Hande Argunsah

Hande Argunsah


Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 July 2018 in The Journal of Foot and Ankle Surgery
Verified
Journal Article
Closed access
HAHande Argunsah
Hande Argunsah
MBMehmed B. Bayram
Mehmed B. Bayram

20 February 2025 in BMC Medical Education
Verified
Journal Article
Open access
PDF file
HAHande Argunsah
Hande Argunsah
LALevent Altıntaş
Levent Altıntaş
MSMelike Sahiner
Melike Sahiner
Individual differences in learning preferences and cognitive strategies play a crucial role in shaping academic outcomes, emphasizing the need to customize educational approaches to meet diverse learner needs. This study explores the relationship between gaze behavior, learning style and academic performance in 20 sophomore Turkish medical students. Eye-tracking metrics, gaze duration, fixation count, fixation duration, and saccadic movements, were recorded using Tobii Pro Glasses 2 eye tracker during Trail Making, Visual Sustained Attention and the Stroop Tests and associated with the Felder-Soloman learning style and academic performance. Eye-tracking data revealed consistent patterns across tasks, with fixation percentages averaging 94% and saccadic movements accounting for 6%, suggesting uniform attention allocation. Pupil diameter variation did not significantly differ between tasks, implying similar cognitive demands across all tasks. Most of the participants demonstrated moderate-to-strong visual learning preferences, particularly females. Significant gender differences were observed in learning preferences and academic performance, with higher Grade Point Averages among female participants with stronger visual learning preferences. The study underscores gender-based differences in learning preferences and the alignment of these preferences with academic performance. The findings suggest the importance of tailoring educational strategies to support diverse learning needs, with a particular emphasis on visually engaging materials.

5 September 2024 in Computer Methods in Biomechanics and Biomedical Engineering
Verified
Journal Article
Closed access
EKEngin Kaya
Engin Kaya
HAHande Argunsah
Hande Argunsah
Machine learning (ML) has been used to predict lower extremity joint torques from joint angles and surface electromyography (sEMG) signals. This study trained three bidirectional Long Short-Term Memory (LSTM) models, which utilize joint angle, sEMG, and combined modalities as inputs, using a publicly accessible dataset to estimate joint torques during normal walking and assessed the performance of models, that used specific inputs independently plus the accuracy of the joint-specific torque prediction. The performance of each model was evaluated using normalized root mean square error (nRMSE) and Pearson correlation coefficient (PCC). Each model's median scores for the PCC and nRMSE values were highly convergent and the bulk of the mean nRMSE values of all joints were less than 10%. The ankle joint torque was the most successfully predicted output, having a mean nRMSE of less than 9% for all models. The knee joint torque prediction has reached the highest accuracy with a mean nRMSE of 11% and the hip joint torque prediction of 10%. The PCC values of each model were significantly high and remarkably comparable for the ankle (∼ 0.98), knee (∼ 0.92), and hip (∼ 0.95) joints. The model obtained significantly close accuracy with single and combined input modalities, indicating that one of either input may be sufficient for predicting the torque of a particular joint, obviating the need for the other in certain contexts.

17 December 2020 in International Journal of Performance Analysis in Sport
Verified
Journal Article
Closed access
HAHande Argunsah
Hande Argunsah
Begüm YalçınBegüm Yalçın
Begüm Yalçın
Biofeedback is used for enhancing performance through providing real-time stimulus to the individual during physical activities. This randomised controlled trial aimed investigating the changes in physiological and kinematic variables of treadmill running in response to biofeedback. 24 age-matched, healthy, and recreationally active participants were randomly assigned to biofeedback and no-biofeedback groups. During 15-min treadmill running, group-based knee ROM, COM and heart rate were collected and analysed. Heart rate change was statistically significant for no-biofeedback group (156.60 ± 5.18, 160.20 ± 5.22 and 163.60 ± 5.93 F (2, 9) = 13.878, p < 0.001). No such change was found for biofeedback group (152.20 ± 6.63, 152.10 ± 6.12 and 155.30 ± 6.22 F (2, 9) = 2.787, p = 0.088). According to the Pearson Correlation Coefficient and RMSE techniques knee movement was highly correlated (initial-mid R2 = 0.995, RMSE = 1.558; initial-final R2 = 0.994, RMSE = 1.407) for biofeedback group compared to no biofeedback group (initial-mid R2 = 0.989, RMSE = 1.988; initial-final R2 = 0.985, RMSE = 4.978). Heart rate and COM variability were minimised; correspondingly, consistent knee movement was obtained in biofeedback group. Future research should determine if muscle involvement in motion and fatigue are regulated with biofeedback.

1 July 2025 in Computers in Biology and Medicine
Verified
Journal Article
Open access
EKEngin Kaya
Engin Kaya
HŞHülya Şirzai
Hülya Şirzai
GYGüneş Yavuzer
Güneş Yavuzer
HAHande Argunsah
Hande Argunsah
This study explored the integration of markerless motion capture and deep learning to classify knee osteoarthritis severity based on gait kinematics, providing an alternative to traditional assessment methods. We employed a Long Short-Term Memory Fully Convolutional Network model to analyze gait patterns and distinguish severity levels corresponding to Kellgren-Lawrence grades. Two data-splitting strategies were compared: random splitting, which allowed learning from a diverse set of individuals, and subject-based splitting, which assessed the model's ability to generalize to unseen patients. The model achieved high classification performance in the random split (accuracy = 0.91), but performance declined in the subject-based split (accuracy = 0.76), indicating challenges in generalizing across individuals. Severe and healthy groups were well classified, while early and moderate severity groups showed higher misclassification rates due to overlapping gait characteristics. Our findings highlight the potential of gait-based deep learning models for automated severity classification, offering a scalable and accessible alternative to conventional assessments. However, challenges related to inter-subject variability suggest the need for enhanced feature extraction, multimodal data integration, and domain adaptation to improve generalizability. Future research should focus on longitudinal data to assess the model's predictive capability for disease progression and treatment outcomes.

1 December 2014 in Clinical Biomechanics
Verified
Journal Article
Closed access
HAHande Argunsah
Hande Argunsah
CCChih-Hao Chien
Chih-Hao Chien
BDBrian L. Davis
Brian L. Davis