43
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      A Telesurveillance System With Automatic Electrocardiogram Interpretation Based on Support Vector Machine and Rule-Based Processing

      research-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Background

          Telehealth care is a global trend affecting clinical practice around the world. To mitigate the workload of health professionals and provide ubiquitous health care, a comprehensive surveillance system with value-added services based on information technologies must be established.

          Objective

          We conducted this study to describe our proposed telesurveillance system designed for monitoring and classifying electrocardiogram (ECG) signals and to evaluate the performance of ECG classification.

          Methods

          We established a telesurveillance system with an automatic ECG interpretation mechanism. The system included: (1) automatic ECG signal transmission via telecommunication, (2) ECG signal processing, including noise elimination, peak estimation, and feature extraction, (3) automatic ECG interpretation based on the support vector machine (SVM) classifier and rule-based processing, and (4) display of ECG signals and their analyzed results. We analyzed 213,420 ECG signals that were diagnosed by cardiologists as the gold standard to verify the classification performance.

          Results

          In the clinical ECG database from the Telehealth Center of the National Taiwan University Hospital (NTUH), the experimental results showed that the ECG classifier yielded a specificity value of 96.66% for normal rhythm detection, a sensitivity value of 98.50% for disease recognition, and an accuracy value of 81.17% for noise detection. For the detection performance of specific diseases, the recognition model mainly generated sensitivity values of 92.70% for atrial fibrillation, 89.10% for pacemaker rhythm, 88.60% for atrial premature contraction, 72.98% for T-wave inversion, 62.21% for atrial flutter, and 62.57% for first-degree atrioventricular block.

          Conclusions

          Through connected telehealth care devices, the telesurveillance system, and the automatic ECG interpretation system, this mechanism was intentionally designed for continuous decision-making support and is reliable enough to reduce the need for face-to-face diagnosis. With this value-added service, the system could widely assist physicians and other health professionals with decision making in clinical practice. The system will be very helpful for the patient who suffers from cardiac disease, but for whom it is inconvenient to go to the hospital very often.

          Related collections

          Most cited references48

          • Record: found
          • Abstract: found
          • Article: not found

          Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: population-based estimates.

          Atrial fibrillation (AF) is the most common of the serious cardiac rhythm disturbances and is responsible for substantial morbidity and mortality in the general population. Its prevalence doubles with each advancing decade of age, from 0.5% at age 50-59 years to almost 9% at age 80-89 years. It is also becoming more prevalent, increasing in men aged 65-84 years from 3.2% in 1968-1970 to 9.1% in 1987-1989. This statistically significant increase in men was not explained by an increase in age, valve disease, or myocardial infarctions in the cohort. The incidence of new onset of AF also doubled with each decade of age, independent of the increasing prevalence of known predisposing conditions. Based on 38-year follow-up data from the Framingham Study, men had a 1.5-fold greater risk of developing AF than women after adjustment for age and predisposing conditions. Of the cardiovascular risk factors, only hypertension and diabetes were significant independent predictors of AF, adjusting for age and other predisposing conditions. Cigarette smoking was a significant risk factor in women adjusting only for age (OR = 1.4), but was just short of significance on adjustment for other risk factors. Neither obesity nor alcohol intake was associated with AF incidence in either sex. For men and women, respectively, diabetes conferred a 1.4- and 1.6-fold risk, and hypertension a 1.5- and 1.4-fold risk, after adjusting for other associated conditions. Because of its high prevalence in the population, hypertension was responsible for more AF in the population (14%) than any other risk factor. Intrinsic overt cardiac conditions imposed a substantially higher risk. Adjusting for other relevant conditions, heart failure was associated with a 4.5- and 5.9-fold risk, and valvular heart disease a 1.8- and 3.4-fold risk for AF in men and women, respectively. Myocardial infarction significantly increased the risk factor-adjusted likelihood of AF by 40% in men only. Echocardiographic predictors of nonrheumatic AF include left atrial enlargement (39%/ increase in risk per 5-mm increment), left ventricular fractional shortening (34% per 5% decrement), and left ventricular wall thickness (28% per 4-mm increment). These echocardiographic features offer prognostic information for AF beyond the traditional clinical risk factors. Electrocardiographic left ventricular hypertrophy increased risk of AF 3-4-fold after adjusting only for age, but this risk ratio is decreased to 1.4 after adjustment for the other associated conditions. The chief hazard of AF is stroke, the risk of which is increased 4-5-fold. Because of its high prevalence in advanced age, AF assumes great importance as a risk factor for stroke and by the ninth decade becomes a dominant factor. The attributable risk for stroke associated with AF increases steeply from 1.5% at age 50-59 years to 23.5% at age 80-89 years. AF is associated with a doubling of mortality in both sexes, which is decreased to 1.5-1.9-fold after adjusting for associated cardiovascular conditions. Decreased survival associated with AF occurs across a wide range of ages.
            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            Telemedicine and remote management of patients with heart failure.

            Advances in telecommunication technologies have created new opportunities to provide telemedical care as an adjunct to medical management of patients with heart failure. Meta-analyses suggest that telemedicine can reduce morbidity and mortality in such patients; however, two prospective clinical trials not included in the analyses do not support these findings. Therefore, the effectiveness of telemedicine in heart failure is not established. Telemedicine approaches range from computer-based support systems to programmes led by nurses and physicians. Standardisation and appropriate classification of telemedical systems are needed to enable accurate interpretation of clinical trials. Here we propose a classification of four generations of telemedicine in heart failure. Not all approaches are the same and not every patient with heart failure will need telemedicine. Crisis prevention and treatment, and stabilisation and self-empowerment of patients are focuses of telemedicine in heart failure. The profile of patients who can potentially benefit from telemedicine is unknown and should be investigated in adequately powered randomised clinical trials. We are optimistic that telemedicine is an efficient approach and will become an important feature of management in heart failure. Copyright © 2011 Elsevier Ltd. All rights reserved.
              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Structured telephone support or telemonitoring programmes for patients with chronic heart failure.

              Specialised disease management programmes for chronic heart failure (CHF) improve survival, quality of life and reduce healthcare utilisation. The overall efficacy of structured telephone support or telemonitoring as an individual component of a CHF disease management strategy remains inconclusive. To review randomised controlled trials (RCTs) of structured telephone support or telemonitoring compared to standard practice for patients with CHF in order to quantify the effects of these interventions over and above usual care for these patients. Databases (the Cochrane Central Register of Controlled Trials (CENTRAL), Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment Database (HTA) on The Cochrane Library, MEDLINE, EMBASE, CINAHL, AMED and Science Citation Index Expanded and Conference Citation Index on ISI Web of Knowledge) and various search engines were searched from 2006 to November 2008 to update a previously published non-Cochrane review. Bibliographies of relevant studies and systematic reviews and abstract conference proceedings were handsearched. No language limits were applied. Only peer reviewed, published RCTs comparing structured telephone support or telemonitoring to usual care of CHF patients were included. Unpublished abstract data was included in sensitivity analyses. The intervention or usual care could not include a home visit or more than the usual (four to six weeks) clinic follow-up. Data were presented as risk ratio (RR) with 95% confidence intervals (CI). Primary outcomes included all-cause mortality, all-cause and CHF-related hospitalisations which were meta-analysed using fixed effects models. Other outcomes included length of stay, quality of life, acceptability and cost and these were described and tabulated. Twenty-five studies and five published abstracts were included. Of the 25 full peer-reviewed studies meta-analysed, 16 evaluated structured telephone support (5613 participants), 11 evaluated telemonitoring (2710 participants), and two tested both interventions (included in counts). Telemonitoring reduced all-cause mortality (RR 0.66, 95% CI 0.54 to 0.81, P < 0.0001) with structured telephone support demonstrating a non-significant positive effect (RR 0.88, 95% CI 0.76 to 1.01, P = 0.08). Both structured telephone support (RR 0.77, 95% CI 0.68 to 0.87, P < 0.0001) and telemonitoring (RR 0.79, 95% CI 0.67 to 0.94, P = 0.008) reduced CHF-related hospitalisations. For both interventions, several studies improved quality of life, reduced healthcare costs and were acceptable to patients. Improvements in prescribing, patient knowledge and self-care, and New York Heart Association (NYHA) functional class were observed. Structured telephone support and telemonitoring are effective in reducing the risk of all-cause mortality and CHF-related hospitalisations in patients with CHF; they improve quality of life, reduce costs, and evidence-based prescribing.
                Bookmark

                Author and article information

                Contributors
                Journal
                JMIR Med Inform
                JMIR Med Inform
                JMI
                JMIR Medical Informatics
                Gunther Eysenbach (JMIR Publications Inc., Toronto, Canada )
                2291-9694
                Apr-Jun 2015
                07 May 2015
                : 3
                : 2
                : e21
                Affiliations
                [1] 1National Taiwan University Graduate Institute of Biomedical Electronics and Bioinformatics TaipeiTaiwan
                [2] 2National Taiwan University Graduate Institute of Communication Engineering TaipeiTaiwan
                [3] 3National Taiwan University Department of Electrical Engineering TaipeiTaiwan
                [4] 4National Taiwan University Hospital Telehealth Center TaipeiTaiwan
                Author notes
                Corresponding Author: Jian-Jiun Ding jjding@ 123456ntu.edu.tw
                Author information
                http://orcid.org/0000-0002-4371-6936
                http://orcid.org/0000-0001-9583-7476
                http://orcid.org/0000-0002-3145-0798
                http://orcid.org/0000-0001-7147-8122
                http://orcid.org/0000-0003-4510-2273
                http://orcid.org/0000-0002-8936-9570
                http://orcid.org/0000-0002-0158-7232
                Article
                v3i2e21
                10.2196/medinform.4397
                4440896
                25953306
                778b6111-9f63-44a6-93ef-f31df7c173b2
                ©Te-Wei Ho, Chen-Wei Huang, Ching-Miao Lin, Feipei Lai, Jian-Jiun Ding, Yi-Lwun Ho, Chi-Sheng Hung. Originally published in JMIR Medical Informatics (http://medinform.jmir.org), 07.05.2015.

                This is an open-access article distributed under the terms of the Creative Commons Attribution License ( http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Medical Informatics, is properly cited. The complete bibliographic information, a link to the original publication on http://medinform.jmir.org/, as well as this copyright and license information must be included.

                History
                : 04 March 2015
                : 19 March 2015
                : 03 April 2015
                Categories
                Original Paper
                Original Paper

                telehealth care,telesurveillance system,electrocardiogram,ecg classification,support vector machine

                Comments

                Comment on this article