Mammographic masses classification: comparison between backpropagation neural network (BNN), K nearest neighbors (KNN), and human readers

Lina Arbach; Joseph M Reinhardt; D Lee Bennett; Ghassan Fallouh

doi:10.1109/CCECE.2003.1226174

Back

Conference proceeding

Mammographic masses classification: comparison between backpropagation neural network (BNN), K nearest neighbors (KNN), and human readers

Lina Arbach, Joseph M Reinhardt, D Lee Bennett and Ghassan Fallouh

CCECE 2003 - Canadian Conference on Electrical and Computer Engineering. Toward a Caring and Humane Technology (Cat. No.03CH37436), Vol.3, pp.1441-1444 vol.3

2003

DOI: 10.1109/CCECE.2003.1226174

View Online

Abstract

We compare mammographic mass classification performance between a backpropagation neural network (BNN), K nearest neighbors, expert radiologists, and residents. Our goal is to reduce false negatives during screening of mammograms. 160 cases were used. Each case contained at least one mass and had an accompanying biopsy result. Masses were extracted using region growing with seed locations identified by an expert radiologist. 10 texture and shape based features were used as inputs to a BNN and KNN. 140 cases were used for training the BNN and the KNN. The remaining 20 cases were used for testing. The testing set was diagnosed by three expert radiologists, three residents, the BNN, and the KNN. We evaluated the human readers and the BNN by computing the area under the ROC curve (AUC). The KNN was evaluated by computing the sensitivity, specificity, and number of false negatives (FN). The AUC was 0.923 for the BNN, 0.846 for the expert radiologists, and 0.648 for the residents. The KNN had a specificity 85.7% with sensitivity 84.6% and with FN=2. These results illustrate the promise of using the BNN as a physician's assistant for breast mass classification.

Breast Cancer

Backpropagation

Biopsy

Humans

Mammography

Nearest neighbor searches

Neural networks

Sensitivity

Shape

Testing

Details

Title: Subtitle: Mammographic masses classification: comparison between backpropagation neural network (BNN), K nearest neighbors (KNN), and human readers
Creators: Lina Arbach - Iowa Univ., Iowa City, IA, USA
Joseph M Reinhardt - Iowa Univ., Iowa City, IA, USA
D Lee Bennett - Iowa Univ., Iowa City, IA, USA
Ghassan Fallouh
Resource Type: Conference proceeding
Publication Details: CCECE 2003 - Canadian Conference on Electrical and Computer Engineering. Toward a Caring and Humane Technology (Cat. No.03CH37436), Vol.3, pp.1441-1444 vol.3
DOI: 10.1109/CCECE.2003.1226174
ISSN: 0840-7789
eISSN: 2576-7046
Publisher: IEEE
Language: English
Date published: 2003
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Radiology
Record Identifier: 9984204105202771

Metrics

25 Record Views