Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects

J. S. Kippenhan, W. W. Barker, S. Pascal, J. Nagel, R. Duara

Research output: Contribution to journalArticle

64 Citations (Scopus)

Abstract

The value of PET as an objective diagnostic tool for dementia may depend on the degree to which abnormal metabolic patterns can be detected by quantitative classification methods. In these studies, a neural-network classifier based on coarse region of interest analyses was used to classify normal and abnormal FDG-PET scans. The performance of neural networks and of an expert reader were evaluated by cross-validation testing. When the 'abnormal' class was represented by subjects with clinical diagnoses of 'Probable Alzheimer's,' the areas under the relative-operating-characteristic (ROC) curves were 0.85 and 0.89 for the neural network and the expert reader, respectively. When testing with abnormal subjects represented by 'Possible AD' cases, ROC areas for both the network and the expert were 0.81. The neural network out-performed discriminant analysis. It is concluded that PET has potential for the detection of abnormal brain function in dementing diseases, and that the combination of neural networks and PET is a useful diagnostic tool. Despite the low-resolution 'view' afforded the neural network, its performance was nearly equivalent to that of an expert reader.

Original languageEnglish
Pages (from-to)1459-1467
Number of pages9
JournalJournal of Nuclear Medicine
Volume33
Issue number8
StatePublished - Jan 1 1992
Externally publishedYes

Fingerprint

Discriminant Analysis
Positron-Emission Tomography
Dementia
Alzheimer Disease
Brain

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology

Cite this

Kippenhan, J. S., Barker, W. W., Pascal, S., Nagel, J., & Duara, R. (1992). Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects. Journal of Nuclear Medicine, 33(8), 1459-1467.

Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects. / Kippenhan, J. S.; Barker, W. W.; Pascal, S.; Nagel, J.; Duara, R.

In: Journal of Nuclear Medicine, Vol. 33, No. 8, 01.01.1992, p. 1459-1467.

Research output: Contribution to journalArticle

Kippenhan, JS, Barker, WW, Pascal, S, Nagel, J & Duara, R 1992, 'Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects', Journal of Nuclear Medicine, vol. 33, no. 8, pp. 1459-1467.
Kippenhan JS, Barker WW, Pascal S, Nagel J, Duara R. Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects. Journal of Nuclear Medicine. 1992 Jan 1;33(8):1459-1467.
Kippenhan, J. S. ; Barker, W. W. ; Pascal, S. ; Nagel, J. ; Duara, R. / Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects. In: Journal of Nuclear Medicine. 1992 ; Vol. 33, No. 8. pp. 1459-1467.
@article{be15ccbcb6fd497abd11692c8619318a,
title = "Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects",
abstract = "The value of PET as an objective diagnostic tool for dementia may depend on the degree to which abnormal metabolic patterns can be detected by quantitative classification methods. In these studies, a neural-network classifier based on coarse region of interest analyses was used to classify normal and abnormal FDG-PET scans. The performance of neural networks and of an expert reader were evaluated by cross-validation testing. When the 'abnormal' class was represented by subjects with clinical diagnoses of 'Probable Alzheimer's,' the areas under the relative-operating-characteristic (ROC) curves were 0.85 and 0.89 for the neural network and the expert reader, respectively. When testing with abnormal subjects represented by 'Possible AD' cases, ROC areas for both the network and the expert were 0.81. The neural network out-performed discriminant analysis. It is concluded that PET has potential for the detection of abnormal brain function in dementing diseases, and that the combination of neural networks and PET is a useful diagnostic tool. Despite the low-resolution 'view' afforded the neural network, its performance was nearly equivalent to that of an expert reader.",
author = "Kippenhan, {J. S.} and Barker, {W. W.} and S. Pascal and J. Nagel and R. Duara",
year = "1992",
month = "1",
day = "1",
language = "English",
volume = "33",
pages = "1459--1467",
journal = "Journal of Nuclear Medicine",
issn = "0161-5505",
publisher = "Society of Nuclear Medicine Inc.",
number = "8",

}

TY - JOUR

T1 - Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects

AU - Kippenhan, J. S.

AU - Barker, W. W.

AU - Pascal, S.

AU - Nagel, J.

AU - Duara, R.

PY - 1992/1/1

Y1 - 1992/1/1

N2 - The value of PET as an objective diagnostic tool for dementia may depend on the degree to which abnormal metabolic patterns can be detected by quantitative classification methods. In these studies, a neural-network classifier based on coarse region of interest analyses was used to classify normal and abnormal FDG-PET scans. The performance of neural networks and of an expert reader were evaluated by cross-validation testing. When the 'abnormal' class was represented by subjects with clinical diagnoses of 'Probable Alzheimer's,' the areas under the relative-operating-characteristic (ROC) curves were 0.85 and 0.89 for the neural network and the expert reader, respectively. When testing with abnormal subjects represented by 'Possible AD' cases, ROC areas for both the network and the expert were 0.81. The neural network out-performed discriminant analysis. It is concluded that PET has potential for the detection of abnormal brain function in dementing diseases, and that the combination of neural networks and PET is a useful diagnostic tool. Despite the low-resolution 'view' afforded the neural network, its performance was nearly equivalent to that of an expert reader.

AB - The value of PET as an objective diagnostic tool for dementia may depend on the degree to which abnormal metabolic patterns can be detected by quantitative classification methods. In these studies, a neural-network classifier based on coarse region of interest analyses was used to classify normal and abnormal FDG-PET scans. The performance of neural networks and of an expert reader were evaluated by cross-validation testing. When the 'abnormal' class was represented by subjects with clinical diagnoses of 'Probable Alzheimer's,' the areas under the relative-operating-characteristic (ROC) curves were 0.85 and 0.89 for the neural network and the expert reader, respectively. When testing with abnormal subjects represented by 'Possible AD' cases, ROC areas for both the network and the expert were 0.81. The neural network out-performed discriminant analysis. It is concluded that PET has potential for the detection of abnormal brain function in dementing diseases, and that the combination of neural networks and PET is a useful diagnostic tool. Despite the low-resolution 'view' afforded the neural network, its performance was nearly equivalent to that of an expert reader.

UR - http://www.scopus.com/inward/record.url?scp=0026680856&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026680856&partnerID=8YFLogxK

M3 - Article

C2 - 1634935

AN - SCOPUS:0026680856

VL - 33

SP - 1459

EP - 1467

JO - Journal of Nuclear Medicine

JF - Journal of Nuclear Medicine

SN - 0161-5505

IS - 8

ER -