Rapid and simple quantitative measurement of α-fetoprotein by combining immunochromatographic strip test and artificial neural network image analysis system

被引:44
作者
Chuang, L
Hwang, JY
Chang, HC
Chang, FM
Jong, SB
机构
[1] Kaohsiung Med Univ, Sch Med Radiat, Kaohsiung, Taiwan
[2] Natl Cheng Kung Univ, Coll Med, Dept Obstet & Gynecol, Tainan, Taiwan
[3] Hosp Tainan, Tainan, Taiwan
关键词
alpha-fetoprotein; immunochromatographic strip; quantification; artificial neural network; image analysis;
D O I
10.1016/j.cccn.2004.05.010
中图分类号
R446 [实验室诊断]; R-33 [实验医学、医学实验];
学科分类号
1001 ;
摘要
Background: Quantitative immunochromatographic strip (ICS) assay can facilitate clinical diagnosis by providing more information than traditional qualitative or semiquantitative strip assay. Methods: We constructed a human serum a-fetoprotein (AFP) measurement system by combining semiquantitative ICS tests and artificial neural network (ANN) image analysis system [immunochromatographic strip analyzed by artificial neural network image analysis system (IAIS)]. After ICS tests completed, the AFP concentration can be obtained from analysis of IAIS software. The equipment required are commercial semiquantitative AFP dipstick, optical scanner, centrifuge (option), personal computer, and software of IAIS. Results: The serum AFP concentrations measured by IAIS were strongly correlated (r=0.9971) with that by RIA. Using Bland-Altman analysis, the IAIS achieved clinical acceptable limits of agreement in comparison with RIA. The within-run precision of IAIS, expressed as coefficients of variation (CV), at 81.7 ng/ml was 1.50% and, at 244.4 ng/ml, was 1.09%. The measurement of serum AFP by IAIS can be completed in 20 min. Conclusions: The newly constructed quantitative immunochromatographic strip assay (IAIS) is a simple, rapid, and reliable method for serum AFP measurement. With the simple equipment required, the IAIS can be performed outside the laboratory and is ideal for outpatient or point-of-care AFP testing. (C) 2004 Elsevier B.V. All rights reserved.
引用
收藏
页码:87 / 93
页数:7
相关论文
共 18 条
  • [1] APPLICATION OF ARTIFICIAL NEURAL NETWORKS TO CLINICAL MEDICINE
    BAXT, WG
    [J]. LANCET, 1995, 346 (8983): : 1135 - 1138
  • [2] Prospective validation of artificial neural network trained to identify acute myocardial infarction
    Baxt, WG
    Skora, J
    [J]. LANCET, 1996, 347 (8993) : 12 - 15
  • [3] DEMONSTRATION OF A NEW PROTEIN FRACTION IN SERUM FROM THE HUMAN FETUS
    BERGSTRAND, CG
    CZAR, B
    [J]. SCANDINAVIAN JOURNAL OF CLINICAL & LABORATORY INVESTIGATION, 1956, 8 (02) : 174 - 174
  • [4] Measuring agreement in method comparison studies
    Bland, JM
    Altman, DG
    [J]. STATISTICAL METHODS IN MEDICAL RESEARCH, 1999, 8 (02) : 135 - 160
  • [5] STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENT
    BLAND, JM
    ALTMAN, DG
    [J]. LANCET, 1986, 1 (8476) : 307 - 310
  • [6] Artificial neural networks applied to outcome prediction for colorectal cancer patients in separate institutions
    Bottaci, L
    Drew, PJ
    Hartley, JE
    Hadfield, MB
    Farouk, R
    Lee, PWR
    Macintyre, IMC
    Duthie, GS
    Monson, JRT
    [J]. LANCET, 1997, 350 (9076) : 469 - 472
  • [7] A rapid, simple measurement of human albumin in whole blood using a fluorescence immunoassay (I)
    Choi, S
    Choi, EY
    Kim, DJ
    Kim, JH
    Kim, TS
    Oh, SW
    [J]. CLINICA CHIMICA ACTA, 2004, 339 (1-2) : 147 - 156
  • [8] Chou SF, 2002, CLIN CHEM, V48, P913
  • [9] Ultrasound estimation of fetal weight with the use of computerized artificial neural network model
    Chuang, L
    Hwang, JY
    Chang, CH
    Yu, CH
    Chang, FM
    [J]. ULTRASOUND IN MEDICINE AND BIOLOGY, 2002, 28 (08) : 991 - 996
  • [10] ARTIFICIAL NEURAL NETWORKS IN PATHOLOGY AND MEDICAL LABORATORIES
    DYBOWSKI, R
    GANT, V
    [J]. LANCET, 1995, 346 (8984): : 1203 - 1207