shanghai Ruijin Hospital of Shanghai Jiaotong University, China. Her research interests include ECG diagnosis and cardiovascular diseases analysis. Liu has an MS in medicine from Shanghai Jiaotong University.

Zhen-jiang Li (zhenjiang.li@ia.ac.cn) is an associate professor at the State Key Laboratory of Management and Control for Complex Systems, Chinese Academy of Sciences. His research focuses on embedded systems. Li has a PhD in pattern recognition and control engineering from the Institute of Automation, Chinese Academy of Sciences.