Plug and play perimetry: Evaluating the use of a self-calibrating digital display for screen-based threshold perimetry

Abstract

This study evaluated the feasibility of using a 'self-calibrating' display (EIZO CG277) to perform screenbased threshold perimetry. Such displays incorporate their own integrated photometer, so could potentially be used 'straight out of the box', without the need for time-consuming and costly luminance calibration by skilled experts. Concerns remain, however, due to the fact that the internal calibration of such devices is imperfect (i.e., is limited to a single screen location only) and due to lingering doubts regarding the accuracy of screen-based perimetry in general. To evaluate such a system, automated static threshold perimetry was performed in thirty-two normal-sighted adults. In one condition, participants performed a novel screen-based perimetry test, for which the screen was extensively calibrated using traditional photometric techniques/equipment. In a second condition, the same test was performed, but the display was calibrated using only the screen's integrated photometer (and assuming uniformity across the display). For reference, participants also completed a traditional visual-field assessment using a Humphrey Field Analyzer (HFA). All three tests were performed twice to assess test-retest repeatability (six tests total). The results showed no differences when comparing screen-based perimetric measurements made with internal self-calibration vs full manual calibration (either in terms of mean sensitivity, pointwise sensitivity, test-retest repeatability, or test duration). Furthermore, the accuracy and precision of both were indistinguishable from the current gold standard (HFA), although the HFA was approximately two minutes (~30%) faster. These results indicate that self-calibrating commercial monitors can be used to perform screen-based perimetry almost as well as current clinical devices, and without the need for any specialized knowledge or equipment to setup or maintain. This could facilitate perimetric testing in currently hard-to-reach settings, such as community centers, stroke wards, homes, rural locations, or developing countries.

An Open-source Static Threshold Perimetry Test Using Remote Eye-tracking (Eyecatcher): Description, Validation, and Preliminary Normative Data

“…However, we have demonstrated subsequently that measurements from this inbuilt photometer can be substituted for a formal photometric calibration, without any appreciable loss of accuracy or reliability in visual field measurements. 58 Furthermore, we have made the necessary Matlab code for controlling the photometer available online ( https://github.com/petejonze/myEIZOSensor ), allowing calibration to be performed automatically (e.g., overnight or before the first test of each day). Finally, we have also shown that this monitor (which comes with a factory-calibrated certificate of uniformity) remains photometrically stable over its spatial extent for at least 20 months.…”

“…Finally, we have also shown that this monitor (which comes with a factory-calibrated certificate of uniformity) remains photometrically stable over its spatial extent for at least 20 months. 58 Taken together, this makes the prospect of a simple plug-and-play perimeter—one that combines open source software with ordinary, commercially available hardware—a genuine, pragmatic possibility.…”

Differences in Contrast Reproduction between Electronic Devices for Visual Assessment: Clinical Implications

“…Visual specialists could provide visual tests but also the characterized devices to ensure correct performance of the patient. Future solutions could involve the development of hardware with integrated photometers for self-calibrating, [35] but the cost is still high, and this solution is not affordable in handheld devices.…”

The Human Touch: Using a Webcam to Autonomously Monitor Compliance During Visual Field Assessments

“…It was modified in the present work to more closely mimic conventional static threshold perimetry, most notably by employing a Zippy Estimation by Sequential Testing (ZEST) thresholding algorithm, 15 a central fixation cross, and a button press response. The software was implemented using Psychtoolbox 3, 16 and we used bit stealing to ensure >10-bit luminance precision, 17 with extensive photometric calibration to ensure stimulus uniformity across the display (see Kyu Han and Jones 18 for technical details regarding the calibration method). Unlike conventional perimetry, participants received visual Figure 1.…”