Document Detail

Estimating three-class ideal observer decision variables for computerized detection and classification of mammographic mass lesions.
MedLine Citation:
PMID:  14761024     Owner:  NLM     Status:  MEDLINE    
We are using Bayesian artificial neural networks (BANNs) to classify mammographic masses in schemes for computer-aided diagnosis, and we are extending this methodology to a three-class classification task. We investigated whether a BANN can estimate ideal observer decision variables to distinguish malignant, benign, and false-positive computer detections. Five features were calculated for 63 malignant and 29 benign computer-detected mass lesions, and for 1049 false-positive computer detections, in 440 mammograms randomly divided into a training and testing set. A BANN was trained on the training set features and applied to the testing set features. We then used a known relation between three-class ideal observer decision variables and that used by a two-class ideal observer when two of three classes are grouped into one class, giving one decision variable for distinguishing malignant from nonmalignant detections, and a second for distinguishing true-positive from false-positive computer detections. For comparison, we grouped the training data into two classes in the same two ways and trained two-class BANNs for these two tasks. The three-class BANN decision variables were essentially identical in performance to the specifically trained two-class BANNs, with the average difference in area under the ROC curves being less than 0.0035 and no differences in area being statistically significant. Thus, the BANN outputs obey the same theoretical relationship as do the three-class and two-class ideal observer decision variables, which is consistent with the claim that the three-class BANN output can provide good estimates of the decision variables used by a three-class ideal observer.
Darrin C Edwards; Li Lan; Charles E Metz; Maryellen L Giger; Robert M Nishikawa
Related Documents :
11570714 - Model testing for reliability and validity of the outcome expectations for exercise scale.
23084704 - Exercise improves behavioral, neurocognitive, and scholastic performance in children wi...
19021674 - A validated lip fullness grading scale.
24734244 - Effect of video-based versus personalized instruction on errors during elastic tubing e...
22062424 - Effect of pre-slaughter physiological conditions on the oxidative stability of colour a...
9825384 - High-intensity physical training in adults with asthma. a comparison between training o...
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't; Research Support, U.S. Gov't, P.H.S.    
Journal Detail:
Title:  Medical physics     Volume:  31     ISSN:  0094-2405     ISO Abbreviation:  Med Phys     Publication Date:  2004 Jan 
Date Detail:
Created Date:  2004-02-05     Completed Date:  2004-08-24     Revised Date:  2007-11-14    
Medline Journal Info:
Nlm Unique ID:  0425746     Medline TA:  Med Phys     Country:  United States    
Other Details:
Languages:  eng     Pagination:  81-90     Citation Subset:  IM    
Department of Radiology, The University of Chicago, Chicago, Illinois 60637, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Breast Neoplasms / radiography*
Neural Networks (Computer)*
Observer Variation
Radiographic Image Interpretation, Computer-Assisted*
Reproducibility of Results
Grant Support

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  Boron neutron capture therapy of skin melanomas at the RA-6 reactor: a procedural approach to beam s...
Next Document:  Comparison of beam characteristics of a gold x-ray target and a tungsten replacement target.