perimetry

R package for visual field analysis.

perimetry will simplify visual field analysis based on current concepts and understanding of visual field data.

Features

Installation

NOT YET ON CRAN

Development version on github (here!)

# install.packages("devtools")
devtools::install_github("tjebo/perimetry")

Example

library(perimetry)

tidyfields

A framework for the analysis of visual field (perimetry) data in R.

Ideas

1. Input

2. Central object: tidyfield object

str(tidyfield_mock)
#> tibble [1 × 24] (S3: tbl_df/tbl/data.frame)
#>  $ Device             : chr "[MAIA / ...]"
#>  $ DeviceID           : chr "[ID]"
#>  $ PatientID          : chr "[PID]"
#>  $ Lastname           : chr "[Lastname]"
#>  $ Firstname          : chr "[Firstname]"
#>  $ DOB                : chr "[DOB]"
#>  $ Age                : chr "[Age]"
#>  $ DOE                : chr "[DOE, format: 2005-10-21 18:47:22]"
#>  $ Background         : chr "[1.27 cd/m2]"
#>  $ StimulusSize       : chr "[G3]"
#>  $ StimulusColor      : chr "[white]"
#>  $ StimulusDuration   : chr "[200]"
#>  $ Staircase          : chr "['4-2']"
#>  $ FoveaPosition_X    : chr "[]"
#>  $ FoveaPosition_Y    : chr "[]"
#>  $ ONHDPosition_X     : chr "[]"
#>  $ ONHDPosition_Y     : chr "[]"
#>  $ AverageSensitivity : chr "[]"
#>  $ AverageReactionTime: chr "[]"
#>  $ BCEA95%            : chr "[]"
#>  $ BCEA63%            : chr "[]"
#>  $ FixationData       : chr "[nested tibble]"
#>  $ blindspot          :List of 1
#>   ..$ : tibble [1 × 1] (S3: tbl_df/tbl/data.frame)
#>   .. ..$ BlindSpotTrials: spec_tbl_df [1 × 7] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
#>   .. .. ..$ ID            : num 0
#>   .. .. ..$ X             : num 14
#>   .. .. ..$ Y             : num 2
#>   .. .. ..$ FinalThreshold: num 26
#>   .. .. ..$ Intensities   : chr "c(10,10,10,10)"
#>   .. .. ..$ Responses     : chr "c(0,0,0,1)"
#>   .. .. ..$ ResponseTimes : chr "c(NA, NA, NA, NA)"
#>   .. .. ..- attr(*, "spec")=List of 3
#>   .. .. .. ..$ cols   :List of 7
#>   .. .. .. .. ..$ ID            : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ X             : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ Y             : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ FinalThreshold: list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ Intensities   : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. .. ..$ Responses     : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. .. ..$ ResponseTimes : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. ..$ default: list()
#>   .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_guess" "collector"
#>   .. .. .. ..$ skip   : int 1
#>   .. .. .. ..- attr(*, "class")= chr "col_spec"
#>  $ testpoints         :List of 1
#>   ..$ : tibble [1 × 1] (S3: tbl_df/tbl/data.frame)
#>   .. ..$ Testpoints: spec_tbl_df [1 × 9] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
#>   .. .. ..$ ID            : num 1
#>   .. .. ..$ X             : num 0
#>   .. .. ..$ Y             : num 0
#>   .. .. ..$ FinalThreshold: num 26
#>   .. .. ..$ Intensities   : chr "c(32,28,24,26)"
#>   .. .. ..$ Responses     : chr "c(0,0,1,1)"
#>   .. .. ..$ ResponseTimes : chr "c(NA, NA, NA, NA)"
#>   .. .. ..$ fixated       : chr "c(NA, NA, NA, NA)"
#>   .. .. ..$ prior         : chr "Nested table with pdf for ZEST"
#>   .. .. ..- attr(*, "spec")=List of 3
#>   .. .. .. ..$ cols   :List of 9
#>   .. .. .. .. ..$ ID            : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ X             : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ Y             : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ FinalThreshold: list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_double" "collector"
#>   .. .. .. .. ..$ Intensities   : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. .. ..$ Responses     : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. .. ..$ ResponseTimes : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. .. ..$ fixated       : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. .. ..$ prior         : list()
#>   .. .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_character" "collector"
#>   .. .. .. ..$ default: list()
#>   .. .. .. .. ..- attr(*, "class")= chr [1:2] "collector_guess" "collector"
#>   .. .. .. ..$ skip   : int 1
#>   .. .. .. ..- attr(*, "class")= chr "col_spec"

3. “Mind the gap” function

Function that completes inputs, that are not stored in the original device output, but that are ‘imputable’

E.g., Response [y/n] not saved in MAIA files, but imputable from the staircase values

4. Coordinate system

Fixed XY - xy converted to deg. in visual field space - Ideal for binocular anlyses

Flipped XY (“left-eye convention”) - Ideal for normal data comparison

Maybe, both should be included in tidyfield object Some convention is needed throughout the package, i.e., each X and Y should have a suffix that specifies the meaning (e.g., “Yvsdeg” for visual field space and degrees )

Normal data

• Systematic comparison of spatial interpolation techniques necessary

Output statistics

Spatially-ignorant metrics (will be spatially-weighted) - Mean Sensitivity [dB] - Mean Deviation [dB] - PSD [dB] - Mean Loss [dB] - Sqrt. of Loss Variance [dB]

Spatially-adjusted metrics - Vol. [dB*sr] - Loss [dB*sr] - PSD [dB*sr]