Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models

Abstract This study presents a deep learning-based approach for the automated detection of Quiet Eye (QE) durations from electrooculography (EOG) signals in archery. QE—the final fixation or tracking of the gaze before executing a motor action—is a critical factor in precision sports. Traditional de...

Full description

Saved in:
Bibliographic Details
Main Authors: Fatma Söğüt, Hüseyin Yanık, Evren Değirmenci, İnci Kesilmiş, Ülkü Çömelekoğlu
Format: Article
Language:English
Published: BMC 2025-08-01
Series:BMC Sports Science, Medicine and Rehabilitation
Subjects:
Online Access:https://doi.org/10.1186/s13102-025-01284-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849343740347416576
author Fatma Söğüt
Hüseyin Yanık
Evren Değirmenci
İnci Kesilmiş
Ülkü Çömelekoğlu
author_facet Fatma Söğüt
Hüseyin Yanık
Evren Değirmenci
İnci Kesilmiş
Ülkü Çömelekoğlu
author_sort Fatma Söğüt
collection DOAJ
description Abstract This study presents a deep learning-based approach for the automated detection of Quiet Eye (QE) durations from electrooculography (EOG) signals in archery. QE—the final fixation or tracking of the gaze before executing a motor action—is a critical factor in precision sports. Traditional detection methods, which rely on expert evaluations, are inherently subjective, time-consuming, and inconsistent. To overcome these limitations, EOG data were collected from 10 licensed archers during controlled shooting sessions and preprocessed using a wavelet transform and a Butterworth bandpass filter for noise reduction. We implemented and compared a traditional model (SVM) and five deep learning models—CNN + LSTM, CNN + GRU, Transformer, UNet, and 1D CNN—for QE detection. The CNN + LSTM model achieved the highest accuracy (95%), followed closely by CNN + GRU (93%), demonstrating superior performance in capturing both spatial and temporal dependencies in the EOG signals. Although Transformer-based and UNet models performed competitively, they exhibited lower precision in distinguishing QE periods. The performance of the traditional model was inferior to deep learning approaches. These results indicate that deep learning provides an effective and scalable solution for objective QE analysis, substantially reducing the dependence on expert annotations. This automated approach can enhance sports training by offering real-time, data-driven feedback to athletes and coaches. Furthermore, the methodology holds promise for broader applications in cognitive and motor skill assessments across various domains. Future work will focus on expanding the dataset, enabling real-time deployment, and evaluating model generalizability across different skill levels and sports disciplines.
format Article
id doaj-art-dc7bfca3dc3e4935b3a8bca6bc21a14b
institution Kabale University
issn 2052-1847
language English
publishDate 2025-08-01
publisher BMC
record_format Article
series BMC Sports Science, Medicine and Rehabilitation
spelling doaj-art-dc7bfca3dc3e4935b3a8bca6bc21a14b2025-08-20T03:42:52ZengBMCBMC Sports Science, Medicine and Rehabilitation2052-18472025-08-0117111910.1186/s13102-025-01284-2Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning modelsFatma Söğüt0Hüseyin Yanık1Evren Değirmenci2İnci Kesilmiş3Ülkü Çömelekoğlu4Vocational School of Health Service, Mersin UniversityInformation Systems and Technologies, Mersin UniversityDepartment of Electrical and Electronics Engineering, Faculty of Engineering, Mersin UniversityDepartment of Coaching Education, Faculty of Sports Science, Mersin UniversityDepartment of Biophysics, Faculty of Medicine, Mersin UniversityAbstract This study presents a deep learning-based approach for the automated detection of Quiet Eye (QE) durations from electrooculography (EOG) signals in archery. QE—the final fixation or tracking of the gaze before executing a motor action—is a critical factor in precision sports. Traditional detection methods, which rely on expert evaluations, are inherently subjective, time-consuming, and inconsistent. To overcome these limitations, EOG data were collected from 10 licensed archers during controlled shooting sessions and preprocessed using a wavelet transform and a Butterworth bandpass filter for noise reduction. We implemented and compared a traditional model (SVM) and five deep learning models—CNN + LSTM, CNN + GRU, Transformer, UNet, and 1D CNN—for QE detection. The CNN + LSTM model achieved the highest accuracy (95%), followed closely by CNN + GRU (93%), demonstrating superior performance in capturing both spatial and temporal dependencies in the EOG signals. Although Transformer-based and UNet models performed competitively, they exhibited lower precision in distinguishing QE periods. The performance of the traditional model was inferior to deep learning approaches. These results indicate that deep learning provides an effective and scalable solution for objective QE analysis, substantially reducing the dependence on expert annotations. This automated approach can enhance sports training by offering real-time, data-driven feedback to athletes and coaches. Furthermore, the methodology holds promise for broader applications in cognitive and motor skill assessments across various domains. Future work will focus on expanding the dataset, enabling real-time deployment, and evaluating model generalizability across different skill levels and sports disciplines.https://doi.org/10.1186/s13102-025-01284-2Quiet eyeElectrooculographyWavelet transformConvolutional neural networksLong-short term memoryTransformer
spellingShingle Fatma Söğüt
Hüseyin Yanık
Evren Değirmenci
İnci Kesilmiş
Ülkü Çömelekoğlu
Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
BMC Sports Science, Medicine and Rehabilitation
Quiet eye
Electrooculography
Wavelet transform
Convolutional neural networks
Long-short term memory
Transformer
title Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
title_full Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
title_fullStr Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
title_full_unstemmed Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
title_short Automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
title_sort automated detection of quiet eye durations in archery using electrooculography and comparative deep learning models
topic Quiet eye
Electrooculography
Wavelet transform
Convolutional neural networks
Long-short term memory
Transformer
url https://doi.org/10.1186/s13102-025-01284-2
work_keys_str_mv AT fatmasogut automateddetectionofquieteyedurationsinarcheryusingelectrooculographyandcomparativedeeplearningmodels
AT huseyinyanık automateddetectionofquieteyedurationsinarcheryusingelectrooculographyandcomparativedeeplearningmodels
AT evrendegirmenci automateddetectionofquieteyedurationsinarcheryusingelectrooculographyandcomparativedeeplearningmodels
AT incikesilmis automateddetectionofquieteyedurationsinarcheryusingelectrooculographyandcomparativedeeplearningmodels
AT ulkucomelekoglu automateddetectionofquieteyedurationsinarcheryusingelectrooculographyandcomparativedeeplearningmodels