Abstract
Background
The use of artificial intelligence, including machine learning, is increasing in medicine. Use of machine learning is rising in the prediction of patient outcomes. Machine learning may also be able to enhance and augment anesthesia clinical procedures such as airway management. In this study, we sought to develop a machine learning algorithm that could classify vocal cords and tracheal airway anatomy real-time during video laryngoscopy or bronchoscopy as well as compare the performance of three novel convolutional networks for detecting vocal cords and tracheal rings.
Methods
Following institutional approval, a clinical dataset of 775 video laryngoscopy and bronchoscopy videos was used. The dataset was divided into two categories for use for training and testing. We used three convolutional neural networks (CNNs): ResNet, Inception and MobileNet. Backpropagation and a mean squared error loss function were used to assess accuracy as well as minimize bias and variance. Following training, we assessed transferability using the generalization error of the CNN, sensitivity and specificity, average confidence error, outliers, overall confidence percentage, and frames per second for live video feeds. After the training was complete, 22 models using 0 to 25,000 steps were generated and compared.
Results
The overall confidence of classification for the vocal cords and tracheal rings for ResNet, Inception and MobileNet CNNs were as follows: 0.84, 0.78, and 0.64 for vocal cords, respectively, and 0.69, 0.72, 0.54 for tracheal rings, respectively. Transfer learning following additional training resulted in improved accuracy of ResNet and Inception for identifying the vocal cords (with a confidence of 0.96 and 0.93 respectively). The two best performing CNNs, ResNet and Inception, achieved a specificity of 0.985 and 0.971, respectively, and a sensitivity of 0.865 and 0.892, respectively. Inception was able to process the live video feeds at 10 FPS while ResNet processed at 5 FPS. Both were able to pass a feasibility test of identifying vocal cords and tracheal rings in a video feed.
Conclusions
We report the development and evaluation of a CNN that can identify and classify airway anatomy in real time. This neural network demonstrates high performance. The availability of artificial intelligence may improve airway management and bronchoscopy by helping to identify key anatomy real time. Thus, potentially improving performance and outcomes during these procedures. Further, this technology may theoretically be extended to the settings of airway pathology or airway management in the hands of experienced providers. The researchers in this study are exploring the performance of this neural network in clinical trials.
This is a preview of subscription content, log in via an institution to check access.
References
Cook, T. M., Woodall, N., Harper, J. et al., Major complications of airway management in the UK: Results of the fourth National Audit Project of the Royal College of Anaesthetists and the difficult airway society. British Journal of Anaesthesia 106(5):617–631, 2011.
Chen, Y.-L., and Wu, K.-H., Airway Management of Patients with Craniofacial Abnormalities: 10-year experience at a teaching Hospital in Taiwan. Journal of the Chinese Medical Association 72(9):468–470, 2009.
Weiss, M., and Engelhardt, T., Proposal for the management of the unexpected difficult pediatric airway. Pediatric Anesthesia 20(5):454–464, 2010.
Aziz, M. F., Dillman, D., Fu, R., and Brambrink, A. M., Comparative effectiveness of the C-MAC video laryngoscope versus direct laryngoscopy in the setting of the predicted difficult airway. Anesthesiology 116(3):629–636, 2012.
Niforopoulou, P., Pantazopoulos, I., Demestiha, T., Koudouna, E., and Xanthos, T., Video-laryngoscopes in the adult airway management: A topical review of the literature. Acta Anaesthesiologica Scandinavica 54:1050–1061, 2010. https://doi.org/10.1111/j.1399-6576.2010.02285.x.
Abdelgadir, I. S., Phillips, R. S., Singh, D., Moncreiff, M. P., and Lumsden, J. L., Videolaryngoscopy versus direct laryngoscopy for tracheal intubation in children (excluding neonates). Cochrane Database of Systematic Reviews. https://doi.org/10.1002/14651858.CD011413.pub2.
Fiadjoe, J. E., Nishisaki, A., Jagannathan, N., Hunyady, A. I., Greenberg, R. S., Reynolds, P. I. et al., Airway management complications in children with difficult tracheal intubation from the pediatric difficult intubation (PeDI) registry: A prospective cohort analysis. Lancet Respir Med 4(1):37–48, 2016.
Howard, A. G., Zhu, M., and Chen, B., Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv, 2017 1704.04861 [cs. CV].
Liu W. et al. (2016) SSD: Single Shot MultiBox Detector. In: Leibe B., Matas J., Sebe N., Welling M. (eds) Computer Vision – ECCV 2016. ECCV 2016. Lecture Notes in Computer Science, vol 9905. Springer, Cham https://doi.org/10.1007/978-3-319-46448-0_2
Szegedy C, et al (2015) Going deeper with convolutions, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, 2015, pp. 1–9. https://doi.org/10.1109/CVPR.2015.7298594
Zhang, Y. C., and Kagen, A. C., Machine Learning Interface for Medical Image Analysis J Digit Imaging 11:1–7, 2016.
Ren, S., He, K., Girshick, R., and Sun, J., Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence 39(6):1137–1149, 2017. https://doi.org/10.1109/TPAMI.2016.2577031.
Girshick R (2015) Fast R-CNN. arXiv:1504.08083 [cs.CV]
He K, Zhang X, Ren S, Sun J (2016) Deep Residual Learning for Image Recognition IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2016.90
Montgomery, D. C., Jennings, C. L., and Kulahci, M., Introduction to time series analysis and forecasting Wiley. New: Jersey, 2015.
Funding
This study was funded by departmental funds.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interest
Clyde Matava declares that he has no conflict of interest. Evelina Pankiv declares that she no conflict of interest. Sam Raisbeck declares that he has no conflict of interest. Monica Caldeira declares that she has no conflict of interest. Fahad Alam declares that he has no conflict of interest.
Ethical Approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
What is already known?
• Failure to classify vocal cords and tracheal rings anatomy during intubation and bronchoscopy during laryngoscopy and bronchoscopy can lead to adverse events
• Machine learning and artificial intelligence may be able to play a role in real-time clinical decision making
What is new?
• We report the successful development and testing of a convolutional neural network that can classify, identify, and label vocal cords and tracheal rings from live video.
This article is part of the Topical Collection on Image & Signal Processing
Electronic supplementary material
ESM 1
(MP4 41816 kb)
Rights and permissions
About this article
Cite this article
Matava, C., Pankiv, E., Raisbeck, S. et al. A Convolutional Neural Network for Real Time Classification, Identification, and Labelling of Vocal Cord and Tracheal Using Laryngoscopy and Bronchoscopy Video. J Med Syst 44, 44 (2020). https://doi.org/10.1007/s10916-019-1481-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10916-019-1481-4