Shannon information and receiver operating characteristic analysis for multiclass classification in imaging

Eric W Clarkson, Johnathan B. Cushing

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We show how Shannon information is mathematically related to receiver operating characteristic (ROC) analysis for multiclass classification problems in imaging. In particular, the minimum probability of error for the ideal observer, as a function of the prior probabilities for each class, determines the Shannon Information for the classification task, also considered as a function of the prior probabilities on the classes. In the process, we show how an ROC hypersurface that has been studied by other researchers is mathematically related to a Shannon information ROC (SIROC) hypersurface. In fact, the ROC hypersurface completely determines the SIROC hypersurface via a non-local integral transform on the ROC hypersurface. We also show that both hypersurfaces are convex and satisfy other geometrical relationships via the Legendre transform.

Original languageEnglish (US)
Pages (from-to)930-937
Number of pages8
JournalJournal of the Optical Society of America A: Optics and Image Science, and Vision
Volume33
Issue number5
DOIs
StatePublished - May 1 2016

Fingerprint

ROC Curve
receivers
Imaging techniques
integral transformations
Research Personnel

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Computer Vision and Pattern Recognition

Cite this

@article{2e559bcb53b5411584207be09d3318a2,
title = "Shannon information and receiver operating characteristic analysis for multiclass classification in imaging",
abstract = "We show how Shannon information is mathematically related to receiver operating characteristic (ROC) analysis for multiclass classification problems in imaging. In particular, the minimum probability of error for the ideal observer, as a function of the prior probabilities for each class, determines the Shannon Information for the classification task, also considered as a function of the prior probabilities on the classes. In the process, we show how an ROC hypersurface that has been studied by other researchers is mathematically related to a Shannon information ROC (SIROC) hypersurface. In fact, the ROC hypersurface completely determines the SIROC hypersurface via a non-local integral transform on the ROC hypersurface. We also show that both hypersurfaces are convex and satisfy other geometrical relationships via the Legendre transform.",
author = "Clarkson, {Eric W} and Cushing, {Johnathan B.}",
year = "2016",
month = "5",
day = "1",
doi = "10.1364/JOSAA.33.000930",
language = "English (US)",
volume = "33",
pages = "930--937",
journal = "Journal of the Optical Society of America A: Optics and Image Science, and Vision",
issn = "1084-7529",
publisher = "The Optical Society",
number = "5",

}

TY - JOUR

T1 - Shannon information and receiver operating characteristic analysis for multiclass classification in imaging

AU - Clarkson, Eric W

AU - Cushing, Johnathan B.

PY - 2016/5/1

Y1 - 2016/5/1

N2 - We show how Shannon information is mathematically related to receiver operating characteristic (ROC) analysis for multiclass classification problems in imaging. In particular, the minimum probability of error for the ideal observer, as a function of the prior probabilities for each class, determines the Shannon Information for the classification task, also considered as a function of the prior probabilities on the classes. In the process, we show how an ROC hypersurface that has been studied by other researchers is mathematically related to a Shannon information ROC (SIROC) hypersurface. In fact, the ROC hypersurface completely determines the SIROC hypersurface via a non-local integral transform on the ROC hypersurface. We also show that both hypersurfaces are convex and satisfy other geometrical relationships via the Legendre transform.

AB - We show how Shannon information is mathematically related to receiver operating characteristic (ROC) analysis for multiclass classification problems in imaging. In particular, the minimum probability of error for the ideal observer, as a function of the prior probabilities for each class, determines the Shannon Information for the classification task, also considered as a function of the prior probabilities on the classes. In the process, we show how an ROC hypersurface that has been studied by other researchers is mathematically related to a Shannon information ROC (SIROC) hypersurface. In fact, the ROC hypersurface completely determines the SIROC hypersurface via a non-local integral transform on the ROC hypersurface. We also show that both hypersurfaces are convex and satisfy other geometrical relationships via the Legendre transform.

UR - http://www.scopus.com/inward/record.url?scp=84969922129&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84969922129&partnerID=8YFLogxK

U2 - 10.1364/JOSAA.33.000930

DO - 10.1364/JOSAA.33.000930

M3 - Article

AN - SCOPUS:84969922129

VL - 33

SP - 930

EP - 937

JO - Journal of the Optical Society of America A: Optics and Image Science, and Vision

JF - Journal of the Optical Society of America A: Optics and Image Science, and Vision

SN - 1084-7529

IS - 5

ER -