Predicting Bacterial Antibiotic Resistance using MALDI-TOF Mass Spectrometry Databases with ELM Applications.

Felipe Tirado; Xaviera Lopez Cortez; Vicente Macaya Mejías; David Zabala-Blanco; José M. Manríquez-Troncoso; Roberto Ahumada-García

doi:10.24215/16666038.24.e08

Authors

Felipe Tirado Universidad Católica del Maule https://orcid.org/0000-0003-2001-658X
Xaviera Lopez Cortez Facultad de Ciencias de la Ingeniería, Universidad Católica del Maule, Talca, Chile https://orcid.org/0000-0002-7514-8777
Vicente Macaya Mejías Facultad de Ciencias de la Ingeniería, Universidad Católica del Maule, Talca, Chile https://orcid.org/0000-0002-1825-0097
David Zabala-Blanco Facultad de Ciencias de la Ingeniería, Universidad Católica del Maule, Talca, Chile https://orcid.org/0000-0002-5692-5673
José M. Manríquez-Troncoso Facultad de Ciencias de la Ingeniería, Universidad Católica del Maule, Talca, Chile https://orcid.org/0000-0001-7571-0048
Roberto Ahumada-García Facultad de Ciencias de la Ingeniería, Universidad Católica del Maule, Talca, Chile https://orcid.org/0000-0003-1107-4606

DOI:

https://doi.org/10.24215/16666038.24.e08

Keywords:

Antibiotic Resistance Prediction, MALDI-TOF Mass Spectrometry, Machine Learning in Medicine, Extreme Learning Machines, Oversampling, Weighted ELM

Abstract

Early detection of antibiotic resistance is a crucial task, especially for vulnerable patients under prolonged treatments with a single antibiotic. To solve this, machine learning approaches have been reported in the state of art. Researchers have used MALDI-TOF MS in order to predict antibiotic resistance and/or susceptibility in bacterial samples. Weis, et al. implemented LR, LightGBM and ANN to study the antibiotic resistance on bacterial strains of Escherichia Coli, Staphylococcus Aureus, and Klebsiella Pneumoniae. Despite promising results, the models have not achieved perfect accuracy, specifically when the classes are unbalanced. On the other hand, Extreme Learning Machine (ELM) is a training algorithm for forward propagation of single hidden layer neural networks, which converges much faster than traditional methods and offers promising performance along with less programmer intervention. In this way, this study introduced improved ELMs, including two weighted ELMs proposed by Zong, and the SMOTE technique in order to create new synthetic samples of the minority class. After heuristic optimization of ELM hiper-parameters, results demonstrated 85% in accuracy and 85% in geometric mean for the classification problem in the case of weighted ELM 1 subject to the SMOTE technique of oversampling.

Downloads

Download data is not yet available.

References

J.O’Neill,“Tacklingdrug-resistantinfectionsglobally: final report and recommendations.” 2016.

G. Mancuso, A. Midiri, E. Gerace, and C. Biondo, “Bacterial antibiotic resistance: The most critical pathogens,” Pathogens, vol. 10, no. 10, p. 1310, 2021.

M. F. Varela, J. Stephen, M. Lekshmi, M. Ojha, N. Wenzel, L. M. Sanford, A. J. Hernandez, A. Par- vathi, and S. H. Kumar, “Bacterial resistance to an- timicrobial agents,” Antibiotics, vol. 10, no. 5, p. 593, 2021.

X. A. Lo ́pez-Corte ́s, J. M. Manr ́ıquez-Troncoso, R. Herna ́ndez-Garc ́ıa, and D. Peralta, “Msdeepamr: an- timicrobial resistance prediction based on deep neural networks and transfer learning,” Frontiers in Microbiology, vol. 15, p. 1361795, 2024.

X. A. Lo ́pez-Corte ́s, J. M. Manr ́ıquez-Troncoso, J. Kandalaft-Letelier, and S. Cuadros-Orellana, “Ma- chine learning and matrix-assisted laser desorp- tion/ionizationtime-of-flightmassspectraforantimi- crobial resistance prediction: A systematic review of recent advancements and future development,” Journal of Chromatography A, p. 465262, 2024.

X. A. Lo ́pez-Corte ́s, F. Matamala, B. Venegas, and C. Rivera, “Machine-learning applications in oral can- cer: A systematic review,” Applied Sciences 2022, Vol. 12, Page 5715, vol. 12, p. 5715, 6 2022. [Online]. Avail- able: https://www.mdpi.com/2076-3417/12/11/5715/ htmhttps://www.mdpi.com/2076-3417/12/11/5715

A. Tapia-Castillo, C. A. Carvajal, X. Lo ́pez-Corte ́s, A. Vecchiola, and C. E. Fardella, “Novel metabolomic profile of subjects with non-classic apparent miner- alocorticoid excess,” Scientific Reports 2021 11:1, vol. 11, pp. 1–12, 8 2021. [Online]. Available: https: //www.nature.com/articles/s41598-021-96628-6

V. R. Olate-Olave, L. Guzma ́n, X. A. Lo ́pez-Corte ́s, R. Cornejo, F. M. Nachtigall, M. Doorn, L. S. San- tos, and A. Bejarano, “Comparison of chilean honeys through maldi-tof-ms profiling and evaluation of their antioxidant and antibacterial potential,” Annals of Agri- cultural Sciences, vol. 66, pp. 152–161, 12 2021.

C.Gonza ́lez,C.A.Astudillo,X.A.Lo ́pez-Corte ́s,and S. Maldonado, “Semi-supervised learning for maldi– tof mass spectrometry data classification: An applica- tion in the salmon industry,” Neural Computing and Applications, vol. 35, no. 13, pp. 9381–9391, 2023.

X. A. Lo ́pez-Corte ́s, F. M. Nachtigall, V. R. Olate, M. Araya, S. Oyanedel, V. Diaz, E. Jakob, M. R ́ıos- Momberg, and L. S. Santos, “Fast detection of pathogens in salmon farming industry,” Aquaculture, vol. 470, pp. 17–24, 3 2017.

X. A. Lopez-Cortes, F. Avila-Salas, C. Orellana, and L. S. Santos, “Strategy based on data mining and maldi-mass spectrometry for control disease of srs in salmo salar,” IEEE ICA-ACCA 2018 - IEEE Interna- tional Conference on Automation/23rd Congress of the Chilean Association of Automatic Control: Towards an Industry 4.0 - Proceedings, 1 2019.

Y. Charretier and J. Schrenzel, “Mass spectrom- etry methods for predicting antibiotic resistance,” PROTEOMICS–Clinical Applications, vol. 10, no. 9- 10, pp. 964–981, 2016.

C. Weis, A. Cue ́nod, B. Rieck, O. Dubuis, S. Graf, C. Lang, M. Oberle, M. Brackmann, K. K. Søgaard, M. Osthoff et al., “Direct antimicrobial resistance pre- diction from clinical maldi-tof mass spectra using ma- chine learning,” Nature Medicine, vol. 28, no. 1, pp. 164–174, 2022.

H. Cruz-Reyes, A. Reyes-Nava, E. R. Lara, and R. Alejo, “Estudio del desbalance de clases en bases de datos de microarrays de expresión genética mediante te ́cnicas de deep learning.” Res. Comput. Sci., vol. 147, no. 5, pp. 197–207, 2018.

W. Zong, G.-B. Huang, and Y. Chen, “Weighted ex- treme learning machine for imbalance learning,” Neu- rocomputing, vol. 101, pp. 229–242, 2013.

S.R.Shrivastava,P.S.Shrivastava,andJ.Ramasamy, “World health organization releases global priority list of antibiotic-resistant bacteria to guide research, dis- covery, and development of new antibiotics,” Journal of Medical Society, vol. 32, no. 1, pp. 76–77, 2018.

D. Zabala-Blanco, M. Mora, R. J. Barrientos, R. Herna ́ndez-Garc ́ıa, and J. Naranjo-Torres, “Finger- print classification through standard and weighted ex- treme learning machines,” applied sciences, vol. 10, no. 12, p. 4125, 2020.

R. Penrose, “A generalized inverse for matrices,” in Mathematical proceedings of the Cambridge philosoph- ical society, vol. 51. Cambridge University Press, 1955, pp. 406–413.

W. Deng, Q. Zheng, and L. Chen, “Regularized ex- treme learning machine,” in 2009 IEEE symposium on computational intelligence and data mining. IEEE, 2009, pp. 389–395.

M. S. Shelke, P. R. Deshmukh, and V. K. Shandilya, “A review on imbalanced data handling using under-

sampling and oversampling technique,” Int. J. Recent Trends Eng. Res, vol. 3, no. 4, pp. 444–449, 2017.

J. Garc ́ıa Abad et al., “Comparativa de te ́cnicas de balanceo de datos. aplicación a un caso real para la prediccio ́n de fuga de clientes,” 2021.

A.Ferna ́ndez,S.Garcia,F.Herrera,andN.V.Chawla, “Smote for learning from imbalanced data: progress and challenges, marking the 15-year anniversary,” Journal of artificial intelligence research, vol. 61, pp. 863– 905, 2018.

A. D. Rasamoelina, F. Adjailia, and P. Sincˇa ́k, “A re- view of activation function for artificial neural network,” in 2020 IEEE 18th World Symposium on Applied Ma- chine Intelligence and Informatics (SAMI). IEEE, 2020, pp. 281–286.