• Home
  • Search
  • Transformer Models and Convolutional Networks with Different Activation Functions for Swallow Classification Using Depth Video Data
  • Cite Icon16
  • https://doi.org/10.3390/math11143081Copy DOI Icon

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

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
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.

Loading PDF

Similar Papers
  • PDF
  • Research Article
  • Citations25

Research on improved convolutional wavelet neural network

  • Sep 09, 2021
  • Scientific Reports
  • Jingwei Liu +4
  • Research Article
  • Citations3

Classification of COVID-19 from CT chest images using Convolutional Wavelet Neural Network

  • Feb 01, 2023
  • International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering
  • Dina A Abdulqader +3
  • Preprint Article

Semantic Segmentation for Disaster Response: Evaluating CNNs and Transformers for Flood Mapping

  • Mar 18, 2025
  • Torit Chakraborty +1
  • PDF
  • Research Article
  • Citations44

The effect of activation functions on accuracy, convergence speed, and misclassification confidence in CNN text classification: a comprehensive exploration

  • Jun 24, 2023
  • The Journal of Supercomputing
  • Rebecca H K Emanuel +3
  • Book Chapter
  • Citations3

Toward Secure and Efficient CNN Recognition with Different Activation and Optimization Functions

  • Jan 01, 2023
  • Fathi Amsaad +8
  • PDF
  • Research Article
  • Citations1

The structural tuning of the convolutional neural network forspeaker identification in mel frequency cepstrumcoefficients space

  • Jul 03, 2023
  • Herald of Advanced Information Technology
  • Anastasiia D Matychenko +1
  • Research Article
  • Citations2

Enhancing emotion classification on the ISEAR dataset using fine-tuning and data augmentation with hybrid transformer models

  • Sep 22, 2025
  • PeerJ Computer Science
  • Uzair Muhammad +5
  • Research Article

Advanced Machine Learning Techniques for Pest Classification: A Comprehensive Review

  • Apr 01, 2026
  • Computational Intelligence
  • C N Hettiarachchi +1
  • Research Article
  • Citations4

Real Time Drone Detection Based on Acoustics Using Hybrid Deep Learning Models

  • May 30, 2025
  • Journal of Internet Services and Information Security
  • Shahad W Jasim +1
  • Research Article
  • Citations20

Detection of aspiration from images of a videofluoroscopic swallowing study adopting deep learning.

  • Feb 08, 2023
  • Oral Radiology
  • Yukihiro Iida +5
  • Research Article
  • Citations2

A Gas Classification Algorithm of Electronic Noses Based on Convolutional Spiking Neural Network

  • May 30, 2021
  • ECS Meeting Abstracts
  • Yizhou Xiong +4
  • Preprint Article

Downscaling a heat stress index in southern South America using deep-learning 

  • Mar 18, 2025
  • Candela Sol Glatstein +2
  • Research Article
  • Citations10

Robust optimization of convolutional neural networks with a uniform experiment design method: a case of phonocardiogram testing in patients with heart diseases

  • Nov 01, 2021
  • BMC Bioinformatics
  • Wen-Hsien Ho +7
  • Conference Article
  • Citations3

Empirical Evaluation on Robustness of Deep Convolutional Neural Networks Activation Functions Against Adversarial Perturbation

  • Nov 01, 2018
  • Jiawei Su +2
  • Book Chapter
  • Citations66

Convolutional Neural Network for Hybrid fNIRS-EEG Mental Workload Classification

  • Jun 12, 2019
  • Marjan Saadati +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.