IDEAS home Printed from https://ideas.repec.org/a/gam/jmathe/v11y2023i14p3081-d1192627.html
   My bibliography  Save this article

Transformer Models and Convolutional Networks with Different Activation Functions for Swallow Classification Using Depth Video Data

Author

Listed:
  • Derek Ka-Hei Lai

    (Department of Biomedical Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China
    These authors contributed equally to this work.)

  • Ethan Shiu-Wang Cheng

    (Department of Electronic and Information Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China
    These authors contributed equally to this work.)

  • Bryan Pak-Hei So

    (Department of Biomedical Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China)

  • Ye-Jiao Mao

    (Department of Biomedical Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China)

  • Sophia Ming-Yan Cheung

    (Department of Mathematics, School of Science, The Hong Kong University of Science and Technology, Hong Kong 999077, China)

  • Daphne Sze Ki Cheung

    (School of Nursing, The Hong Kong Polytechnic University, Hong Kong 999077, China
    Research Institute of Smart Ageing, The Hong Kong Polytechnic University, Hong Kong 999077, China)

  • Duo Wai-Chi Wong

    (Department of Biomedical Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China)

  • James Chung-Wai Cheung

    (Department of Biomedical Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China
    Research Institute of Smart Ageing, The Hong Kong Polytechnic University, Hong Kong 999077, China)

Abstract

Dysphagia is a common geriatric syndrome that might induce serious complications and death. Standard diagnostics using the Videofluoroscopic Swallowing Study (VFSS) or Fiberoptic Evaluation of Swallowing (FEES) are expensive and expose patients to risks, while bedside screening is subjective and might lack reliability. An affordable and accessible instrumented screening is necessary. This study aimed to evaluate the classification performance of Transformer models and convolutional networks in identifying swallowing and non-swallowing tasks through depth video data. Different activation functions (ReLU, LeakyReLU, GELU, ELU, SiLU, and GLU) were then evaluated on the best-performing model. Sixty-five healthy participants ( n = 65) were invited to perform swallowing (eating a cracker and drinking water) and non-swallowing tasks (a deep breath and pronouncing vowels: “/eɪ/”, “/iː/”, “/aɪ/”, “/oʊ/”, “/u:/”). Swallowing and non-swallowing were classified by Transformer models (TimeSFormer, Video Vision Transformer (ViViT)), and convolutional neural networks (SlowFast, X3D, and R(2+1)D), respectively. In general, convolutional neural networks outperformed the Transformer models. X3D was the best model with good-to-excellent performance (F1-score: 0.920; adjusted F1-score: 0.885) in classifying swallowing and non-swallowing conditions. Moreover, X3D with its default activation function (ReLU) produced the best results, although LeakyReLU performed better in deep breathing and pronouncing “/aɪ/” tasks. Future studies shall consider collecting more data for pretraining and developing a hyperparameter tuning strategy for activation functions and the high dimensionality video data for Transformer models.

Suggested Citation

  • Derek Ka-Hei Lai & Ethan Shiu-Wang Cheng & Bryan Pak-Hei So & Ye-Jiao Mao & Sophia Ming-Yan Cheung & Daphne Sze Ki Cheung & Duo Wai-Chi Wong & James Chung-Wai Cheung, 2023. "Transformer Models and Convolutional Networks with Different Activation Functions for Swallow Classification Using Depth Video Data," Mathematics, MDPI, vol. 11(14), pages 1-22, July.
  • Handle: RePEc:gam:jmathe:v:11:y:2023:i:14:p:3081-:d:1192627
    as

    Download full text from publisher

    File URL: https://www.mdpi.com/2227-7390/11/14/3081/pdf
    Download Restriction: no

    File URL: https://www.mdpi.com/2227-7390/11/14/3081/
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Jing Lei, 2020. "Cross-Validation With Confidence," Journal of the American Statistical Association, Taylor & Francis Journals, vol. 115(532), pages 1978-1997, December.
    2. Bryan Pak-Hei So & Derek Ka-Hei Lai & Daphne Sze-Ki Cheung & Wing-Kai Lam & James Chung-Wai Cheung & Duo Wai-Chi Wong, 2022. "Virtual Reality-Based Immersive Rehabilitation for Cognitive- and Behavioral-Impairment-Related Eating Disorders: A VREHAB Framework Scoping Review," IJERPH, MDPI, vol. 19(10), pages 1-18, May.
    3. Takaya Saito & Marc Rehmsmeier, 2015. "The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets," PLOS ONE, Public Library of Science, vol. 10(3), pages 1-21, March.
    4. Hyo-Jung Lim & Derek Ka-Hei Lai & Bryan Pak-Hei So & Calvin Chi-Kong Yip & Daphne Sze Ki Cheung & James Chung-Wai Cheung & Duo Wai-Chi Wong, 2023. "A Comprehensive Assessment Protocol for Swallowing (CAPS): Paving the Way towards Computer-Aided Dysphagia Screening," IJERPH, MDPI, vol. 20(4), pages 1-9, February.
    5. Andy Yiu-Chau Tam & Li-Wen Zha & Bryan Pak-Hei So & Derek Ka-Hei Lai & Ye-Jiao Mao & Hyo-Jung Lim & Duo Wai-Chi Wong & James Chung-Wai Cheung, 2022. "Depth-Camera-Based Under-Blanket Sleep Posture Classification Using Anatomical Landmark-Guided Deep Learning Model," IJERPH, MDPI, vol. 19(20), pages 1-12, October.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Christopher J Greenwood & George J Youssef & Primrose Letcher & Jacqui A Macdonald & Lauryn J Hagg & Ann Sanson & Jenn Mcintosh & Delyse M Hutchinson & John W Toumbourou & Matthew Fuller-Tyszkiewicz &, 2020. "A comparison of penalised regression methods for informing the selection of predictive markers," PLOS ONE, Public Library of Science, vol. 15(11), pages 1-14, November.
    2. Jie-Huei Wang & Cheng-Yu Liu & You-Ruei Min & Zih-Han Wu & Po-Lin Hou, 2024. "Cancer Diagnosis by Gene-Environment Interactions via Combination of SMOTE-Tomek and Overlapped Group Screening Approaches with Application to Imbalanced TCGA Clinical and Genomic Data," Mathematics, MDPI, vol. 12(14), pages 1-24, July.
    3. Hyo-Jung Lim & Derek Ka-Hei Lai & Bryan Pak-Hei So & Calvin Chi-Kong Yip & Daphne Sze Ki Cheung & James Chung-Wai Cheung & Duo Wai-Chi Wong, 2023. "A Comprehensive Assessment Protocol for Swallowing (CAPS): Paving the Way towards Computer-Aided Dysphagia Screening," IJERPH, MDPI, vol. 20(4), pages 1-9, February.
    4. Le, Hong Hanh & Viviani, Jean-Laurent, 2018. "Predicting bank failure: An improvement by implementing a machine-learning approach to classical financial ratios," Research in International Business and Finance, Elsevier, vol. 44(C), pages 16-25.
    5. João Chang Junior & Fábio Binuesa & Luiz Fernando Caneo & Aida Luiza Ribeiro Turquetto & Elisandra Cristina Trevisan Calvo Arita & Aline Cristina Barbosa & Alfredo Manoel da Silva Fernandes & Evelinda, 2020. "Improving preoperative risk-of-death prediction in surgery congenital heart defects using artificial intelligence model: A pilot study," PLOS ONE, Public Library of Science, vol. 15(9), pages 1-21, September.
    6. Arthur De Sá Ferreira & Ney Meziat-Filho & Ana Paula Antunes Ferreira, 2021. "Double threshold receiver operating characteristic plot for three-modal continuous predictors," Computational Statistics, Springer, vol. 36(3), pages 2231-2245, September.
    7. Fan, Xudong & Wang, Xiaowei & Zhang, Xijin & ASCE Xiong (Bill) Yu, P.E.F., 2022. "Machine learning based water pipe failure prediction: The effects of engineering, geology, climate and socio-economic factors," Reliability Engineering and System Safety, Elsevier, vol. 219(C).
    8. Zhang, Han, 2021. "How Using Machine Learning Classification as a Variable in Regression Leads to Attenuation Bias and What to Do About It," SocArXiv 453jk, Center for Open Science.
    9. Masabho P Milali & Samson S Kiware & Nicodem J Govella & Fredros Okumu & Naveen Bansal & Serdar Bozdag & Jacques D Charlwood & Marta F Maia & Sheila B Ogoma & Floyd E Dowell & George F Corliss & Maggy, 2020. "An autoencoder and artificial neural network-based method to estimate parity status of wild mosquitoes from near-infrared spectra," PLOS ONE, Public Library of Science, vol. 15(6), pages 1-16, June.
    10. Daniel R Jeske, 2018. "Metrics Used When Evaluating the Performance of Statistical Classifiers," Biostatistics and Biometrics Open Access Journal, Juniper Publishers Inc., vol. 8(1), pages 7-9, August.
    11. Juliet Chebet Moso & Stéphane Cormier & Cyril de Runz & Hacène Fouchal & John Mwangi Wandeto, 2021. "Anomaly Detection on Data Streams for Smart Agriculture," Agriculture, MDPI, vol. 11(11), pages 1-17, November.
    12. Kajal Lahiri & Cheng Yang, 2023. "ROC and PRC Approaches to Evaluate Recession Forecasts," Journal of Business Cycle Research, Springer;Centre for International Research on Economic Tendency Surveys (CIRET), vol. 19(2), pages 119-148, September.
    13. Tzu-Hsuan Lin & Jehn-Ruey Jiang, 2021. "Credit Card Fraud Detection with Autoencoder and Probabilistic Random Forest," Mathematics, MDPI, vol. 9(21), pages 1-16, October.
    14. Robert A. Blair & Nicholas Sambanis, 2021. "Is Theory Useful for Conflict Prediction? A Response to Beger, Morgan, and Ward," Journal of Conflict Resolution, Peace Science Society (International), vol. 65(7-8), pages 1427-1453, August.
    15. Mieke Deschepper & Willem Waegeman & Dirk Vogelaers & Kristof Eeckloo, 2020. "Using structured pathology data to predict hospital-wide mortality at admission," PLOS ONE, Public Library of Science, vol. 15(6), pages 1-11, June.
    16. Alfred Krzywicki & David Muchlinski & Benjamin E. Goldsmith & Arcot Sowmya, 2022. "From academia to policy makers: a methodology for real-time forecasting of infrequent events," Journal of Computational Social Science, Springer, vol. 5(2), pages 1489-1510, November.
    17. Falco J. Bargagli-Dtoffi & Massimo Riccaboni & Armando Rungi, 2020. "Machine Learning for Zombie Hunting. Firms Failures and Financial Constraints," Working Papers 01/2020, IMT School for Advanced Studies Lucca, revised Jun 2020.
    18. Marco Due~nas & V'ictor Ortiz & Massimo Riccaboni & Francesco Serti, 2021. "Assessing the Impact of COVID-19 on Trade: a Machine Learning Counterfactual Analysis," Papers 2104.04570, arXiv.org.
    19. Bryan Pak-Hei So & Tim Tin-Chun Chan & Liangchao Liu & Calvin Chi-Kong Yip & Hyo-Jung Lim & Wing-Kai Lam & Duo Wai-Chi Wong & Daphne Sze Ki Cheung & James Chung-Wai Cheung, 2022. "Swallow Detection with Acoustics and Accelerometric-Based Wearable Technology: A Scoping Review," IJERPH, MDPI, vol. 20(1), pages 1-14, December.
    20. Jianyuan Deng & Zhibo Yang & Hehe Wang & Iwao Ojima & Dimitris Samaras & Fusheng Wang, 2023. "A systematic study of key elements underlying molecular property prediction," Nature Communications, Nature, vol. 14(1), pages 1-20, December.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:gam:jmathe:v:11:y:2023:i:14:p:3081-:d:1192627. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: MDPI Indexing Manager (email available below). General contact details of provider: https://www.mdpi.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.