TY - JOUR
T1 - Fiberoptic Evanescent Wave Spectroscopy (FEWS) for blood diagnosis
T2 - Proceedings of the 1999 Specialty Fiber Optics for Medical Applications
AU - Eytan, Osnat
AU - Bormashenko, Edward
AU - Pogreb, Roman
AU - Sutovski, Semion
AU - Vasserman, Irena
AU - Sela, Ben Ami
AU - Katzir, Abraham
PY - 1999
Y1 - 1999
N2 - Infrared spectral analysis of human blood serum was carried out using FTIR - FEWS (Fourier Transform Infrared Fiberoptic Evanescent Wave Spectroscopy). The fiberoptic sensor elements were short (10.5 cm) lengths of unclad AgClBr fibers of diameter 0.9 mm. In order to protect the fibers from interaction with the blood serum we developed a dip coating technique, which made it possible to control the coating thickness. Three polymer materials were tested: polysulfone, polystyrene and room temperature vulcanized silicone elastomer. It was found that layers of thickness 2-3 μm provide chemical protection, while making it possible to carry out FEWS measurements. The measured spectra were analyzed by neural network analysis, to predict the concentrations of urea, uric acid, cholesterol, total protein and cretonne. The predicted concentrations were compared with the results obtained using standard chemical analysis of the blood (SMAC) and good correlation was observed, with errors of only few percent.
AB - Infrared spectral analysis of human blood serum was carried out using FTIR - FEWS (Fourier Transform Infrared Fiberoptic Evanescent Wave Spectroscopy). The fiberoptic sensor elements were short (10.5 cm) lengths of unclad AgClBr fibers of diameter 0.9 mm. In order to protect the fibers from interaction with the blood serum we developed a dip coating technique, which made it possible to control the coating thickness. Three polymer materials were tested: polysulfone, polystyrene and room temperature vulcanized silicone elastomer. It was found that layers of thickness 2-3 μm provide chemical protection, while making it possible to carry out FEWS measurements. The measured spectra were analyzed by neural network analysis, to predict the concentrations of urea, uric acid, cholesterol, total protein and cretonne. The predicted concentrations were compared with the results obtained using standard chemical analysis of the blood (SMAC) and good correlation was observed, with errors of only few percent.
UR - http://www.scopus.com/inward/record.url?scp=0032689297&partnerID=8YFLogxK
U2 - 10.1117/12.346718
DO - 10.1117/12.346718
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.conferencearticle???
AN - SCOPUS:0032689297
SN - 0277-786X
VL - 3596
SP - 74
EP - 81
JO - Proceedings of SPIE - The International Society for Optical Engineering
JF - Proceedings of SPIE - The International Society for Optical Engineering
Y2 - 24 January 1999 through 25 January 1999
ER -