TY - GEN
T1 - An Eye-Tracking Based Machine Learning Model Towards the Prediction of Visual Expertise for Electrocardiogram Interpretation
AU - Sqalli, Mohammed Tahri
AU - Al-Thani, Dena
AU - Elshazly, Mohamed B.
AU - Al-Hijji, Mohammed
AU - Alahmadi, Alaa
AU - Houssaini, Yahya Sqalli
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2023
Y1 - 2023
N2 - The electrocardiogram, known as the ECG or EKG, is considered among the mostly used medical diagnostic tests worldwide. Despite the test’s prevalence in the healthcare sector, there still exist gaps into training medical practitioners become skilled and efficient ECG interpreters. Moreover, this also brings the challenge of assessing the expertise of those practitioners. This is primarily due to the difficulty of assessing visual expertise. Visual expertise is the skill of interpreting images relating to a certain technical field. Due to the limited quantitative research methodologies that could not capture this subtle skill during the previous two decades, a limited number of models are being conceptualized and assessed. In addition, automated ECG interpretation models based on artificial intelligence are still not accurate enough to be fully deployed in the medical field. This therefore leaves only one choice, which is to focus on improving on methodologies to train and assess medical practitioners’ visual expertise. This approach will contribute towards increasing the accuracy of ECG interpretations within medical institutions by forming competent medical staff. In this paper, we present a road map for the development of an eye-tracking based machine learning model that leads towards the prediction of visual expertise within medical practitioners. To develop the model, we built on top of our previously conducted research that aimed at understanding the differences in visual patterns within medical practitioners with different expertise levels. The developed model could predict the expertise level of the ECG interpreter with an accuracy of 94.08%. This is thanks to the eye movement patterns of the participant.
AB - The electrocardiogram, known as the ECG or EKG, is considered among the mostly used medical diagnostic tests worldwide. Despite the test’s prevalence in the healthcare sector, there still exist gaps into training medical practitioners become skilled and efficient ECG interpreters. Moreover, this also brings the challenge of assessing the expertise of those practitioners. This is primarily due to the difficulty of assessing visual expertise. Visual expertise is the skill of interpreting images relating to a certain technical field. Due to the limited quantitative research methodologies that could not capture this subtle skill during the previous two decades, a limited number of models are being conceptualized and assessed. In addition, automated ECG interpretation models based on artificial intelligence are still not accurate enough to be fully deployed in the medical field. This therefore leaves only one choice, which is to focus on improving on methodologies to train and assess medical practitioners’ visual expertise. This approach will contribute towards increasing the accuracy of ECG interpretations within medical institutions by forming competent medical staff. In this paper, we present a road map for the development of an eye-tracking based machine learning model that leads towards the prediction of visual expertise within medical practitioners. To develop the model, we built on top of our previously conducted research that aimed at understanding the differences in visual patterns within medical practitioners with different expertise levels. The developed model could predict the expertise level of the ECG interpreter with an accuracy of 94.08%. This is thanks to the eye movement patterns of the participant.
KW - ECG
KW - ECG interpretation
KW - Electrocardiogram
KW - Eye-tracking
KW - Human-computer interaction
UR - http://www.scopus.com/inward/record.url?scp=85180814133&partnerID=8YFLogxK
U2 - 10.1007/978-981-16-6775-6_25
DO - 10.1007/978-981-16-6775-6_25
M3 - Conference contribution
AN - SCOPUS:85180814133
SN - 9789811667749
T3 - Lecture Notes in Electrical Engineering
SP - 305
EP - 315
BT - Medical Imaging and Computer-Aided Diagnosis - Proceedings of 2022 International Conference on Medical Imaging and Computer-Aided Diagnosis MICAD 2022
A2 - Su, Ruidan
A2 - Zhang, Yudong
A2 - Liu, Han
A2 - F Frangi, Alejandro
PB - Springer Science and Business Media Deutschland GmbH
T2 - International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2022
Y2 - 20 November 2022 through 21 November 2022
ER -