Document Detail


Dependence of computer classification of clustered microcalcifications on the correct detection of microcalcifications.
MedLine Citation:
PMID:  11585226     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
Our purpose was to study the dependence of computer performance in classifying clustered microcalcifications as malignant or benign on the correct detection of microcalcifications. Specifically, we studied the effects of computer-detected true-positive microcalcifications and computer-detected false-positive microcalcifications in true microcalcification clusters. Using a database of 100 mammograms, we compared computer classification performance obtained from computer-detected microcalcifications to (1) computer classification performance obtained from manually identified microcalcifications, and (2) radiologists' performance. When an artificial neural network (ANN) was trained with manually identified microcalcifications, computer classification performance was comparable to or better than radiologists' performance as the number of computer-detected true-positive microcalcifications decreased to 40% and as the number of computer-detected false-positive microcalcifications increased to 50%. Further loss in computer-detected true-positive microcalcifications degraded classification performance substantially. Moreover, training the ANN with computer-detected microcalcifications also degraded computer classification performance. These results show that computer performance in classifying clustered microcalcifications as malignant or benign is insensitive to moderate decreases in computer-detected true-positive microcalcifications and moderate increases in computer-detected false-positive microcalcifications.
Authors:
Y Jiang; R M Nishikawa; J Papaioannou
Related Documents :
22652476 - A pilot evaluation study of high resolution digital thermal imaging in the assessment o...
14502466 - Chemical machine vision: automated extraction of chemical metadata from raster images.
19443926 - Supervised learning of quantizer codebooks by information loss minimization.
19507676 - Correlations among acoustic, texture and morphological features for breast ultrasound cad.
11071376 - New modalities for imaging the fetal heart.
19719726 - Validation of methodologies for quantifying isolated human islets: an islet cell resour...
Publication Detail:
Type:  Comparative Study; Journal Article; Research Support, Non-U.S. Gov't; Research Support, U.S. Gov't, P.H.S.    
Journal Detail:
Title:  Medical physics     Volume:  28     ISSN:  0094-2405     ISO Abbreviation:  Med Phys     Publication Date:  2001 Sep 
Date Detail:
Created Date:  2001-10-04     Completed Date:  2002-03-25     Revised Date:  2007-11-14    
Medline Journal Info:
Nlm Unique ID:  0425746     Medline TA:  Med Phys     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1949-57     Citation Subset:  IM    
Affiliation:
Department of Radiology, The University of Chicago, Illinois 60637, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Breast Neoplasms / diagnosis,  radiography
Calcinosis / classification*,  diagnosis*,  radiography
Databases, Factual
Diagnosis, Computer-Assisted*
False Positive Reactions
Female
Humans
Mammography
Radiographic Image Enhancement
Radiographic Image Interpretation, Computer-Assisted
Grant Support
ID/Acronym/Agency:
R01 CA 60187/CA/NCI NIH HHS

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


Previous Document:  Selection of an optimal neural network architecture for computer-aided detection of microcalcificati...
Next Document:  Microwave-induced thermoacoustic tomography using multi-sector scanning.