Skip to main content

Validation of a constraint satisfaction neural network for breast cancer diagnosis: New results from 1,030 cases

Publication ,  Journal Article
Tourassi, GD; Lo, JY; Markey, MK
Published in: Proceedings of SPIE - The International Society for Optical Engineering
September 15, 2003

Previously, we presented a Constraint Satisfaction Neural Network (CSNN) to predict the outcome of breast biopsy using mammographic and clinical findings. Based on 500 cases, the study showed that CSNN was able to operate not only as a predictive but also as a knowledge discovery tool. The purpose of this study is to validate the CSNN on a database of additional 1,030 cases. An auto-associative backpropagation scheme was used to determine the CSNN constraints based on the initial 500 patients. Subsequently, the CSNN was applied to 1,030 new patients (358 patients with malignant and 672 with benign lesions) to predict breast lesion malignancy. For every test case, the CSNN reconstructed the diagnosis node given the network constraints and the external inputs to the network. The activation level achieved by the diagnosis node was used as the decision variable for ROC analysis. Overall, the CSNN continued to perform well over this large dataset with ROC area of Az = 0.81 ± 0.02. However, the diagnostic performance of the network was inferior in cases with missing clinical findings (Az = 0.80 ± 0.02) compared to those with complete findings (Az = 0.84 ± 0.03). The study also demonstrated the ability of the CSNN to effectively impute missing findings while performing as a predictive tool.

Duke Scholars

Published In

Proceedings of SPIE - The International Society for Optical Engineering

DOI

ISSN

0277-786X

Publication Date

September 15, 2003

Volume

5032 I

Start / End Page

207 / 214

Related Subject Headings

  • 5102 Atomic, molecular and optical physics
  • 4009 Electronics, sensors and digital hardware
  • 4006 Communications engineering
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Tourassi, G. D., Lo, J. Y., & Markey, M. K. (2003). Validation of a constraint satisfaction neural network for breast cancer diagnosis: New results from 1,030 cases. Proceedings of SPIE - The International Society for Optical Engineering, 5032 I, 207–214. https://doi.org/10.1117/12.481111
Tourassi, G. D., J. Y. Lo, and M. K. Markey. “Validation of a constraint satisfaction neural network for breast cancer diagnosis: New results from 1,030 cases.” Proceedings of SPIE - The International Society for Optical Engineering 5032 I (September 15, 2003): 207–14. https://doi.org/10.1117/12.481111.
Tourassi GD, Lo JY, Markey MK. Validation of a constraint satisfaction neural network for breast cancer diagnosis: New results from 1,030 cases. Proceedings of SPIE - The International Society for Optical Engineering. 2003 Sep 15;5032 I:207–14.
Tourassi, G. D., et al. “Validation of a constraint satisfaction neural network for breast cancer diagnosis: New results from 1,030 cases.” Proceedings of SPIE - The International Society for Optical Engineering, vol. 5032 I, Sept. 2003, pp. 207–14. Scopus, doi:10.1117/12.481111.
Tourassi GD, Lo JY, Markey MK. Validation of a constraint satisfaction neural network for breast cancer diagnosis: New results from 1,030 cases. Proceedings of SPIE - The International Society for Optical Engineering. 2003 Sep 15;5032 I:207–214.

Published In

Proceedings of SPIE - The International Society for Optical Engineering

DOI

ISSN

0277-786X

Publication Date

September 15, 2003

Volume

5032 I

Start / End Page

207 / 214

Related Subject Headings

  • 5102 Atomic, molecular and optical physics
  • 4009 Electronics, sensors and digital hardware
  • 4006 Communications engineering