Nystagmus analysis tool
***********************

Software to accompany {Dunn MJ, Harris CM, Ennis FA, Margrain TH, Woodhouse JM, McIlreavy L, Erichsen JT (In Press as of Dec 2018)}

Code written by Matt J Dunn (note exceptions to this listed below). No aspects of this software may be reproduced without the written permission of the author.
The software is provided as-is, with no warranty. This software must not be used to inform a medical diagnosis.
The software provided is limited to demonstrating concepts outlined in the above paper. It is not a complete eye movement analysis suite and should not be treated as such.
For a more complete suite of tools for analysing eye movements (including nystagmus), please contact Matt.

This software uses two pieces of software written by others. Both authors have generously given their permission for the software to be packaged here.
1) sasg, written by Weiwei Dai {Dai, W., Selesnick, I., Rizzo, J.-R., Rucker, J., & Hudson, T. (2017)}
2) findsaccades_pander, using routines written by Tomasz Pander {Pander, T., Czaba?ski, R., Przyby?a, T., & Pojda-Wilczek, D. (2014)}

In addition, the function that calculates NOFF was co-authored by Joost Felius.

The above packages may not be reproduced without written permission from their respective authors.

** How to use **
Requires MATLAB (2018b or later), with the following toolboxes installed:
- Signal Processing
- Statistics and Machine Learning

Input eyetrace should be formatted in the same manner as the example data file (example_data.csv).
Alternatively, a CSV file with one or two columns of data may be supplied, where each column contains gaze position data. In this case, you will be asked to supply the tracker sampling rate when you launch an analysis.
If only one axis is supplied, there are assumed to be no eye movements in the orthogonal axis.

1) Run nystools.mlapp from MATLAB
2) Click 'Browse' to select your input data file
3) a) If your data are already calibrated, click on the 'Analyse nystagmus' tab, check the box labelled 'Input data are calibrated in degrees', and click 'Analyse' to analyse. If your data aren't calibrated, you can still analyse, but you will not have access to output metrics such as amplitude and velocity.
   b) If the data aren't calibrated, you can use the 'Determine point of regard' tab to find the ideal gaze position to be fed into a calibration algorithm. This is the specific algorithm presented in the paper.

Any settings changed in 'Preferences' apply to both other tabs.
If you choose 'Use cleanest' eye, the eye with the least dropped data samples will be used. This software does not attempt to combine data across both eyes.

You can adjust some other details, including default settings, by editing the file 'defaults.csv'.