Search PubMed⌕ Search

Biomedical subjects

M P Eckstein

Publications and source records attributed to M P Eckstein.

12 recordsLinked to original sources
PubMed2001

Quantifying the performance limits of human saccadic targeting during visual search.

In previous studies of saccadic targeting, the issue how visually guided saccades to unambiguous targets are programmed and executed has been examined. These studies have found different degrees of guidance for saccades depending on the task and task difficulty. In this study, we use ideal-observer analysis to estimate the visual information used for the first saccade during a search for a target disk in noise. We quantitatively compare the performance of the first saccadic decision to that of the ideal observer (ie absolute efficiency of the first saccade) and to that of the associated final perceptual decision at the end of the search (ie relative efficiency of the first saccade). Our results show, first, that at all levels of salience tested, the first saccade is based on visual information from the stimulus display, and its highest absolute efficiency is approximately 20%. Second, the efficiency of the first saccade is lower than that of the final perceptual decision after active search (with eye movements) and has a minimum relative efficiency of 19% at the lowest level of saliency investigated. Third, we found that requiring observers to maintain central fixation (no saccades allowed) decreased the absolute efficiency of their perceptual decision by up to a factor of two, but that the magnitude of this effect depended on target salience. Our results demonstrate that ideal-observer analysis can be extended to measure the visual information mediating saccadic target-selection decisions during visual search, which enables direct comparison of saccadic and perceptual efficiencies.

Decision Theory↗
PubMed2001

Effects of luminance oscillations on simulated lightness discriminations.

The speed of processes underlying lightness constancy was studied by having observers discriminate small differences in simulated lightness under an oscillating illumination. The period of oscillation varied from 0.25 to 120 sec. The target was a 1 degrees square which appeared for 150 msec at random intervals either directly against a uniform background or separated from the background by a 1 degrees dark gap. When the target and background were adjacent to each other, discrimination accuracy approached control levels (fixed illumination) at all but the shortest periods of oscillation. When the gap was introduced, accuracy increased as the period of oscillation increased, but never approached control levels. The results suggest that a fast local contrast mechanism is the primary mediator of lightness constancy for this task, but that there is also a slower mechanism that may be related to adaptation.

Adult↗
PubMed2000

Visual signal detection in structured backgrounds. III. Calculation of figures of merit for model observers in statistically nonstationary backgrounds.

Models of human visual detection have been successfully used in computer-generated noise. For these backgrounds, which are generally statistically stationary, model performance can be readily calculated by computing the index of detectability d' from the noise power spectrum, the signal profile, and the model template. However, model observers are ultimately needed in more real backgrounds, which may be statistically non-stationary. We investigated different methods to calculate figures of merit for model observers in real backgrounds based on different assumptions about image stationarity. We computed performance of the nonpre-whitening matched-filter observer with an eye filter on mammography and coronary angiography for an additive or a multiplicative signal. Performance was measured either by applying the model template to the images or by computing closed-form expressions with various assumptions about image stationarity. Results show first that the structured backgrounds investigated cannot be considered stationary. Second, traditional closed-form expressions of detectability calculated from the noise power spectra with the assumption of background stationarity lead to erroneous estimates of model performance. Third, the most accurate way of measuring model performances is by directly applying the model template on the images or by computing a closed-form expression that does not assume image stationarity.

Choice Behavior↗
PubMed2000

Visual signal detection in structured backgrounds. IV. Figures of merit for model performance in multiple-alternative forced-choice detection tasks with correlated responses.

Many investigators are currently developing models to predict human performance in detecting a signal embedded in complex backgrounds. A common figure of merit for model performance is d', an index of detectability that can be mathematically related to the proportion correct (Pc) when the responses of the model are Gaussian distributed and statistically independent. However, in many multiple-alternative forced-choice (MAFC) detection tasks, the target appears in one of M different locations within an image. If the image contains slow spatially varying luminance changes (low-pass noise), the pixel luminance values at the possible signal locations are correlated and therefore the model/human responses to the different locations might also be correlated. We investigate the effect of response correlations on model performance and compare different figures of merit for these conditions. Our results show that use of the standard d' index of detectability assuming statistical independence can lead to erroneous underestimates of Pc and misleading comparisons of models. We introduce a novel figure of merit d'(r) that takes into account response correlations and can be used to accurately estimate Pc. Furthermore, we show that d'(r) can be readily related to the standard index of detectability d' by d'(r) = d'/square root of (1 - r), where r is the correlation between the responses in any MAFC detection task. We illustrate the use of the theory by computing figures of merit for two linear models detecting a signal in one of four locations within medical image backgrounds.

Choice Behavior↗
PubMed2000

Derivation of a detectability index for correlated responses in multiple-alternative forced-choice experiments.

In a recent paper [J. Opt. Soc. Am. A 17, 206 (2000)], a modified detectability index, d'r, was proposed to accommodate correlations between internal responses in multiple-alternative forced-choice (MAFC) experiments. The derivation given in that work pertained only to two-alternative forced choice, although it was shown empirically that the result held for general MAFC tasks when the correlation between responses is constant. Here we present a rigorous derivation that shows that the d'r result generalizes to MAFC tasks in this case.

Choice Behavior↗
PubMed2000

A signal detection model predicts the effects of set size on visual search accuracy for feature, conjunction, triple conjunction, and disjunction displays.

Recently, quantitative models based on signal detection theory have been successfully applied to the prediction of human accuracy in visual search for a target that differs from distractors along a single attribute (feature search). The present paper extends these models for visual search accuracy to multidimensional search displays in which the target differs from the distractors along more than one feature dimension (conjunction, disjunction, and triple conjunction displays). The model assumes that each element in the display elicits a noisy representation for each of the relevant feature dimensions. The observer combines the representations across feature dimensions to obtain a single decision variable, and the stimulus with the maximum value determines the response. The model accurately predicts human experimental data on visual search accuracy in conjunctions and disjunctions of contrast and orientation. The model accounts for performance degradation without resorting to a limited-capacity spatially localized and temporally serial mechanism by which to bind information across feature dimensions.

Adult↗
PubMed1998

Why do anatomic backgrounds reduce lesion detectability?

RATIONALE AND OBJECTIVES: Developing metrics of medical image quality requires an understanding of how anatomic backgrounds reduce human visual detection performance. Visual psychophysics has shown that there are two distinct ways in which a complex background can degrade performance: (1) the presence of a deterministic high-contrast background, (2) variability in the background from location to location. The authors investigated how these two sources of performance degradation reduce human visual performance locating a lesion in anatomic backgrounds. METHODS: Human performance localizing a disk-shaped lesion in one of four locations (four alternative forced choice) was measured for three background conditions. In the first condition the background was a uniform gray. In the second condition (the repeated background condition) an anatomic background was sampled on each trial and used as a background for the four possible lesion locations. In the third condition (the different background condition) four different anatomic backgrounds were sampled on each trial and used for the four possible lesion locations. Test images consisted of computer simulated lesions mathematically projected on digital x-ray coronary angiograms. RESULTS: For five levels of lesion contrast, visual detection performance for two observers decreased significantly from the uniform background condition to the repeated background condition, and decreased even further for the different background condition. CONCLUSIONS: Study results show that both the presence of a deterministic high-contrast background and the background variations contribute to performance degradation of human visual detection of signals in anatomic backgrounds.

Diagnostic Imaging↗
PubMed1997

Visual signal detection in structured backgrounds. II. Effects of contrast gain control, background variations, and white noise.

Studies of visual detection of a signal superimposed on one of two identical backgrounds show performance degradation when the background has high contrast and is similar in spatial frequency and/or orientation to the signal. To account for this finding, models include a contrast gain control mechanism that pools activity across spatial frequency, orientation and space to inhibit (divisively) the response of the receptor sensitive to the signal. In tasks in which the observer has to detect a known signal added to one of M different backgrounds grounds due to added visual noise, the main sources of degradation are the stochastic noise in the image and the suboptimal visual processing. We investigate how these two sources of degradation (contrast gain control and variations in the background) interact in a task in which the signal is embedded in one of M locations in a complex spatially varying background (structured background). We use backgrounds extracted from patient digital medical images. To isolate effects of the fixed deterministic background (the contrast gain control) from the effects of the background variations, we conduct detection experiments with three different background conditions: (1) uniform background, (2) a repeated sample of structured background, and (3) different samples of structured background. Results show that human visual detection degrades from the uniform background condition to the repeated background condition and degrades even further in the different backgrounds condition. These results suggest that both the contrast gain control mechanism and the background random variations degrade human performance in detection of a signal in a complex, spatially varying background. A filter model and added white noise are used to generate estimates of sampling efficiencies, an equivalent internal noise, an equivalent contrast-gain-control-induced noise, and an equivalent noise due to the variations in the structured background.

Contrast Sensitivity↗
PubMed1996

Visual signal detection in structured backgrounds. I. Effect of number of possible spatial locations and signal contrast.

Several studies have investigated the effect of signal location uncertainty on the detectability of simple visual signals in uncorrelated Gaussian noise with a deterministic background. For this case, human performance in locating a signal in a forced-choice experiment has been successfully predicted for 2-1800 alternative locations with the use of signal detection theory and the usual assumption that the observer's internal response is Gaussian distributed. Gaussian uncorrelated noise is far from realistic medical image noise, which includes not only fluctuations in intensity of quantum origin but also other anatomical objects lying in the x-ray path (structured backgrounds). Our goal is to determine whether signal detection theory with the Gaussian assumption is adequate for the case of structured backgrounds, or whether other more complex models need to be developed to predict human performance as a function of the number of possible signal locations in structured backgrounds. We present experimental data suggesting that an assumed Gaussian internal response accurately predicts the decrease in observer performance as the number of alternative locations is increased. The one exception is a lower-than-predicted performance for the detection of low-contrast signals for two alternative locations. Performance as measured by the index of detectability d' is also found to be linear with signal contrast. Together these findings extend the applicability of signal detection theory with Gaussian internal response functions to the case of complex structured backgrounds.

Computer Simulation↗
PubMed1996

Role of knowledge in human visual temporal integration in spatiotemporal noise.

Previous studies have shown how human observers' knowledge about the signal's spatial frequency, spatial phase, and spatial locations affects human performance in detecting and identifying signals in spatial noise. These results have led to the idea that human observers can be modeled as suboptimal Bayesian observers that use a priori information to generate probabilities or likelihoods for hypothesis. This approach has also been applied more recently to object recognition. We investigate whether human observers have the ability to use information about the temporal profile of a temporally modulated signal in temporal information processing. We measure human performance in detecting a time-varying signal embedded in spatiotemporal (dynamic) noise with and without a cue that contains information about the temporal phase of the signal. Results show improvement in performance in the phase-cued condition, suggesting that human observers act as if they have the ability to use knowledge about the temporal shape of the signal when performing temporal information processing. Human performance is consistent with a suboptimal Bayesian observer and a newly proposed Max-Min observer. The results also suggest that models based solely on the integration of the early temporal filters in the human visual system and/or any further integration (e.g., probability summation), which do not make use of knowledge about the signals' temporal profile, are incomplete models of human visual detection in spatiotemporal noise.

Adult↗
PubMed1994

Improving detection of coronary morphological features from digital angiograms. Effect of stenosis-stabilized display.

BACKGROUND: We have developed a digital display method that stabilizes the motion of a stenosis in sequential frames of a coronary angiogram, allowing it to be scrutinized at high display frame rates. The purpose of this study was to determine whether this technique improves visual detection of low-contrast luminal morphological features. METHODS AND RESULTS: An observer detection study was conducted using computer-simulated arterial segments containing known target features, inserted into clinical digital coronary angiograms. Four observers performed a forced-choice detection of a simulated filling defect in each of 320 angiograms using the conventional and stenosis-stabilized dynamic displays (at 7.5, 15, and 32 frames per second) and a single-frame static display (total of 8960 detections). In a second simulated clinical task, three observers detected a bridging stenotic lumen in 600 angiograms using the two displays (3600 detections). In a third experiment, two angiographers rated the likelihood of intraluminal thrombus in 89 right coronary digital angiograms by consensus reading with both dynamic displays. Detectability of the simulated filling defect was similar for both dynamic display methods at 7.5 frames per second (averaging twice that for static images). As display rate was increased to 32 frames per second, detectability for the conventional display declined, whereas the stabilized display detectability increased for all observers (P < .05). On average, stabilization allowed detection of filling defects equivalent to a 71% increase in feature contrast. Response time for the conventional display averaged 12.9 +/- 4.7 seconds. For the stenosis-stabilized display, response time fell with increased frame rate (P < .05) to 4.9 +/- 1.2 seconds at 32 Hz, similar to the time for static images (4.6 +/- 0.8 seconds). The detectability of the bridging stenotic lumen was increased by 62% with the stabilization compared with conventional dynamic display (P < .00001). Consensus reading of coronary angiograms showed differences between the two dynamic display methods (kappa = 0.11) that may be explained by an improvement in observer uncertainty. A rating of definite for thrombus present or absent was more frequent with the stabilized display (39% versus 15%, P < .0001). CONCLUSIONS: These data suggest that stabilized display of coronary angiograms significantly increases detectability, reduces the time required for detection, and improves observer uncertainty for the presence of small luminal morphological features. The method of angiographic display may thus have an impact on clinical coronary angiographic interpretation.

Coronary Angiography↗