A Novel Non-contact Infection Screening System Based on Self-Organizing Map with K-means Clustering

This paper aims to evaluate the efficacy of our non-contact infection screening system which uses Kohonen’s self-organizing map (SOM) with Kmeans clustering algorithm. In this study, the linear discriminant analysis (LDA) used in our previous system was replaced by SOM with K-means clustering algorithm to increase accuracy. The system simultaneously measures heart rate, respiratory rate, and facial skin temperature. The evaluation was done using the same data which we used in our previous study. The data was based on the test on 57 influenza patients and 35 normal control subjects at Japan Self-defense Forces Central Hospital. The system showed higher sensitivity of 98% and negative predictive value (NPV) of 96% compared to our previous system (sensitivity of 89%, NPV of 83%). The system can be used as a public health measure at points of entry where high sensitivity is most required in order to prevent the spread of the pandemic.

[1]  J. Desenclos,et al.  International travels and fever screening during epidemics: a literature review on the effectiveness and potential use of non-contact infrared thermometers. , 2009, Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin.

[2]  T Matsui,et al.  Development of a non-contact screening system for rapid medical inspection at a quarantine depot using a laser Doppler blood-flow meter, microwave radar and infrared thermography , 2009, Journal of medical engineering & technology.

[3]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

[4]  Takemi Matsui,et al.  A novel screening method for influenza patients using a newly developed non-contact screening system , 2010, Journal of Infection.

[5]  I. Lauder,et al.  Screening for fever by remote-sensing infrared thermographic camera. , 2006, Journal of travel medicine.

[6]  T. Tam,et al.  Border Screening for SARS , 2005, Emerging infectious diseases.

[7]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[8]  Esa Alhoniemi,et al.  Self-organizing map in Matlab: the SOM Toolbox , 1999 .

[9]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[10]  F Allerberger,et al.  Update of Clostridium difficile-associated disease due to PCR ribotype 027 in Europe. , 2007, Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin.

[11]  Masashi Sugiyama,et al.  Dimensionality Reduction of Multimodal Labeled Data by Local Fisher Discriminant Analysis , 2007, J. Mach. Learn. Res..

[12]  Abdul Ghaaliq Lalkhen,et al.  Clinical tests: sensitivity and specificity , 2008 .

[13]  Hiroshi Nishiura,et al.  Fever screening during the influenza (H1N1-2009) pandemic at Narita International Airport, Japan , 2011, BMC infectious diseases.

[14]  Wen-Cheng Chang,et al.  Limitations of Forehead Infrared Body Temperature Detection for Fever Screening for Severe Acute Respiratory Syndrome , 2004, Infection Control & Hospital Epidemiology.