Movatterモバイル変換

rPACI

This Github repository is devoted to the development of the R packagerPACI (R Placido Analysis of Corneal Irregularity),also available in thethe CRANrepository.

Check out therPACIshiny app.

## Contributors The contributors/developers of this project are: *Darío Ramos-López, University of Almería, Spain. dramoslopez at ual.es *Ana D. Maldonado, University of Almería, Spain. ana.d.maldonado atual.es

To contribute to the package, report a bug, or any other issues,please use ourGitHubrepository rPACI.

## rPACI aim and description

This package performs analyses of corneal data obtained from aPlacido disk corneal topographer, with the calculation of the Placidoirregularity indices and the posterior analysis. The package is intendedto be easy to use by a practitioner, providing a simple interface andyielding easily interpretable results.

A corneal topographer is an ophthalmic clinical device that obtainsmeasurements in the cornea (the anterior part of the eye). A Placidodisk corneal topographer makes use of the Placido diskRowsey etal. (1981), which produce a circular pattern of measurement nodes.The raw information measured by such a topographer is used bypractitioners to analyze curvatures, to study optical aberrations, or todiagnose specific conditions of the eye (e.g. keratoconus, an importantcorneal disease).

The rPACI package allows the calculation of the corneal irregularityindices described inCastro-Luna etal. (2020),Ramos-Lopez etal. (2013), andRamos-Lopez etal. (2011). It provides a simple interface to read cornealtopography data files as exported by a typical Placido disk topographer,to compute the irregularity indices mentioned before, and to displaysummary plots that are easy to interpret for a clinician.

We encourage the users to get started by reading thepackage’s vignettes. We recommend beginning with ‘Workflow of the rPACIpackage’ and then consulting the other vignettes for additionaldetails.

## Corneal topography

The cornea is the external (outermost) part of the eye, and it isresponsible for most of the eye’s refractive power. Given its importancein vision, its analysis is especially important in clinicalophthalmology, since it can be affected by several diseases (e.g.,keratoconus). Besides, refractive surgery to correct vision defects actson the cornea, changing its curvature. Thus, the correct analysis anddiagnosis of the cornea are crucial for practitioners.

Corneal topography refers to the measurement of the corneal surfaceshape using an automated device, which is called a corneal topographer.Most corneal topographers used in clinical practice rely on the Placidodisk technology (either alone or in combination with others, such as theScheimpflug camera).

For a deeper insight about these and other related topics (cornealtopography, Placido disk technology, keratoconus, …), we refer thereader to the rPACI’s vignette entitled ‘Corneal topographers and dataformats’, or to the websites byUniversityof Iowa Health Care andAmerican Academy ofOphthalmology.

## rPACI’s file formats

TherPACI package is able to read corneal topographyfiles in two different formats, by now. It will possibly be expanded inthe future allowing formats used by other manufacturers. Both formatsrely on plain text files. The first file format is basically the oneemployed byCSO topographers, butwith some more flexibility. The second file format follows the structureused internally byrPACI. It has been developed tomanage directly the datasets handled by the package, allowing to saveand read data easily. These two formats are described in the nextsubsections.

You can find more details about supported formats in the vignette‘Corneal topographers and data formats’.

### File format by CSO topographers

A corneal topography file exported by CSO topographers can be readdirectly inrPACI. The raw measurement file should havethe following structure:

• A header (of any length, possibly missing) with meta-data: patientdata, optical measurements, date of exploration, etc. Its size dependson the device. For CSO topographers, the header typically has 24 lines.The function used inrPACI to read the files,readCSO, is able to detect the header lines (those that donot contain only real numbers) and to skip them.
• A list of size N_R x N_A with the rho coordinates (distance to theorigin in polar coordinates, measured in mm) of the digitized points, atequally spaced angles (theta, the argument in polar coordinates). Thetheta coordinate is inferred from the position (assuming a uniformdistribution), as it is not explicitly given in the data file.
• A list of size N_R x N_A with the z coordinate, which can bealtitude (elevation) or curvature, depending on the topographer and theexportation settings. These data are post-processed by the device usingdifferent algorithms, not directly measured, and the z values are notused inrPACI.

### File format by rPACI

rPACI also reads another format of cornealtopography files. This format was designed to follow the internaldatasets’ structure. A file formatted in this manner is obtained whenusing the writing functions of the package. It should have the followingstructure:

• A header (of any length, possibly missing) with meta-data. In casethat the saved dataset was obtained by simulation, the header includesthe list of simulation parameters. Else (if the dataset was originallyread from a real measurement file in other formats), no header will bepresent.
• A block of three separated columns (with some specified characterseparator), according to the usual format used byrPACI, i.e., a list with three columns (x and ycoordinates of each point, and its ring index) and a row per datapoint.

## Placido irregularity indices

TherPACI package is able to read corneal topographyfiles and then analyze them. This analysis includes the computation ofsome Placido irregularity indices. The indices computed byrPACI allow to discriminate between normal andirregular corneas. Most of them were introduced and validated with realdatasets in the scientific publicationsRamos-Lopez etal. (2013), andRamos-Lopez etal. (2011). An additional index, based on a naive Bayes classifier,was proposed later, and tested with a different database, inCastro-Luna etal. (2020). In these papers, all indices demonstrated a goodsensitivity for the detection of keratoconus, a corneal disease.

Consult more information about the Placido irregularity indices inthe vignette ‘Mathematical definition of the Placido irregularityindices’.

## Analyzing datasets

The package includes several easy-to-use functions that allowanalyzing data from a single file or dataset, or from a folder (withseveral files either corresponding to the same or to differentpatients). All these functions include easily interpretable plots andyield well-formatted data.frames to facilitate the posterior analysis ofresults.

You can find more details about supported formats in the vignette‘Workflow of the rPACI package’.