Blood diagnostics using fiberoptic evanescent wave spectroscopy and neural networks analysis

Yaron Gotshal; Ronit Simhi; Ben Ami Sela; Abraham Katzir

doi:10.1016/S0925-4005(97)80331-0

Blood diagnostics using fiberoptic evanescent wave spectroscopy and neural networks analysis

Yaron Gotshal, Ronit Simhi, Ben Ami Sela, Abraham Katzir

School of Physics and Astronomy

Research output: Contribution to journal › Article › peer-review

Abstract

A spectral analysis of human blood serum was undertaken by fiberoptic evanescent wave spectroscopy (FEWS) using a Fourier transform infrared (FTIR) spectrometer. The blood serum samples were introduced into a special cell designed for FEWS of liquids, with an IR transmitting silver halide fiber as the sensing element. The spectra were analyzed by models of neural networks (NN), and the concentrations of protein, cholesterol and uric acid in human blood serum were obtained. Our results are in agreement with those obtained by ordinary chemical enzymatic methods and multivariate calibration methods. The estimated prediction errors obtained (in percent of the average value) were 4.7% for total protein, 22% for cholesterol and 35% for uric acid. This method can be used for in-situ real-time blood analysis.

Original language	English
Pages (from-to)	157-161
Number of pages	5
Journal	Sensors and Actuators, B: Chemical
Volume	42
Issue number	3 B
DOIs	https://doi.org/10.1016/S0925-4005(97)80331-0
State	Published - Aug 1997

Keywords

Blood analysis
FEWS
Neural network
Silver halide

Access to Document

10.1016/S0925-4005(97)80331-0

Cite this

@article{c852c06186bf4591936972c031003fee,

title = "Blood diagnostics using fiberoptic evanescent wave spectroscopy and neural networks analysis",

abstract = "A spectral analysis of human blood serum was undertaken by fiberoptic evanescent wave spectroscopy (FEWS) using a Fourier transform infrared (FTIR) spectrometer. The blood serum samples were introduced into a special cell designed for FEWS of liquids, with an IR transmitting silver halide fiber as the sensing element. The spectra were analyzed by models of neural networks (NN), and the concentrations of protein, cholesterol and uric acid in human blood serum were obtained. Our results are in agreement with those obtained by ordinary chemical enzymatic methods and multivariate calibration methods. The estimated prediction errors obtained (in percent of the average value) were 4.7% for total protein, 22% for cholesterol and 35% for uric acid. This method can be used for in-situ real-time blood analysis.",

keywords = "Blood analysis, FEWS, Neural network, Silver halide",

author = "Yaron Gotshal and Ronit Simhi and Sela, {Ben Ami} and Abraham Katzir",

note = "Funding Information: The authors wish to thank F. Moser for helpful discussions, A. Levite for help in producing the fibers, A. Yekuel and A. Gur for their technical assistance, and the Israeli Ministry of Science and Arts and the Commission of the European Communities for partial support of this work.",

year = "1997",

month = aug,

doi = "10.1016/S0925-4005(97)80331-0",

language = "אנגלית",

volume = "42",

pages = "157--161",

journal = "Sensors and Actuators B: Chemical",

issn = "0925-4005",

publisher = "Elsevier",

number = "3 B",

}

TY - JOUR

T1 - Blood diagnostics using fiberoptic evanescent wave spectroscopy and neural networks analysis

AU - Gotshal, Yaron

AU - Simhi, Ronit

AU - Sela, Ben Ami

AU - Katzir, Abraham

N1 - Funding Information: The authors wish to thank F. Moser for helpful discussions, A. Levite for help in producing the fibers, A. Yekuel and A. Gur for their technical assistance, and the Israeli Ministry of Science and Arts and the Commission of the European Communities for partial support of this work.

PY - 1997/8

Y1 - 1997/8

N2 - A spectral analysis of human blood serum was undertaken by fiberoptic evanescent wave spectroscopy (FEWS) using a Fourier transform infrared (FTIR) spectrometer. The blood serum samples were introduced into a special cell designed for FEWS of liquids, with an IR transmitting silver halide fiber as the sensing element. The spectra were analyzed by models of neural networks (NN), and the concentrations of protein, cholesterol and uric acid in human blood serum were obtained. Our results are in agreement with those obtained by ordinary chemical enzymatic methods and multivariate calibration methods. The estimated prediction errors obtained (in percent of the average value) were 4.7% for total protein, 22% for cholesterol and 35% for uric acid. This method can be used for in-situ real-time blood analysis.

AB - A spectral analysis of human blood serum was undertaken by fiberoptic evanescent wave spectroscopy (FEWS) using a Fourier transform infrared (FTIR) spectrometer. The blood serum samples were introduced into a special cell designed for FEWS of liquids, with an IR transmitting silver halide fiber as the sensing element. The spectra were analyzed by models of neural networks (NN), and the concentrations of protein, cholesterol and uric acid in human blood serum were obtained. Our results are in agreement with those obtained by ordinary chemical enzymatic methods and multivariate calibration methods. The estimated prediction errors obtained (in percent of the average value) were 4.7% for total protein, 22% for cholesterol and 35% for uric acid. This method can be used for in-situ real-time blood analysis.

KW - Blood analysis

KW - FEWS

KW - Neural network

KW - Silver halide

UR - http://www.scopus.com/inward/record.url?scp=0031197264&partnerID=8YFLogxK

U2 - 10.1016/S0925-4005(97)80331-0

DO - 10.1016/S0925-4005(97)80331-0

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0031197264

VL - 42

SP - 157

EP - 161

JO - Sensors and Actuators B: Chemical

JF - Sensors and Actuators B: Chemical

SN - 0925-4005

IS - 3 B

ER -

Blood diagnostics using fiberoptic evanescent wave spectroscopy and neural networks analysis

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this