Neural network diagnosis of avascular necrosis from magnetic resonance images

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Artificial neural networks has been used to diagnose avascular necrosis (AVN) of the femoral head from magnetic resonance images. Multilayer perceptron networks, trained with conjugate gradient optimization, which diagnose AVN from single sagittal images of the femoral head with 100% accuracy on the training data and 97% accuracy on test data has been developed. These networks use only the raw image as input (with minimal preprocessing to average the images down to 32 × 32 size and to scale the input data values) and learn to extract their own features for the diagnosis decision. Various experiments with these networks, whose results are considered to be very encouraging for the use of neural networks in diagnostic radiology, are described.

Original languageEnglish (US)
Title of host publicationProceedings of the Annual Conference on Engineering in Medicine and Biology
Place of PublicationPiscataway, NJ, United States
PublisherPubl by IEEE
Pages1429-1431
Number of pages3
Volume13
Editionpt 3
ISBN (Print)0780302168
StatePublished - 1991
EventProceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Orlando, FL, USA
Duration: Oct 31 1991Nov 3 1991

Other

OtherProceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
CityOrlando, FL, USA
Period10/31/9111/3/91

Fingerprint

Magnetic resonance
Neural networks
Radiology
Multilayer neural networks
Experiments

ASJC Scopus subject areas

  • Bioengineering

Cite this

Manduca, A., Christy, P., & Ehman, R. L. (1991). Neural network diagnosis of avascular necrosis from magnetic resonance images. In Proceedings of the Annual Conference on Engineering in Medicine and Biology (pt 3 ed., Vol. 13, pp. 1429-1431). Piscataway, NJ, United States: Publ by IEEE.

Neural network diagnosis of avascular necrosis from magnetic resonance images. / Manduca, Armando; Christy, P.; Ehman, Richard Lorne.

Proceedings of the Annual Conference on Engineering in Medicine and Biology. Vol. 13 pt 3. ed. Piscataway, NJ, United States : Publ by IEEE, 1991. p. 1429-1431.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Manduca, A, Christy, P & Ehman, RL 1991, Neural network diagnosis of avascular necrosis from magnetic resonance images. in Proceedings of the Annual Conference on Engineering in Medicine and Biology. pt 3 edn, vol. 13, Publ by IEEE, Piscataway, NJ, United States, pp. 1429-1431, Proceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Orlando, FL, USA, 10/31/91.
Manduca A, Christy P, Ehman RL. Neural network diagnosis of avascular necrosis from magnetic resonance images. In Proceedings of the Annual Conference on Engineering in Medicine and Biology. pt 3 ed. Vol. 13. Piscataway, NJ, United States: Publ by IEEE. 1991. p. 1429-1431
Manduca, Armando ; Christy, P. ; Ehman, Richard Lorne. / Neural network diagnosis of avascular necrosis from magnetic resonance images. Proceedings of the Annual Conference on Engineering in Medicine and Biology. Vol. 13 pt 3. ed. Piscataway, NJ, United States : Publ by IEEE, 1991. pp. 1429-1431
@inproceedings{f4c5d8fc73ad4e51bc9c836c711a5cb2,
title = "Neural network diagnosis of avascular necrosis from magnetic resonance images",
abstract = "Artificial neural networks has been used to diagnose avascular necrosis (AVN) of the femoral head from magnetic resonance images. Multilayer perceptron networks, trained with conjugate gradient optimization, which diagnose AVN from single sagittal images of the femoral head with 100{\%} accuracy on the training data and 97{\%} accuracy on test data has been developed. These networks use only the raw image as input (with minimal preprocessing to average the images down to 32 × 32 size and to scale the input data values) and learn to extract their own features for the diagnosis decision. Various experiments with these networks, whose results are considered to be very encouraging for the use of neural networks in diagnostic radiology, are described.",
author = "Armando Manduca and P. Christy and Ehman, {Richard Lorne}",
year = "1991",
language = "English (US)",
isbn = "0780302168",
volume = "13",
pages = "1429--1431",
booktitle = "Proceedings of the Annual Conference on Engineering in Medicine and Biology",
publisher = "Publ by IEEE",
edition = "pt 3",

}

TY - GEN

T1 - Neural network diagnosis of avascular necrosis from magnetic resonance images

AU - Manduca, Armando

AU - Christy, P.

AU - Ehman, Richard Lorne

PY - 1991

Y1 - 1991

N2 - Artificial neural networks has been used to diagnose avascular necrosis (AVN) of the femoral head from magnetic resonance images. Multilayer perceptron networks, trained with conjugate gradient optimization, which diagnose AVN from single sagittal images of the femoral head with 100% accuracy on the training data and 97% accuracy on test data has been developed. These networks use only the raw image as input (with minimal preprocessing to average the images down to 32 × 32 size and to scale the input data values) and learn to extract their own features for the diagnosis decision. Various experiments with these networks, whose results are considered to be very encouraging for the use of neural networks in diagnostic radiology, are described.

AB - Artificial neural networks has been used to diagnose avascular necrosis (AVN) of the femoral head from magnetic resonance images. Multilayer perceptron networks, trained with conjugate gradient optimization, which diagnose AVN from single sagittal images of the femoral head with 100% accuracy on the training data and 97% accuracy on test data has been developed. These networks use only the raw image as input (with minimal preprocessing to average the images down to 32 × 32 size and to scale the input data values) and learn to extract their own features for the diagnosis decision. Various experiments with these networks, whose results are considered to be very encouraging for the use of neural networks in diagnostic radiology, are described.

UR - http://www.scopus.com/inward/record.url?scp=0026298264&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026298264&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0026298264

SN - 0780302168

VL - 13

SP - 1429

EP - 1431

BT - Proceedings of the Annual Conference on Engineering in Medicine and Biology

PB - Publ by IEEE

CY - Piscataway, NJ, United States

ER -