Abstract
Doubt and the ability to point out details that identify an object or categories of objects are peculiarities of human intelligence. Roughly speaking, artificial intelligence aims to mimic the behavior of human intelligence. This work is a first attempt at joint use of previously existing technologies to mimic these characteristics of human intelligence. This work aims to help in the diagnosis of X-ray chest images with pneumonia, and covid-19; and images of healthy individuals by applying deep neural networks. These deep neural networks are modified so that they can generate predictions with uncertainty. Subsequently, on the previously generated predictions, the salient feature maps are generated to identify on which parts of the image the forecast decision is based. As a result of the work, examples of X-ray chest images will be shown where independent executions predict different labels, focusing attention on different areas of the radiography. So that the different areas indicated by the independent runs might help in the diagnosis of the different pathologies.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Change history
15 September 2022
Correction to: Chapter “Uncertainty Propagation and Salient Features Maps in Deep Learning Architectures for Supporting Covid-19 Diagnosis” in:V. Chang et al. (eds.), Artificial Intelligence and Machine Learning Methods in COVID-19 and Related Health Diseases, Studies in Computational Intelligence 1023,https://doi.org/10.1007/978-3-031-04597-4_1
Notes
- 1.
Where \(p\left( \textbf{y}^{*} | \textbf{f}^{\omega }\left( \textbf{x}^{*}\right) \right) \) is the posterior probability distribution of the theoretical BNN prediction \(\textbf{y}^{*}\) conditioned to the output of the model \(\textbf{f}^{\omega }\left( \textbf{x}^{*}\right) \).
It is supposed that it follows a normal distribution with mean \(\textbf{f}^{\omega }\left( \textbf{x}^{*}\right) \) and variances \(\tau ^{-1} \textbf{I}\), where \(\textbf{I}\) is the identity matrix.
References
Abdar, M., Pourpanah, F., Hussain, S., Rezazadegan, D., Liu, L., Ghavamzadeh, M., & Nahavandi, S. (2021). A review of uncertainty quantification in deep learning: Techniques, applications and challenges. Information Fusion, 76, 243–297. Retrieved from https://doi.org/10.1016/j.inffus.2021.05.008, https://doi.org/10.1016/j.inffus.2021.05.008
Alizadehsani, R., Roshanzamir, M., Hussain, S., Khosravi, A., Koohestani, A., Zangooei, M. H., & Acharya, U. R. (2021). Handling of uncertainty in medical data using machine learning and probability theory techniques: A review of 30 years (1991–2020). Annals of Operations Research. Retrieved from https://doi.org/10.1007/s10479-021-04006-2, https://doi.org/10.1007/s10479-021-04006-2
Arora, S., Bhaskara, A., Ge, R., & Ma, T. (2014, 22-24 Jun). Provable bounds for learning some deep representations. In E. P. Xing & T. Jebara (Eds.), Proceedings References 19 of the 31st international conference on machine learning (Vol. 32, pp. 584-592). Bejing, China: PMLR.
Basu, S., Mitra, S., & Saha, N. (2020). Deep learning for screening covid-19 using chest x-ray images. In 2020 IEEE symposium series on computational intelligence (ssci) (pp. 2521–2527).
Bauer, S., Kohler, S., Schulz, M. H., & Robinson, P. N. (2012). Bayesian ontology querying for accurate and noise-tolerant semantic searches. Bioinformatics, 28(19), 2502–2508. Retrieved from https://doi.org/10.1093/bioinformatics/bts471, https://doi.org/10.1093/bioinformatics/bts471
Brinati, D., Campagner, A., Ferrari, D., Locatelli, M., Banfi, G., & Cabitza, F. (2020). Detection of covid-19 infection from routine blood exams with machine learning: A feasibility study. Journal of Medical Systems, 44(8), 1–12.
Butcher, R., & Fenton, N. (2020). Extending the range of symptoms in a bayesian network for the predictive diagnosis of covid-19. medRxiv.
Chung, A., Wang, L., Wong, A., Lin, Z. Q., McInnis, P., & Gunraj, H. (2020). Figure 1 Covid-19 chest x-ray data initiative. Retrieved from https://github.com/agchung/Figure1-COVID-chestxray-dataset
Cohen, J. P.,Morrison, P., & Dao, L. (2020). Covid-19 image data collection. Retrieved from https://github.com/ieee8023/covid-chestxray-dataset
Das, D., Santosh, K., & Pal, U. (2020). Truncated inception net: Covid-19 outbreak screening using chest x-rays. Physical and Engineering Sciences in Medicine, 43(3), 915–925.
Estrategia Nacional de Inteligencia Artificial (Tech. Rep.). (2020, Nov.). Secretaría General Técnica del Ministerio de Ciencia, Innovación y Universidades.: Gobierno de España.
Gal, Y. (2016a). Bayesian deep learning. In Uncertainty in deep learning (pp. 29–61).
Gal, Y. (2016b). Uncertainty in deep learning (Unpublished doctoral dissertation). Cambridge: University of Cambridge.
Goodfellow, I. J., Bengio, Y., & Courville, A. (2016). Deep learning. Cambridge, MA, USA: MIT Press. http://www.deeplearningbook.org
Gozes, O., Frid-Adar, M., Sagie, N., Zhang, H., Ji, W., & Greenspan, H. (2020). Coronavirus detection and analysis on chest ct with deep learning. arXiv:2004.02640.
He, X., Yang, X., Zhang, S., Zhao, J., Zhang, Y., Xing, E., & Xie, P. (2020). Sampleefficient deep learning for covid-19 diagnosis based on ct scans. medRxiv.
Hinton, G. E., & van Camp, D. (1993). Keeping the neural networks simple by minimizing the description length of the weights. In Proceedings of the sixth annual conference on computational learning theory (pp. 5–13). New York, NY, USA: Association for Computing Machinery, https://doi.org/10.1145/168304.168306.
Kassani, S. H., Kassasni, P. H., Wesolowski, M. J., Schneider, K. A., & Deters, R. (2020). Automatic detection of coronavirus disease (covid-19) in x-ray and ct images: A machine learning-based approach.
Kullback, S. (1959). Information theory and statistics. New York: Wiley
Kwon, Y., Won, J.-H., Kim, B. J., & Paik, M. C. (2018). Uncertainty quantification using bayesian neural networks in classification: Application to ischemic stroke lesion segmentation.
Kwon, Y., Won, J.-H., Kim, B. J., & Paik, M. C. (2020). Uncertainty quantification using bayesian neural networks in classification: Application to biomedical image segmentation. Computational Statistics and Data Analysis, 142, 106816. https://doi.org/10.1016/j.csda.2019.106816.
LeCun, Y. (1989). Generalization and network design strategies (Tech. Rep. No. CRGTR- 89-4). Department of Computer Science, University of Toronto, 10 Kings College Rad, Toronto, Canada: University of Toronto.
Li, L., Qin, L., Xu, Z., Yin, Y., Wang, X., Kong, B., & Xia, J. (2020). Using artificial intelligence to detect covid-19 and community-acquired pneumonia based on pulmonary ct: Evaluation of the diagnostic accuracy.Radiology, 296(2), E65–E71. (PMID: 32191588). https://doi.org/10.1148/radiol.2020200905.
Murdoch, W. J., Singh, C., Kumbier, K., Abbasi-Asl, R., & Yu, B. (2019). Definitions, methods, and applications in interpretable machine learning. Proceedings of the National Academy of Sciences, 116(44), 22071–22080. https://doi.org/10.1073/pnas.1900654116
Neal, R. M. (1995). Bayesian learning for neural networks (Unpublished doctoral dissertation). University of Toronto.
North America, R. S. (2019). Rsna pneumo- nia detection challenge. Retrieved from https://www.kaggle.com/c/rsna-pneumonia-detection-challenge/data
Onisko, A., Druzdzel, M. J., D, P., Wasyluk, H., & D, P. (1999). A bayesian network model for diagnosis of liver disorders (Tech. Rep. No. CBMI-99-27). University of Pittsburgh, Pittsburgh, PA 15260, U.S.A: Center for Biomedical Informatics, University of Pittsburgh.
Paleyes, A., Urma, R., & Lawrence, N. D. (2020). Challenges in deploying machine learning: a survey of case studies. CoRR. ArXiv:abs/2011.09926. Retrieved from https://arxiv.org/abs/2011.09926.
Panwar, H., Gupta, P., Siddiqui, M. K., Morales-Menendez, R., & Singh, V. (2020). Application of deep learning for fast detection of covid-19 in x-rays using ncovnet. Chaos, Solitons & Fractals, 138, 109944.
Robbins, H., & Monro, S. (1951). A stochastic approximation method. Annals of Mathematical Statistics, 22, 400–407.
Rubin, D. B. (1981, 01). The bayesian bootstrap. Annals of Statistics, 9(1), 130–134. Retrieved from https://doi.org/10.1214/aos/1176345338, https://doi.org/10.1214/aos/1176345338.
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning Representations by Back-propagating Errors. Nature, 323(6088), 533–536. Retrieved from http://www.nature.com/articles/323533a0, https://doi.org/10.1038/323533a0.
Sánchez-Pastor, T., Rodríguez-García, I., & Cárdenas-Montes, M. (2020). Clasificación de radiograféas covid con inteligencia artificial. Vértices, 33, 24–26.
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2019). Grad-cam: Visual explanations from deep networks via gradientReferences 21 based localization. International Journal of Computer Vision, 128(2), 336- 359. Retrieved from http://dx.doi.org/10.1007/s11263-019-01228-7, https://doi.org/10.1007/s11263-019-01228-7.
Shen, Y., Zhang, L., Zhang, J., Yang, M., Tang, B., Li, Y., & Lei, K. (2018). Cbn: Constructing a clinical bayesian network based on data from the electronic medical record. Journal of Biomedical Informatics, 88, 1–10. https://doi.org/10.1016/j.jbi.2018.10.007.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15, 1929–1958.
Szegedy, C., Ioffe, S., Vanhoucke, V., & Alemi, A. (2017). Inception-v4, inception-resnet and the impact of residual connections on learning. In Proceedings of the AAAI conference on artificial intelligence (Vol. 31(1)).
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., & Rabinovich, A. (2015). Going deeper with convolutions. In Computer vision and pattern recognition (cvpr). Retrieved from http://arxiv.org/abs/1409.4842.
Tishby, L., & Solla. (1989). Consistent inference of probabilities in layered networks: predictions and generalizations. In International 1989 joint conference on neural networks (Vol. 2, pp. 403–409).
Vázquez-Escobar, J., Hernández, J., & Cárdenas-Montes, M. (2021). Estimation of machine learning model uncertainty in particle physics event classifiers. Computer Physics Communications, 268, 108100. https://doi.org/10.1016/j.cpc.2021.108100.
Wang, L., Lin, Z. Q., & Wong, A. (2020). Covid-net: A tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray images. Scientific Reports, 10(1), 1–12.
Wang, Z., & Yang, J. (2017). Diabetic retinopathy detection via deep convolutional networks for discriminative localization and visual explanation. CoRR. Retrieved from http://arxiv.org/abs/1703.10757
Zech, J. R., Badgeley, M. A., Liu, M., Costa, A. B., Titano, J. J., & Oermann, E. K. (2018). Confounding variables can degrade generalization performance of radiological deep learning models. CoRR. ArXiv:abs/1807.00431.
Zhu, L., & Laptev, N. (n.d.). Engineering uncertainty estimation in neural networks for time series prediction at uber. https://eng.uber.com/neural-networks-uncertaintyestimatio
Acknowledgements
IRG is funded through the PEJ2018-003089-P project to promote Young Employment and Implementation of the Youth Guarantee in R+D+i within the framework of the State Subprogram for the Incorporation of the State Program for the Promotion of Talent and Employability in R+D+i within the framework of the State Plan for Scientific and Technical Research and Innovation 2017–2020. This contract is co-funded by the European Social Fund and the Youth Employment Initiative through the Youth Employment Operational Program. TSP is co-funded in a 91.89% by the European Social Fund within the Youth Employment Operating Program, as well as the Youth Employment Initiative (YEI), and co-found in a 8.11 by the “Comunidad de Madrid (Regional Government of Madrid)” through the project PEJ-2018-AI/TIC-10290. JLGG is co-funded in a 91.89% by the European Social Fund within the Youth Employment Operating Program, as well as the Youth Employment Initiative (YEI), and co-found in a 8,11 by the “Comunidad de Madrid (Regional Government of Madrid)” through the project PEJ-2019-AI/TIC-12440. JVE and MCM are funded by the Spanish Ministry of Economy and Competitiveness (MINECO) for funding support through the grant “Unidad de Excelencia María de Maeztu”: CIEMAT - FÍSICA DE PARTÍCULAS through the grant MDM-2015-0509, and the Spanish Ministry of Science and Innovation for funding support through the grant PID2020-113807RA-I00 “SERVICIOS INNOVADORES DE ANALISIS DE DATOS PARA EL EXPERIMENTO CMS”.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Rodríguez-García, I., Sánchez-Pastor, T., Vázquez-Escobar, J., Gómez-González, J.L., Cárdenas-Montes, M. (2022). Uncertainty Propagation and Salient Features Maps in Deep Learning Architectures for Supporting Covid-19 Diagnosis. In: Chang, V., Kaur, H., Fong, S.J. (eds) Artificial Intelligence and Machine Learning Methods in COVID-19 and Related Health Diseases. Studies in Computational Intelligence, vol 1023. Springer, Cham. https://doi.org/10.1007/978-3-031-04597-4_1
Download citation
DOI: https://doi.org/10.1007/978-3-031-04597-4_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-04596-7
Online ISBN: 978-3-031-04597-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)