2008
DOI: 10.1118/1.3002413
|View full text |Cite
|
Sign up to set email alerts
|

Model for the detection of signals in images with multiple suspicious locations

Abstract: A signal detection model is presented that combines a signal model and a noise model providing mathematical descriptions of the frequency of appearance of the signals, and of the signal-like features naturally occurring in the background. We derive expressions for the likelihood functions for the whole ensemble of observed suspicious locations, in various possible combinations of signals and false signal candidates. As a result, this formalism is able to describe several new types of detection tests using like… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2010
2010
2018
2018

Publication Types

Select...
6
1
1

Relationship

0
8

Authors

Journals

citations
Cited by 9 publications
(3 citation statements)
references
References 27 publications
(33 reference statements)
0
3
0
Order By: Relevance
“…Due to the limitations of SKE tasks, recent studies examine the possibility of allowing for unknown signal location. 22,[25][26][27][28][29][30] For forcedchoice experiments, the figure of merit remains the percentage correct. More generally, the probability of correct joint detection and localization can be plotted vs the probability of false positives; the figure of merit is the area under this curve.…”
Section: B2 the Role Of Signal Searchmentioning
confidence: 99%
“…Due to the limitations of SKE tasks, recent studies examine the possibility of allowing for unknown signal location. 22,[25][26][27][28][29][30] For forcedchoice experiments, the figure of merit remains the percentage correct. More generally, the probability of correct joint detection and localization can be plotted vs the probability of false positives; the figure of merit is the area under this curve.…”
Section: B2 the Role Of Signal Searchmentioning
confidence: 99%
“…In order to test the EFROC estimator and to compare it with its nonparametric counterparts for ROC, LROC, and AFROC analyses, we carried out a simulation study using a previously presented model, 34 with parameterization derived from work involving searching for signals on PET images. 6 The model considers small signals of size comparable to the spatial resolution of the imaging device so that their contrast relative to the surrounding background is practically the only measurable variable indicative to their degree of suspiciousness.…”
Section: A the Simulation Modelmentioning
confidence: 99%
“…Proper quantification of the diagnostic accuracy in the FROC data is not trivial and requires specialized tools (Chapter 9.3 in [1], Chapter 8 in [11], [12]). The fundamental tool in the FROC analysis is the FROC curve 5,[13][14][15] which reflects the tradeoff between the false-positives and true-positive findings with changing decision threshold. Some FROC approaches rely directly on the FROC curve and associated indices, [16][17][18] while others focus on specific summary indices which do not necessarily agree with the empirical FROC curve [11(Ch.…”
Section: Introductionmentioning
confidence: 99%