Visualizing the hill of vision in 3D using the free programming language ‘R’

Dear Editor, In recent years, technological innovations have enabled us to visualize complex 3D information on 2D video displays. Such developments have made a sizeable impact in ophthalmological imaging devices, including in optical coherence tomography and Heidelberg retina tomography (Heidelberg Engineering GmbH, Heidelberg, Germany). However, representations of the hill of vision (HOV) have remained planar, such as the grayscale images displayed by the Humphrey Visual Field Analyzer II (HFA II; Carl Zeiss Meditec, Inc, Dublin, CA, USA) shown in Fig. 1. To overcome this shortcoming, we have written a program in R [1] to display an interactive 3D graphical view of the HOV. R is an open-source, free software environment for statistical computing and graphics. For clinicians, a 3D representation of the HOV offers a clear snapshot of a patient’s visual field (VF) with more detail than possible in a grayscale image. In particular, in glaucoma patients, the edge and shape of a scotoma is immediately apparent in 3D. The rendered 3D image can also be colour-coded to represent the depth of defect according to the significance of the total deviation (td) or pattern deviation (pd) value. Furthermore, the 3D image could be shown to the patient as a means to better illustrate their VF, and progression, since the 3D image is more intuitive than a grayscale image. VFs were measured using the HFA II, 24–2 test pattern, a stimulus size of Goldmann III, and the SITA standard strategy. Subsequently an R program was written to display the HOVs of these data, although the program can also readily accept data from different VF machines. The R code to produce a 3D HOV can be downloaded at http://www. staff.city.ac.uk/d.crabb/visual%20fields.html. A screenshot of a glaucomatous patient’s (patient ‘a’) 3D HOV (generated from the R code) is shown next to that of a normal subject for comparison in Fig. 1. In the R software, the 3D HOV is interactive and can be rotated and examined more closely using the computer mouse. In Fig. S1 (online supplemental material), a series of ten VFs from a glaucomatous patient (patient ‘b’) who developed glaucomatous progression are shown, while in Fig. S2 (online supplemental material) the associated pattern deviation values are plotted (http://www.staff.city.ac.uk/d.crabb/ visual%20fields.html). Furthermore, animations of the 3D HOV over several VF tests (i.e., over time) can be created using software such as ImageMagick [2] to visualize changes (progression) in the HOV. In conclusion, we have developed a simple program to easily obtain a 3D interactive HOV, which gives a clear Electronic supplementary material The online version of this article (doi:10.1007/s00417-011-1861-z) contains supplementary material, which is available to authorized users. R. Asaoka : R. A. Russell :D. F. Garway-Heath NIHR Biomedical Research Centre for Ophthalmology, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, UK