News |  Sitemap |  Contact
PDF Export  | 

ImageJ Plugins for Automatic Bacteria Colony Enumeration

Abstract:
Clonogenic assay is a broadly used biological assay in biomedical examinations, food and drug safety test, environmental monitoring, and public health. In clonogenic assay, bacteria colony enumeration is a tedious but essential step because hundreds or thousands of bacteria colonies on a Petri dish must be manually counted by well-trained technicians. In spite of the time wasting, colony recognition is also an error-prone process since the identification of bacteria colonies is in consistency and subjective from person to person. Some colony counters are available on the market, but most of them are semi-automatic counters which still depend on manual labor to select plate region and specify parameters. Few deluxe models are designed to enumerate bacteria colonies in a fully automatic manner, but they are usually equipped with a high quality image acquiring device, and/or add fluorogenic substrates in the medium for produce fluorescent product for detection. However, the aforementioned requirements make the bacteria colony enumeration system very expensive, and thus, are not economical to most research labs. In this research project, we propose an algorithm to simulate the human recognition behavior on the basis of the hierarchical layout of bacteria colony images. Based on the proposed algorithm, we introduce a fully automatic yet cost-effective bacterial colony enumeration plugins for ImageJ. The plugins can accept general digital camera images as its input. This software-centered counter adopts proposed algorithm for detecting various dish regions, identifying colonies, separating aggregated colonies, and finally report counting results without human intervention. In our experiments, we applied the proposed ImageJ plugins on images with blue color Mitis-Salivarius agar which is used for isolating Mutans Strptococci, a kind of acid-producing bacteria that attack tooth enamel minerals and cause dental caries. The experimental results show the proposed counter demonstrates a promising performance in terms of both precision and recall, and is robust and efficient in terms of labor- and time-savings.

Keywords:
medical image, colony enumeration, colony counting, bacteria colony, image processing

Authors
Wei-Bang Chen, Wen-Lin Liu

Organisation   
University of Alabama at Birmingham

Homepage   

Short Biography   
Wei-Bang Chen is a Ph.D. student in the Computer and Information Sciences Department at University of Alabama at Birmingham (UAB), Birmingham, AL, USA. He received his M.S. degree in Genetics from National Yang-Ming University in Taipei, Taiwan in 1999 and a M.S. degree in Computer Sciences from UAB in 2005. His main research interests include bioinformatics, multimedia data mining, image processing, and computer vision. His current research involves microarray image and data analysis, biological sequence clustering, biomedical video and image mining, and spam image mining.

© Luxembourg Institute of Science and Technology | Legal Notice