Japan Association for Medical Informatics

The 40th Joint Conference on Medical Informatics / APAMI2020

The 40th Joint Conference on Medical Informatics / APAMI2020

Nov 18, 2020 - Nov 23, 2020
Actcity Hamamatsu

[AP2-E2-4-01] Effects of the Sampling Frequency Change in Eye Movement Analysis Using Deep Convolutional Neural Network: Comparison with Other Analyses by Open Annotated Gaze Data

*Takayoshi Terashita1, Tetsuo Sato1, Shoko Tsutsumi1, Mitsuru Sato1, Toshihiro Ogura1, Kunio Doi1,2 (1. Graduate School of Radiological Technology, Gunma Prefectural College of Health Sciences, Japan, 2. Department of Radiology, University of Chicago, USA)

Eye Movement Analysis, Sampling Frequency of Gaze Data, Deep Convolutional Neural Network

In our previous work, an eye movement analysis using a deep convolutional neural network was proposed. However, the low sampling frequency decreased the detection accuracy of eye events. The purpose of this study is to evaluate the effects of the sampling frequency change in our eye movement analysis. The gaze data of Lund University were used as open annotated data. It was then clarified that our method had high accuracy rates of approximately 90%. Moreover, the change in the accuracy was continuously high regardless of the sampling frequency in comparison to other methods.