Residual neural networks in single instance-driven identification of fungal pathogens

Main Article Content

Rafał Wyszyński
Karol Struniawski

Keywords : Residual Neural Networks, fungal image classification, deep learning, microscopic images, majority voting, machine learning, image processing

The rise in fungal infections, attributed to various factors including medical interventions and compromised immune systems, necessitates rapid and accurate identification methods. While traditional mycological diagnostics are time-consuming, machine learning offers a promising alternative. Nevertheless, the scarcity of well-curated datasets is a significant obstacle. To address this, a novel approach for identifying fungi in microscopic images using Residual Neural Networks and a subimage retrieval mechanism is proposed, with the final step involving the implementation of majority voting. The new method, applied to the Digital Images of Fungus Species database, surpassed the original patch-based classification using Convolutional Neural Networks, obtaining an overall classification accuracy of 94.7% compared to 82.4% with AlexNet FV SVM. The observed MCC metric exceeds 0.9, while AUC is near to one. This improvement is attributed to the optimization of hyperparameters and top layer architecture, as well as the effectiveness of the Mish activation function in ResNet-based architectures. Noteworthy, the proposed method achieved 100% accurate classification for images from 8 out of 9 classes after majority voting and is high resistant to overfitting, highlighting its potential for rapid and accurate fungal species identification in medical diagnostics and research.

Article Details

How to Cite
Wyszyński, R., & Struniawski, K. (2023). Residual neural networks in single instance-driven identification of fungal pathogens. Machine Graphics and Vision, 32(3/4), 45–64.

A. Agnihotri, P. Saraf, and K. R. Bapnad. A Convolutional Neural Network approach towards self-driving cars. In: IEEE India Conference (INDICON), pp. 1-4. Rajkot, India, 13-15 Dec 2019. (Crossref)

A. Ali, A. Shehzad, M. R. Khan, et al. Yeast, its types and role in fermentation during bread making process - A review. Pak. J. Food Sci., 22:170-178, 2012.

S. Anwar, M. Majid, A. Qayyum, et al. Medical image analysis using convolutional neural networks: A review. J. Med. Syst., 42(11):226, 2018. (Crossref)

S. Bozinovski. Reminder of the first paper on transfer learning in neural networks, 1976. Informatica (Slovenia), 44(3), 2020. (Crossref)

S. Cagatan, T. Mustapha, C. Bagkur, et al. An alternative diagnostic method for C. Neoformans: Preliminary results of deep-learning based detection model. Diagnostics, 13(1):81, 2022. (Crossref)

D. Chicco, N. Tötsch, and G. Jurman. The Matthews correlation coefficient (MCC) is more reliable than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix evaluation. BioData Min., 14(1):13, 2021. (Crossref)

V. M. Corbu, I. Gheorghe-Barbu, A. S. Dumbrava, et al. Current insights in fungal importance - A comprehensive review. Microorganisms, 11(6), 2023. (Crossref)

I. Culjak, D. Abram, T. Pribanic, et al. A brief introduction to OpenCV. In: Proc. of the International Convention (MIPRO), pp. 1725-1730. Opatija, Croatia, 21-25 May 2012.

R. J. W. David E. Rumelhart, Geoffrey E. Hinton. Learning representations by back-propagating errors. Nature, 323:533-536, 1986. (Crossref)

J. Deng, W. Dong, R. Socher, et al. Imagenet: A large-scale hierarchical image database. In: IEEE Conference on Computer Vision and Pattern Recognition (CCVPR), pp. 248-255. Miami, USA, 20-25 Jun 2009. (Crossref)

B. Dutta, D. Lahiri, M. Nag, et al. Fungi in pharmaceuticals and production of antibiotics. In: R. Mahrendra and B. Paul D., eds., Applied Mycology, pp. 233-257. Springer, 2022. (Crossref)

E. Gedraite and M. Hadad. Investigation on the effect of a Gaussian blur in image filtering and segmentation. In: Proc. International Symposium on Electronics in Marine (ELMAR), pp. 393-396. Zadar, Croatia, 14-16 Sep 2011.

J. A. Hagerty, J. Ortiz, D. Reich, et al. Fungal infections in solid organ transplant patients. Surg. Infect., 4(3):263-271, 2003. (Crossref)

K. He, X. Zhang, S. Ren, and J. Sun. Identity mappings in Deep Residual Networks. In: Proc. of European Conf. of Computer Vision, vol. 4, pp. 630-645. Amsterdam, The Netherlands, 11-14 Oct 2016. (Crossref)

K. He, X. Zhang, S. Ren, et al. Deep Residual Learning for Image Recognition. arXiv, 2015. ArXiv:1512.03385.

A. Hosna, E. Merry, J. Gyalmo, et al. Transfer learning: a friendly introduction. J. Big Data, 9:102, 2022. (Crossref)

M. Iorizzo, F. Coppola, F. Letizia, et al. Role of yeasts in the brewing process: Tradition and innovation. Processes, 9(5):839, 2021. (Crossref)

S. R. Kollem and B. Panlal. Enhancement of images using morphological transformations. Glob. J. Comput. Sci. Technol., 4(1), 2012. (Crossref)

A. Konopka, K. Struniawski, and R. Kozera. Performance analysis of Residual Neural Networks in soil bacteria microscopic image classification. In: Modelling and Simulation: The European Simulation and Modelling Conference (ESM), pp. 144-149. Toulouse, France, 24-26 Oct 2023.

A. Krizhevsky, I. Sutskever, and G. E. Hinton. ImageNet classification with deep convolutional neural networks. 60(6):84-90, 2017. (Crossref)

C.-C. J. Kuo. Understanding convolutional neural networks with a mathematical model. J. Vis. Commun. Image Represent., 41:406-413, 2016. (Crossref)

L. Lam and S. Suen. Application of majority voting to pattern recognition: an analysis of its behavior and performance. IEEE Trans. Syst. Man Cybern. Syst., 27(5):553-568, 1997. (Crossref)

C. Lass-Florl. The changing face of epidemiology of invasive fungal disease in europe. Mycoses, 52(3):197-205, 2009. (Crossref)

Q. Li, W. Cai, X. Wang, et al. Medical image classification with convolutional neural network. In: International Conference on Control, Automation, Robotics and Vision (ICARCV), pp. 844-848. Singapore, 10-12 Dec 2014. (Crossref)

S. Liu, W. Cai, Y. Song, et al. Localized sparse code gradient in Alzheimer's disease staging. In: Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 5398-5401. Osaka, Japan, 3-7 Jul 2013. (Crossref)

D. Misra. Mish: A self regularized non-monotonic activation function. ArXiv, 2020. ArXiv:1908.08681.

O. M. Niall, C. Sean, and C. Anderson. In: Computer Vision Conference (CVC), vol. 1, pp. 128-144. Las Vegas, USA, 25-26 Apr 2020. (Crossref)

F. C. Odds. Candida and candidosis. Bailliere Tindall, 2nd edn., 1988.

R. Pascanu, T. Mikolov, and Y. Bengio. On the difficulty of training Recurrent Neural Networks. arXiv, 2013. ArXiv:1211.5063.

M. A. Pfaller and D. J. Diekema. Epidemiology of invasive candidiasis: a persistent public health problem. Clin. Microbiol. Rev., 20(1):133-163, 2007. (Crossref)

S. Rawat, B. Bisht, V. Bisht, et al. Mefunx: A novel meta-learning-based deep learning architecture to detect fungal infection directly from microscopic images. Franklin Open, 6:100069, 2024. (Crossref)

C. F. Rodrigues, S. Silva, and M. Henriques. Candida glabrata: a review of its features and resistance. Eur. J. Clin. Microbiol., 33(5):673-688, 2014. (Crossref)

P. Sahoo, S. Soltani, and A. Wong. A survey of thresholding techniques. Comput. Graph. Image Process., 41:233-260, 1988. (Crossref)

M. Schaefer, S. Migge-Kleian, and S. Scheu. The Role of Soil Fauna for Decomposition of Plant Residues, pp. 207-230. Springer Berlin Heidelberg, 2009. (Crossref)

J. Schmidhuber. Deep learning in neural networks: An overview. Neural Networks, 61:85-117, 2015. (Crossref)

T. R. Singh, S. Roy, O. I. Singh, T. Sinam, and K. M. Singh. A new local adaptive thresholding technique in binarization. ArXiv, 2012. ArXiv:1201.5227.

R. Srisha and A. Khan. Morphological operations for image processing : Understanding and its applications. pp. 17-19. Tamil Nadu, India, 26-27 Apr 2013.

K. Struniawski, A. Konopka, and R. Kozera. Identification of soil bacteria with machine learning and image processing techniques applying single cells' region isolation. In: Modelling and Simulation: The European Simulation and Modelling Conference (ESM), pp. 76-81. Porto, Portugal, 26-28 Oct 2022.

K. Struniawski, R. Kozera, P. Trzcinski, et al. Automated identification of soil fungi and chromista through convolutional neural networks. Eng. Appl. Artif. Intell., 127, 2024. (Crossref)

S. van der Walt, J. L. Schönberger, J. Nunez-Iglesias, et al. scikit-image: image processing in python. PeerJ, 2:e453, 2014. (Crossref)

G. Xie and W. Lu. Image edge detection based on OpenCv. Int. J. Electr. Electron. Eng. Telecommun., 1(1):104-106, 2013. (Crossref)

Z. Xie, J. Li, and H. Shi. A face recognition method based on cnn. In: High Performance Computing and Computational Intelligence Conf., vol. 1395, p. 012006. Chengdu, China, 25–27 Oct 2019. (Crossref)

K. You, M. Long, J. Wang, and M. I. Jordan. How does learning rate decay help modern neural networks. arXiv, 2019. ArXiv:1908.01878.

T. Yu and H. Zhu. Hyper-parameter optimization: A review of algorithms and applications. ArXiv, 2020. ArXiv:2003.05689.

F. Zhuang, Z. Qi, K. Duan, et al. A comprehensive survey on transfer learning. Proc. IEEE, 109(1):43-76, 2021. (Crossref)

B. Zieliński, A. Sroka-Oleksiak, D. Rymarczyk, et al. Deep learning approach to describe and classify fungi microscopic images. PLOS ONE, 15(6):e0234806, 2020. (Crossref)



Download data is not yet available.
Recommend Articles