View Item 

Show simple item record

Classifier design for computerâ aided diagnosis: Effects of finite sample size on the mean performance of classical and neural network classifiers
dc.contributor.authorChan, Heang‐ping
dc.contributor.authorSahiner, Berkman
dc.contributor.authorWagner, Robert F.
dc.contributor.authorPetrick, Nicholas
dc.date.accessioned2017-01-06T20:49:41Z
dc.date.available2017-01-06T20:49:41Z
dc.date.issued1999-12
dc.identifier.citationChan, Heang‐ping ; Sahiner, Berkman; Wagner, Robert F.; Petrick, Nicholas (1999). "Classifier design for computerâ aided diagnosis: Effects of finite sample size on the mean performance of classical and neural network classifiers." Medical Physics 26(12): 2654-2668.
dc.identifier.issn0094-2405
dc.identifier.issn2473-4209
dc.identifier.urihttps://hdl.handle.net/2027.42/135032
dc.publisherWiley Periodicals, Inc.
dc.publisherAmerican Association of Physicists in Medicine
dc.subject.otherLinear regression
dc.subject.othercomputerâ aided diagnosis
dc.subject.otherclassifier design
dc.subject.otherlinear classifier
dc.subject.otherquadratic classifier
dc.subject.otherneural network
dc.subject.othersample size
dc.subject.otherfeature space dimensionality
dc.subject.otherROC analysis
dc.subject.otherImage analysis
dc.subject.otherNeural networks, fuzzy logic, artificial intelligence
dc.subject.otherLinear algebra
dc.subject.otherMultilinear algebra
dc.subject.otherNumerical linear algebra
dc.subject.otherimage classification
dc.subject.otherbackpropagation
dc.subject.othercovariance matrices
dc.subject.otherneural nets
dc.subject.othermedical image processing
dc.subject.otherperformance index
dc.subject.otherComputer aided diagnosis
dc.subject.otherArtificial neural networks
dc.subject.otherPhysicists
dc.subject.otherComputer simulation
dc.subject.otherStatistical properties
dc.titleClassifier design for computerâ aided diagnosis: Effects of finite sample size on the mean performance of classical and neural network classifiers
dc.typeArticleen_US
dc.rights.robotsIndexNoFollow
dc.subject.hlbsecondlevelMedicine (General)
dc.subject.hlbtoplevelHealth Sciences
dc.description.peerreviewedPeer Reviewed
dc.contributor.affiliationumDepartment of Radiology, University of Michigan, Ann Arbor, Michigan 48109â 0030
dc.contributor.affiliationumDepartment of Radiology, University of Michigan, Ann Arbor, Michigan 48109â 0030
dc.contributor.affiliationotherCenter for Devices and Radiology Health, FDA, Rockville, Maryland 20852
dc.description.bitstreamurlhttp://deepblue.lib.umich.edu/bitstream/2027.42/135032/1/mp8805.pdf
dc.identifier.doi10.1118/1.598805
dc.identifier.sourceMedical Physics
dc.identifier.citedreferenceC. A. Roe and C. E. Metz, â Varianceâ component modeling in the analysis of receiver operating characteristic index estimates,â Acad. Radiol. 4, 587 â 600 ( 1997 ).
dc.identifier.citedreferenceK. Fukunaga and R. R. Hayes, â Effects of sample size on classifier design,â IEEE Trans. Pattern. Anal. Mach. Intell. ITPIDJ --> 11, 873 â 885 ( 1989 ).
dc.identifier.citedreferenceR. F. Wagner, D. G. Brown, J.â P. Guedon, K. J. Myers, and K. A. Wear, in Information Processing in Medical Imaging, edited by H. H. Barrett and A. F. Gmitro (Springerâ Verlag, Berlin, 1993).
dc.identifier.citedreferenceR. F. Wagner, D. G. Brown, J.â P. Guedon, K. J. Myers, and K. A. Wear, â On combining a few diagnostic tests or features,â Proc. SPIE PSISDG --> 2167, 503 â 512 ( 1994 ).
dc.identifier.citedreferenceD. G. Brown, A. C. Schneider, M. P. Anderson, and R. F. Wagner, â Effect of finite sample size and correlated/noisy input features on neural network pattern classification,â Proc. SPIE PSISDG --> 2167, 180 â 190 ( 1994 ).
dc.identifier.citedreferenceH. P. Chan, B. Sahiner, R. F. Wagner, N. Petrick, and J. Mossoba, â Effects of sample size on classifier design: quadratic and neural network classifiers,â Proc. SPIE PSISDG --> 3034, 1102 â 1113 ( 1997 ).
dc.identifier.citedreferenceH. P. Chan, B. Sahner, R. F. Wagner, and N. Petrick, â Effects of sample size on classifier design for computerâ aided diagnosis,â Proc. SPIE PSISDG --> 3338, 845 â 858 ( 1998 ).
dc.identifier.citedreferenceR. F. Wagner, H. P. Chan, B. Sahiner, N. Petrick, and J. T. Mossoba, â Finiteâ sample effects and resampling plans: applications to linear classifiers in computerâ aided diagnosis,â Proc. SPIE PSISDG --> 3034, 467 â 477 ( 1997 ).
dc.identifier.citedreferenceR. F. Wagner, H. P. Chan, J. T. Mossoba, B. Sahiner, and N. Petrick, â Components of variance in ROC analysis of CADx Classifier performance,â Proc. SPIE PSISDG --> 3338, 859 â 875 ( 1998 ).
dc.identifier.citedreferenceR. F. Wagner, H. P. Chan, B. Sahiner, N. Petrick, and J. T. Mossoba, â Components of variance in ROC analysis of CADx classifier performance. II: Applications of the bootstrap,â Proc. SPIE PSISDG --> 3661, 523 â 532 ( 1999 ).
dc.identifier.citedreferenceD. J. Hand, Discrimination and Classification (Wiley, New York, 1981).
dc.identifier.citedreferenceP. A. Lachenbruch, Discriminant Analysis (Hafner, New York, 1975).
dc.identifier.citedreferenceR. A. Johnson and D. W. Wichern, Applied Multivariate Statistical Analysis (Prenticeâ Hall, Englewood Cliffs, NJ, 1982).
dc.identifier.citedreferenceH. P. Chan, B. Sahiner, R. F. Wagner, and N. Petrick, â Classifier design for computerâ aided diagnosis in mammography: effects of finite sample size,â Med. Phys. MPHYA6 --> 24, 1034 â 1035 ( 1997 ).
dc.identifier.citedreferenceJ. A. Freeman and D. M. Skapura, Neural Networksâ Algorithms, Applications, and Programming Techniques (Addisonâ Wesley, Reading, 1991).
dc.identifier.citedreferenceK. Fukunaga, Introduction to Statistical Pattern Recognition, 2nd ed. (Academic, New York, 1990).
dc.identifier.citedreferenceS. Raudys and V. Pikelis, â On dimensionality, sample size, classification error, and complexity of classification algorithm in pattern recognition,â IEEE Trans. Pattern. Anal. Mach. Intell. ITPIDJ --> PAMIâ 2, 242 â 252 ( 1980 ).
dc.owningcollnameInterdisciplinary and Peer-Reviewed


Files in this item

Name:
mp8805.pdf
Size:
267.7KB
Format:
PDF
View/Open

Show simple item record

Remediation of Harmful Language

The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.

Accessibility

If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.