Deep learning-based bacterial genus identification

Authors

  • Md Shafiur Rahman Khan Department of Computer Science and Mathematics, Faculty of Agricultural Engineering & Technology, Bangladesh Agricultural University, Mymensingh, Bangladesh. https://orcid.org/0000-0001-9110-5913
  • Ishrat Khan Department of Computer Science and Engineering, Faculty of Agricultural Engineering and Technology, Khulna Agricultural University, Khulna, Bangladesh. https://orcid.org/0000-0002-7343-6396
  • Md Abdus Sattar Bag Department of Microbiology and Hygiene, Faculty of Veterinary Science, Bangladesh Agricultural University, Mymensingh, Bangladesh. https://orcid.org/0000-0002-6938-6250
  • Machbah Uddin Department of Computer Science and Mathematics, Faculty of Agricultural Engineering & Technology, Bangladesh Agricultural University, Mymensingh, Bangladesh. https://orcid.org/0000-0002-1572-8606
  • Md Rakib Hassan Department of Computer Science and Mathematics, Faculty of Agricultural Engineering & Technology, Bangladesh Agricultural University, Mymensingh, Bangladesh. https://orcid.org/0000-0001-9368-6942
  • Jayedul Hassan Department of Microbiology and Hygiene, Faculty of Veterinary Science, Bangladesh Agricultural University, Mymensingh, Bangladesh. https://orcid.org/0000-0003-2895-8950

Keywords:

Convolutional neural network; deep learning; digital image bacterial genera; automated bacteria identification

Abstract

Objectives: This study aimed to develop a computerized deep learning (DL) technique to identify bacterial genera more precisely in minimum time than the usual, traditional, and commonly used techniques like cultural, staining, and morphological characteristics. Materials and Methods: A convolutional neural network as a part of machine learning (ML) for bacterial genera identification methods was developed using python programming language and the Keras API with TensorFlow ML or DL framework to discriminate bacterial genera, e.g., Streptococcus, Staphylococcus, Escherichia, Salmonella, and Corynebacterium. A total of 200 digital microscopic cell images comprising 40 of each of the genera mentioned above were used in this study. Results: The developed technique could identify and distinguish microscopic images of Streptococcus, Staphylococcus, Escherichia, Salmonella, and Corynebacterium with the highest accuracy of 92.20% for Staphylococcus and the lowest of 77.40% for Salmonella. Among the five epochs, the accuracy rate of bacterial genera identification of Staphylococcus was graded 1, and Streptococcus, Escherichia, Corynebacterium, and Salmonella as 2, 3, 4, and 5, respectively. Conclusion: The experimental results suggest using the DL method to predict bacterial genera included in this study. However, further improvement with more bacterial genera, especially of similar morphology, is necessary to make the technique widely used for bacterial genera identification.

J. Adv. Vet. Anim. Res., 9(4): 573–582, December 2022

http://doi.org/10.5455/javar.2022.i626

Downloads

Abstract
5
PDF
3

Downloads

Published

2022-11-18

How to Cite

Khan, M. S. R., Khan, I., Bag, M. A. S., Uddin, M., Hassan, M. R., & Hassan, J. (2022). Deep learning-based bacterial genus identification. Journal of Advanced Veterinary and Animal Research, 9(4), 573–582. Retrieved from https://banglajol.info/index.php/JAVAR/article/view/80176

Issue

Vol. 9 No. 4 (2022)

Section

Original Articles

License

Copyright (c) 2022 Md Shafiur Rahman Khan, Ishrat Khan, Md Abdus Sattar Bag, Machbah Uddin, Md Rakib Hassan, Jayedul Hassan

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Authors who publish with this journal agree to the following terms:

Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).