Print ISSN : 0919-2719
Volume 12 , Issue 1
Showing 1-1 articles out of 1 articles from the selected issue
Regular Article
  • Nobuyasu Yamaguchi, Tomoaki Ichijo, Michihiro Ogawa, Katsuji Tani, Mas ...
    Type: scientific monograph
    Subject area: Infomation Science
    2004 Volume 12 Issue 1 Pages 1-7
    Published: 2004
    Released: October 28, 2005
    The digital image analysis software BACS II was developed to enumerate multicolor stained cells rapidly and easily by fluorescence microscopy with multiple excitation. Image analysis requires the following: (i) acquisition of RGB images of the same microscopic fields under different excitations, (ii) subtraction of background, (iii) smoothing and edge detection, (iv) particle recognition and measurement of characteristics (area, length, RGB intensity, etc.), (v) discrimination of cells from other particles, (vi) classification of each cell, and (vii) enumeration. Using the algorithm of BACS II, bacterial cells can be counted based on more than three excitations. The accuracy of enumeration by BACS II was examined in a mixture of three cultured strains (Escherichia coli O157:H7, E. coli K-12, and Staphylococcus epidermidis) triple stained with DAPI, FITC-labeled anti-E. coli O157:H7 antibody, and Cy3-labeled rRNA-targeted probes for specific detection of E. coli cells. Pond water containing indigenous bacteria, algae, and genetically engineered E. coli with a green fluorescent protein-producing gene was stained with DAPI and evaluated as well. There was a strong linear correlation between the counts determined by microscopic visual counting and image analysis for each microbial population (r2=0.94-0.98) over the range of 102-107 cells/ml. Enumeration of multicolor stained microbial cells by BACS II required less than 15 min from image acquisition to obtain results, while conventional microscopic visual counting took more than 30 min. BACS II should prove useful for any research using fluorescence microscopy, especially when multicolor analysis is needed.
    Download PDF (2404K)