Abstract
This work presents twelve fine-tuned deep learning architectures to solve the bacterial classification problem over the digital image of bacterial species dataset. The base architectures were mainly published as mobile or efficient solutions to the ImageNet challenge, and all experiments presented in this work consisted in making several modifications to the original designs, in order to make them able to solve the bacterial classification problem by using fine-tuning and transfer learning techniques. This work also proposes a novel data augmentation technique for this dataset, which is based on the idea of artificial zooming, strongly increasing the performance of every tested architecture, even doubling it in some cases. In order to get robust and complete evaluations, all experiments were performed with 10-fold cross-validation and evaluated with five different metrics: top-1 and top-5 accuracy, precision, recall, and F1 score. This paper presents a complete comparison of the twelve different architectures, cross-validated with the original and the augmented version of the dataset, the results are also compared with several literature methods. Overall, eight of the eleven architectures surpassed the 0.95 score in top-1 accuracy with our data augmentation method, being 0.9738 the highest top-1 accuracy. The impact of the data augmentation technique is reported with relative improvement scores.
Similar content being viewed by others
Code Availability
The software requirements, Jupyter notebooks, Python scripts and detailed instructions to replicate our experiments are available on the following GitHub repository: github.com/gallardorafael/EfficientMobileDL_Bacterial
References
Abd Elaziz M, Hosny KM, Hemedan AA, Darwish MM (2020) Improved recognition of bacterial species using novel fractional-order orthogonal descriptors. Appl Soft Comput 95:106504
Ahmed WM, Bayraktar B, Bhunia AK, Hirleman ED, Robinson JP, Rajwa B (2012) Classification of bacterial contamination using image processing and distributed computing. IEEE J Biomed Health Inform 17(1):232–239
Aslam B, Basit M, Nisar MA, Khurshid M, Rasool MH (2017) Proteomics: technologies and their applications. J Chromatogr Sci 55(2):182–196
Deng J, Dong W, Socher R, Li LJ, Li K, Fei-Fei L (2009) Imagenet: a large-scale hierarchical image database. IN: 2009 IEEE Conference on computer vision and pattern recognition, IEEE, pp 248–255
Ferri C, Hernández-Orallo J, Modroiu R (2009) An experimental comparison of performance measures for classification. Pattern Recogn Lett 30(1):27–38
Forero MG, Cristóbal G, Desco M (2006) Automatic identification of mycobacterium tuberculosis by gaussian mixture models. J Microsc 223 (2):120–132
Franco-Duarte R et al, Çernáková L, Kadam S, Kaushik K, Salehi B, Bevilacqua A, Corbo MR, Antolak H, Dybka-Stępień K, Leszczewicz M (2019) Advances in chemical and biological methods to identify microorganisms—from past to present. Microorganisms 7(5):130
Gallardo-García R, Jarquín-Rodríguez A, Beltrán-Martínez B, Martínez R (2020) Deep learning for fast identification of bacterial strains in resource constrained devices. In: Aplicaciones científicas y tecnológicas de las ciencias computacionales, pp 67–78
Guo Y, Liu Y, Oerlemans A, Lao S, Wu S, Lew MS (2016) Deep learning for visual understanding: A review. Neurocomputing 187:27–48
Holmberg M, Gustafsson F, Hörnsten EG, Winquist F, Nilsson LE, Ljung L, Lundström I (1998) Bacteria classification based on feature extraction from sensor data. Biotechnol Tech 12(4):319–324
Howard A, Sandler M, Chu G, Chen LC, Chen B, Tan M, Wang W, Zhu Y, Pang R, Vasudevan V et al (2019) Searching for mobilenetv3. In: Proceedings of the IEEE international conference on computer vision, pp 1314–1324
Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K (2016) Squeezenet: Alexnet-level accuracy with 50x fewer parameters and< 0.5 mb model size. arXiv preprint arXiv:1602.07360
Jenkins SA, Drucker D, Keaney MG, Ganguli LA (1991) Evaluation of the rapid id 32a system for the identification of bacteroides fragilis and related organisms. J Appl Bacteriol 71(4):360–365
Khalifa NEM, Taha MHN, Hassanien AE, Hemedan AA (2019) Deep bacteria: Robust deep learning data augmentation design for limited bacterial colony dataset. Int J Reason Intell Syst 11(3):256–264
Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097–1105
Liu J, Dazzo FB, Glagoleva O, Yu B, Jain AK (2001) Cmeias: A computer-aided system for the image analysis of bacterial morphotypes in microbial communities. Microb Ecol 41(3):173–194
Loshchilov I, Hutter F (2017) Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101
Ma N, Zhang X, Zheng HT, Sun J (2018) Shufflenet v2: Practical guidelines for efficient cnn architecture design. In: Proceedings of the European conference on computer vision (ECCV), pp 116–131
MicrobiologySociety: Microbes and disease (2021). Accessed on April 20, 2021. https://microbiologysociety.org/why-microbiology-matters/what-is-microbiology/microbes-and-the-human-body/microbes-and-disease.html
Mishra M, Chauhan P (2016) Applications of microscopy in bacteriology. Microsc Res 4(1):1–9
Nasip ÖF, Zengin K (2018) Deep learning based bacteria classification. In: 2018 2Nd international symposium on multidisciplinary studies and innovative technologies (ISMSIT), IEEE, pp 1–5
Nobile CJ, Johnson AD (2015) Candida albicans biofilms and human disease. Annu Rev Microbiol 69:71–92
Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L et al (2019) Pytorch: an imperative style, high-performance deep learning library. In: Advances in neural information processing systems, pp 8026–8037
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: Machine learning in Python. J Mach Learn Res 12:2825–2830
Perronnin F, Dance C (2007) Fisher kernels on visual vocabularies for image categorization. In: 2007 IEEE Conference on computer vision and pattern recognition, IEEE, pp 1–8
Rujichan C, Vongserewattana N, Phasukkit P (2019) Bacteria classification using image processing and deep convolutional neural network. In: 2019 12Th biomedical engineering international conference (BMEiCON), IEEE, pp 1–4
Sandler M, Howard A, Zhu M, Zhmoginov A, Chen LC (2018) Mobilenetv2: Inverted residuals and linear bottlenecks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4510–4520
Satoto BD, Utoyo MI, Rulaningtyas R, Koendhori EB (2020) An auto contrast custom convolutional neural network to identifying gram-negative bacteria. In: 2020 International conference on computer engineering, network, and intelligent multimedia (CENIM), IEEE, pp 70–75
Sauer S, Kliem M (2010) Mass spectrometry tools for the classification and identification of bacteria. Nat Rev Microbiol 8(1):74–82
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556
Singhal N, Kumar M, Kanaujia PK, Virdi JS (2015) Maldi-tof mass spectrometry: An emerging technology for microbial identification and diagnosis. Front Microbiol 6:791
Sommer C, Gerlich DW (2013) Machine learning in cell biology–teaching computers to recognize phenotypes. J Cell Sci 126(24):5529–5539
Spratt DA (2004) Significance of bacterial identification by molecular biology methods. Endod Top 9(1):5–14
Talo M (2019) An automated deep learning approach for bacterial image classification. arXiv preprint arXiv:1912.08765
Tan M, Le QV (2019) Efficientnet: Rethinking model scaling for convolutional neural networks. arXiv preprint arXiv:1905.11946
Trattner S, Greenspan H, Tepper G, Abboud S (2004) Automatic identification of bacterial types using statistical imaging methods. IEEE Trans Med Imaging 23(7):807–820
Váradi L, Luo JL, Hibbs DE, Perry JD, Anderson RJ, Orenga S, Groundwater PW (2017) Methods for the detection and identification of pathogenic bacteria: Past, present, and future. Chem Soc Rev 46(16):4818–4832
Voulodimos A, Doulamis N, Doulamis A, Protopapadakis E (2018) Deep learning for computer vision: A brief review. Computational intelligence and neuroscience 2018
Zieliński B, Plichta A, Misztal K, Spurek P, Brzychczy-włoch M., Ochońska D. (2017) Deep learning approach to bacterial colony classification. Plos One 12(9):e0184554
Funding
This work did not receive funding of any kind.
Author information
Authors and Affiliations
Contributions
Rafael Gallardo García and Sofía Jarquín Rodríguez contributed to the study conception, design, experimentation and writing of the manuscript. Chemical and biological background of bacterial identification provided in this paper was performed by Sofía Jarquín Rodríguez. Rafael Gallardo García worked on data processing, system design, programming, training and evaluation related tasks. Beatriz Beltrán Martínez and Carlos Hernández Gracidas provided useful and complete feedback to ensure strong and robust research results, also, they contributed with detailed revisions of the paper structure, content and language-quality. Rodolfo Martínez Torres provided feedback and assistance for some engineering-related tasks. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethics approval
As the original data was acquired in [40], none of the authors of this work were exposed to dangerous bacteria, biological material or to dangers of any kind. With the aforementioned, we consider that this section is not applicable to this paper.
Conflict of Interests
The authors have declared that no conflict of interests or competing interests exist.
Additional information
Availability of data and material
In the Data Availability section of their paper [40], authors of the DIBaS dataset provided the following link to access the data: http://misztal.edu.pl/software/databases/dibas
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Gallardo García, R., Jarquín Rodríguez, S., Beltrán Martínez, B. et al. Efficient deep learning architectures for fast identification of bacterial strains in resource-constrained devices. Multimed Tools Appl 81, 39915–39944 (2022). https://doi.org/10.1007/s11042-022-13022-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-13022-8