Visual test

About: Visual test is a research topic. Over the lifetime, 915 publications have been published within this topic receiving 12843 citations.

Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation

Abstract: This article presents individual conditional expectation (ICE) plots, a tool for visualizing the model estimated by any supervised learning algorithm. Classical partial dependence plots (PDPs) help visualize the average partial relationship between the predicted response and one or more features. In the presence of substantial interaction effects, the partial response relationship can be heterogeneous. Thus, an average curve, such as the PDP, can obfuscate the complexity of the modeled relationship. Accordingly, ICE plots refine the PDP by graphing the functional relationship between the predicted response and the feature for individual observations. Specifically, ICE plots highlight the variation in the fitted values across the range of a covariate, suggesting where and to what extent heterogeneities might exist. In addition to providing a plotting suite for exploratory analysis, we include a visual test for additive structure in the data-generating model. Through simulated examples and real datasets, we...

The Disappearance of Steadily Fixated Visual Test Objects

Abstract: A system has been devised for causing an image to remain at one point on the retina regardless of eye movements. A beam of light, reflected from a plane mirror on a contact lens, is used to project onto a screen an image of a dark line against a bright background. The screen is viewed by the same eye through an optical system which compensates for the doubling of the angle of rotation of the beam projected from the mirror on the contact lens. Thus, any motion of the eye causes a deviation of the beam such that the retinal image of the projected line undergoes the same displacement as do the retinal receptor cells. By comparison with normal viewing of the same test objects it is found that (1) when first presented, the finest lines are seen with normal or slightly better than normal acuity, (2) within a few seconds the lines begin to disappear, and (3) within one minute even coarse lines are seen only intermittently. The results may be interpreted in terms of local retinal adaptation to a stationary field.

Evaluation of visual and instrument shade matching

Abstract: Statement of problem. Visual color matching to determine shades in dentistry is inconsistent and unreliable. If accurate, instrumental measurement of tooth color would provide objective, quantified data to match natural teeth to clinical shade guides. Purpose. This study evaluated and compared the ability of a new computerized colorimeter and a simple visual test to match ceramic shade guide teeth. Material and methods. Thirty-one (n = 31) observers with normal color vision were allowed unlimited time to match one set of Vita Lumin shade guide teeth to the corresponding shade guide teeth of a second Vita Lumin shade guide. The same test was administered to 14 of the observers several months later to determine within-subject variability. A computerized colorimeter (Colortron II) equipped with a positioning guide was used to measure the middle third of each shade guide tooth. Through a "match tool" present in the computer's software, readings from one shade guide were matched with readings of the other shade guide by using CIELAB measurements and ΔE values. The mean number of correct matches by the colorimeter and of correct matches in visual test were compared with a 1-tailed t test. Repeatability for both tests was determined with a paired t test. Results. The Colortron II instrument correctly matched 8 of the 16 tabs (50% correct), whereas visual matching by examiners averaged 7.7 of 16 correct matches (48% correct) (standard deviation 2.7). No statistically significant differences existed between the 2 methods. The colorimeter demonstrated 100% repeatability and the visual test demonstrated fair repeatability (correlation coefficient r=.60). Conclusions. Shade determination by visual means was inconsistent. Accuracy of a new colorimeter in matching porcelain shade guide teeth was only slightly better. (J Prosthet Dent 1998;80:642-8.)

Peeking Inside the Black Box: Visualizing Statistical Learning with Plots of Individual Conditional Expectation

Abstract: This article presents Individual Conditional Expectation (ICE) plots, a tool for visualizing the model estimated by any supervised learning algorithm. Classical partial dependence plots (PDPs) help visualize the average partial relationship between the predicted response and one or more features. In the presence of substantial interaction effects, the partial response relationship can be heterogeneous. Thus, an average curve, such as the PDP, can obfuscate the complexity of the modeled relationship. Accordingly, ICE plots refine the partial dependence plot by graphing the functional relationship between the predicted response and the feature for individual observations. Specifically, ICE plots highlight the variation in the fitted values across the range of a covariate, suggesting where and to what extent heterogeneities might exist. In addition to providing a plotting suite for exploratory analysis, we include a visual test for additive structure in the data generating model. Through simulated examples and real data sets, we demonstrate how ICE plots can shed light on estimated models in ways PDPs cannot. Procedures outlined are available in the R package ICEbox.