Paper Info
Title
Impact Of Missing Data In Training Artificial Neural Networks For Computer-Aided Diagnosis
Abstract
Artificial neural networks (ANN) are frequently used in the development of Computer-Aided Diagnosis systems for breast cancer detection and diagnosis. One class of models uses descriptions of mammographic lesions encoded following the BI-RADS (TM) lexicon. Data sets that have been carefully curated to ensure completeness are generally used; however, in routine practice, some information is typically missing in clinical databases. The impact of missing data on the performance of a feed-forward, back-propagation ANN, as measured by the area under the Receiver Operating Characteristic curve, was found to be much higher when data were missing from the testing set than when data were missing from the training set. This empirical study highlights the need for additional research on developing robust clinical decision support systems for realistic environments in which key information may be unknown or inaccessible.
Year
Venue
Keywords
2004
PROCEEDINGS OF THE 2004 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA'04)
missing data,breast cancer,clinical decision support system,data systems,predictive models,artificial neural network,feed forward,artificial neural networks,pathology,empirical study,back propagation,intelligent networks,receiver operating characteristic curve
Field
DocType
Citations 
Data mining,Data set,Receiver operating characteristic,Computer science,Data system,Computer-aided diagnosis,Artificial intelligence,Missing data,Clinical decision support system,Artificial neural network,Empirical research,Machine learning
Conference
4
PageRank 
References 
Authors
0.63
2
2
Name
Order
Citations
PageRank
Mia K. Markey135333.66
Amit C. Patel2111.35