A multitarget training method for artificial neural network with application to computer‐aided diagnosis. Issue 1 (26th December 2012)

• Record Type: Journal Article
• Authors: Liu, Bei; Jiang, Yulei
• Is Part Of: Medical physics. Volume 40:Issue 1(2013)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2012-12-26
• Subjects: Medical imaging -- Neural networks -- Monte Carlo simulations; cancer -- computer aided analysis -- image classification -- mammography -- medical image processing -- Monte Carlo methods -- neural nets -- tumours; artificial neural network (ANN) -- ANN training -- breast cancer -- histological subtype -- computer‐aided diagnosis (CAD); Digital computing or data processing equipment or methods, specially adapted for specific applications -- Image data processing or generation, in general; Mammography -- Cancer -- Computer aided diagnosis -- Medical imaging -- Artificial neural networks -- Monte Carlo methods -- Medical diagnosis -- Number theory -- Probability theory -- Testing procedures; Medical physics -- Periodicals; Medical physics; Geneeskunde; Natuurkunde; Toepassingen; Biophysics; Periodicals; Periodicals; Electronic journals; 610.153
• DOI: 10.1118/1.4772021
• Languages: English
• ISSNs: 0094-2405

Abstract

Abstract : Purpose: : The authors propose a new training method for artificial neural networks (ANNs) in two‐class classification tasks such as classifying breast lesions on a mammogram as malignant or benign. Methods: : Whereas the conventional binary training method uses binary training target values based on the diagnostic truth of a lesion being malignant or benign, the authors use multiple training target values based on more detailed histological diagnosis that presumably are related to the posterior probability of a lesion being malignant. The authors performed Monte Carlo simulation studies in which training target values were assigned based on posterior probability, and they also performed a mammography study in which training target values were assigned according to histological subtypes. Results: : These studies showed that the multitarget training method produced less variability in the ANN outputs than the binary training method. The simulation studies also showed that except for when the number of training cases was extremely large, the multitarget training method produced improved overall classification performance over the binary training method. Conclusions: : Therefore, the multitarget ANN training method is potentially useful for ANN applications in computer‐aided diagnosis of breast cancer.