Patient-attentive sequential strategy for perimetry-based visual field acquisition

TORONTO: A trial-oriented multidimensional psychometric testing algorithm

Bayesian adaptive methods for sensory threshold determination were conceived originally to track a single threshold. When applied to the testing of vision, they do not exploit the spatial patterns that underlie thresholds at different locations in the visual field. Exploiting these patterns has been recognized as key to further improving visual field test efficiency. We present a new approach (TORONTO) that outperforms other existing methods in terms of speed and accuracy. TORONTO generalizes the QUEST/ZEST algorithm to estimate simultaneously multiple thresholds. After each trial, without waiting for a fully determined threshold, the trial-oriented approach updates not only the location currently tested but also all other locations based on patterns in a reference data set. Since the availability of reference data can be limited, techniques are developed to overcome this limitation. TORONTO was evaluated using computer-simulated visual field tests: In the reliable condition (false positive [FP] = false negative [FN] = 3%), the median termination and root mean square error (RMSE) of TORONTO was 153 trials and 2.0 dB, twice as fast with equal accuracy as ZEST. In the FP = FN = 15% condition, TORONTO terminated in 151 trials and was 2.2 times faster than ZEST with better RMSE (2.6 vs. 3.7 dB). In the FP = FN = 30% condition, TORONTO achieved 4.2 dB RMSE in 148 trials, while all other techniques had > 6.5 dB RMSE and terminated much slower. In conclusion, TORONTO is a fast and accurate algorithm for determining multiple thresholds under a wide range of reliability and subject conditions.

Improving the robustness of the Sequentially Optimized Reconstruction Strategy (SORS) for visual field testing

Perimetry, or visual field test, estimates differential light sensitivity thresholds across many locations in the visual field (e.g., 54 locations in the 24-2 grid). Recent developments have shown that an entire visual field may be relatively accurately reconstructed from measurements of a subset of these locations using a linear regression model. Here, we show that incorporating a dimensionality reduction layer can improve the robustness of this reconstruction. Specifically, we propose to use principal component analysis to transform the training dataset to a lower dimensional representation and then use this representation to reconstruct the visual field. We named our new reconstruction method the transformed-target principal component regression (TTPCR). When trained on a large dataset, our new method yielded results comparable with the original linear regression method, demonstrating that there is no underfitting associated with parameter reduction. However, when trained on a small dataset, our new method used on average 22% fewer trials to reach the same error. Our results suggest that dimensionality reduction techniques can improve the robustness of visual field testing reconstruction algorithms.

A reinforcement learning approach for VQA validation: An application to diabetic macular edema grading

Recent advances in machine learning models have greatly increased the performance of automated methods in medical image analysis. However, the internal functioning of such models is largely hidden, which hinders their integration in clinical practice. Explainability and trust are viewed as important aspects of modern methods, for the latter's widespread use in clinical communities. As such, validation of machine learning models represents an important aspect and yet, most methods are only validated in a limited way. In this work, we focus on providing a richer and more appropriate validation approach for highly powerful Visual Question Answering (VQA) algorithms. To better understand the performance of these methods, which answer arbitrary questions related to images, this work focuses on an automatic visual Turing test (VTT). That is, we propose an automatic adaptive questioning method, that aims to expose the reasoning behavior of a VQA algorithm. Specifically, we introduce a reinforcement learning (RL) agent that observes the history of previously asked questions, and uses it to select the next question to pose. We demonstrate our approach in the context of evaluating algorithms that automatically answer questions related to diabetic macular edema (DME) grading. The experiments show that such an agent has similar behavior to a clinician, whereby asking questions that are relevant to key clinical concepts.

A Method for Reducing the Number of Presentations in Perimetric Test Procedures

Purpose

To introduce a new method (ARBON) for decreasing the test time of psychophysical procedures and examine its application to perimetry.

Methods

ARBON runs in parallel with an existing psychophysical procedure injecting occasional responses of seen or unseen into that procedure. Using computer simulation to mimic human responses during perimetry, we assess the performance of ARBON relative to an underlying test procedure and a version of that procedure truncated to be faster. Simulations used 610 normal eyes (age 20 to 80 years) and 163 glaucoma eyes (median mean deviation = −1.81 dB, 5th percentile = +2.14 dB, 95th percentile = −22.55 dB). Outcome measures were number of presentations and mean absolute error in threshold estimation. We also examined the probability distribution of measured thresholds.

Results

ARBON and the Truncated procedure reduced presentations by 16% and 18%, respectively. Mean error was increased by 8% to 10% for the Truncated procedure but decreased by 5% to 7% for ARBON. The probability distributions of measured thresholds using ARBON overlapped with the Underlying procedure by over 80%, whereas the Truncated procedure overlapped by 50%.

Conclusions

ARBON offers a principled method for reducing test time. ARBON can be added to any existing psychophysical procedure without requiring any change to the logic or parameters controlling the procedure, resulting in distributions of measured thresholds similar to those of the underlying procedure.

Translational Relevance

ARBON can be added to a perimetry test procedure to speed up the test while largely preserving the distribution of returned sensitivities, thus producing normative data similar to the data for the original, underlying perimetric test.

Patients' Views of Visual Field Testing and Priorities for Research Development and Translation into Practice

PURPOSE

Information regarding the views of patients, on visual field testing is limited, and no information exists regarding their preferences for test developments. This study aimed to increase knowledge of patients' experiences of visual field assessment and to explore their opinions and priorities regarding current areas of research and development.

DESIGN

Online questionnaire with purposive sampling design.

PARTICIPANTS

Adults who regularly undergo visual field tests in Australia who report having glaucoma or being at glaucomatous risk.

METHODS

An anonymous survey, implemented using the Qualtrics webtool, with both closed- and-open ended questions designed to explore opinions regarding visual field testing, visit attendance for perimetry, as well as priorities for developments.

MAIN OUTCOME MEASURES

The survey assessed 3 domains: (1) opinions regarding test duration and visit frequency, (2) subjective experience, and (3) perspectives on future developments.

RESULTS

One hundred fifty-two complete survey responses were obtained. The median age of participants was 66 years (interquartile range [IQR], 60-72 years). Most participants (70%) had experience of undergoing more than 11 visual field tests. Participants recalled that they completed visual field tests in median of 6 minutes (IQR, 5-8 minutes) and were willing to accept additional time (median, 5 minutes; IQR, 3-6 minutes) to obtain more information. Participants were prepared to increase both the number of visual field tests per eye and the frequency of visual field tests (median, 3 visits per year; IQR, 2-4 visits per year) to gain more information about their visual status. Regarding future developments, the most preferred option was "similar test times but an increase in the level of information about my visual field," which ranked significantly higher than all other options, including "shorter test times that maintain the currently available level of information about my visual field."

CONCLUSIONS

Our study confirms, in a different population and health care system, previous research reporting patient perspectives on visual field assessment. We further revealed that health care consumers show a strong preference for accurate information about their vision and report being prepared to undergo longer visual field tests or more visual field tests to achieve that outcome.