Wireless Mobile Communication and Healthcare. Second International ICST Conference, MobiHealth 2011, Kos Island, Greece, October 5-7, 2011. Revised Selected Papers

Research Article

A Support Vector Machine Approach for Categorization of Patients Suffering from Chronic Diseases

Download
502 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-29734-2_36,
        author={Christos Bellos and Athanasios Papadopoulos and Roberto Rosso and Dimitrios Fotiadis},
        title={A Support Vector Machine Approach for Categorization of Patients Suffering from Chronic Diseases},
        proceedings={Wireless Mobile Communication and Healthcare. Second International ICST Conference, MobiHealth 2011, Kos Island, Greece, October 5-7, 2011. Revised Selected Papers},
        proceedings_a={MOBIHEALTH},
        year={2012},
        month={10},
        keywords={Personalized treatment Wearable Monitoring Management of Chronic Diseases Real-time classification},
        doi={10.1007/978-3-642-29734-2_36}
    }
    
  • Christos Bellos
    Athanasios Papadopoulos
    Roberto Rosso
    Dimitrios Fotiadis
    Year: 2012
    A Support Vector Machine Approach for Categorization of Patients Suffering from Chronic Diseases
    MOBIHEALTH
    Springer
    DOI: 10.1007/978-3-642-29734-2_36
Christos Bellos1,*, Athanasios Papadopoulos1,*, Roberto Rosso2,*, Dimitrios Fotiadis1,*
  • 1: Foundation for Research and Technology - Hellas
  • 2: TESAN Telematic & Biomedical Services S.p.A.
*Contact email: cbellos@cc.uoi.gr, thpapado@cc.uoi.gr, rosso@tesan.it, fotiadis@cs.uoi.gr

Abstract

The CHRONIOUS system is an open-architecture integrated platform aiming at the management of chronic disease patients. The system consists of a body sensor network collecting patient’s vital signals, a Personal Digital Assistance (PDA) for the real-time data analysis based on a Decision Support System (DSS) and a central system for the deeper analysis of patient’s status and data storing. The DSS combines several data sources to decide upon the severity of patient’s current health status. The first pilot study has been designed and carried out using patients suffering from Chronic Obstructive Pulmonary Disease(COPD). The DSS facilitates a one-against-all multi-class Support Vector Machine (SVM) classification system. The performance of the categorization scheme provides high classification results for most of the patient’s health status levels. The involvement of a larger number of patients might increase further the performance of the system.