Parameters for Statistical Evaluation of Time to Fixate Efectiveness for Assessment of Fitness to Drive

This repository contains a table with parameters (including time to fixate - TTF parameter) and an R script for statistical analysis. This is supplementary material for the paper titled "Effectiveness of a Time to Fixate for Fitness to Drive Evaluation in Neurological Patients" and authored by Nadica Miljković and Jaka Sodnik (published in Behavior Research Methods and previously shared on arXiv).

TTF parameter was calculated in overall 56 patients during selected scenario with pedestrian collision in a driving simulator produced by Nervtech. Together with other parameters, ST parameters are stored in tableParametersAll.csv, while R programming code for statistical analysis of all parameters is placed in statisticalAnalysisAll.R.

Dataset contents

tableParametersAll.csv, table with parameters, csv (comma-separated values) format

statisticalAnalysisAll.R, code in R programming language for statistical analysis

Table with parameters has the following structure

column - no which is ordinary number in consecutive order from 1 to 56

column - id presents an internal patient's id

column - ttf presents TTF parameter in ms

column - fitness presents a categorical variable and can be either fit-, unfit-, or conditionally-fit-to-drive (cond fit)

column - speed at the collision onset in km/h

column - ttc presents time-to-collision in s

column - manual_correction is categorical variable: 0 means that no manual correction was required for ST calculation, while 1 means that manual correction was required

column - igd presents initial gaze distance in pixels

column - R1 presents the first measurement of perception response time (PRT)

column - R2 presents the second measurement of PRT

column - R3 presents the third measurement of PRT

Missing data are presented with NA (Not Available).

NOTE: Python code for ST calculation and sample eye tracker video are available on GitHub repository https://github.com/NadicaSm/Time-To-Fixate-Calculation-from-the-Eye-Tracker-Videos under GNU GPL license and released on Zenodo with doi (https://doi.org/10.5281/zenodo.6560419).

If you find these parameters and R code useful for your own research and teaching class, please cite the following references:

Miljković, N., & Sodnik, J. (2023). NadicaSm/Time-To-Fixate-Calculation-from-the-Eye-Tracker-Videos: v2. [Software code], Zenodo. https://doi.org/10.5281/zenodo.6560419

Miljković, N., & Sodnik, J. (2023). Effectiveness of a time to fixate for fitness to drive evaluation in neurological patients. Behavior Research Methods. https://doi.org/10.3758/s13428-023-02177-3

Motnikar, L., Stojmenova, K., Štaba, U. Č., Klun, T., Robida, K. R., & Sodnik, J. (2020). Exploring driving characteristics of fit-and unfit-to-drive neurological patients: A driving simulator study. Traffic Injury Prevention, 21(6), 359-364. https://doi.org/10.1080/15389588.2020.1764547

Acknowledgements

J.S. kindly acknowledges University Rehabilitation Institute Soča employees and the Nervtech team. Authors gratefully appreciate the support from Nenad B. Popović, PhD from University of Belgrade – School of Electrical Engineering for his valuable assistance in design of illustrations and for provided feedback for the initial manuscript structure. Also, both Authors thank Nebojša Jovanović, MSc from University of Belgrade - School of Electrical Engineering for his kind contribution to earlier stages of the project, especially for his work on developing Python code to capture time to fixate parameter. Last but not least, we are very thankful to Damjan Krstajić, founder and director of the Research Centre for Cheminformatics for his precious advices on statistical analysis in a retrospective study and to student Gregor Kovač from Faculty of Electrical Engineering, University of Ljubljana for his diligent work on YOLO application in driving simulation.