Abstract
Machine learning-based prediction has been effec6
tively applied for many healthcare applications. Predicting breast
7 screening attendance using machine learning (prior to the actual
8 mammogram) is a new field. This paper presents new predictor
9 attributes for such an algorithm. It describes a new hybrid algo10
rithm that relies on back-propagation and radial basis function11
based neural networks for prediction. The algorithm has been de12
veloped in an open source-based environment. The algorithm was
13 tested on a 13-year dataset (1995–2008). This paper compares the
14 algorithm and validates its accuracy and efficiency with different
15 platforms. Nearly 80%accuracy and 88%positive predictive value
16 and sensitivity were recorded for the algorithm. The results were
17 encouraging; 40–50% of negative predictive value and specificity
18 warrant further work. Preliminary results were promising and
19 provided ample amount of reasons for testing the algorithm on a
20 larger scale.
tively applied for many healthcare applications. Predicting breast
7 screening attendance using machine learning (prior to the actual
8 mammogram) is a new field. This paper presents new predictor
9 attributes for such an algorithm. It describes a new hybrid algo10
rithm that relies on back-propagation and radial basis function11
based neural networks for prediction. The algorithm has been de12
veloped in an open source-based environment. The algorithm was
13 tested on a 13-year dataset (1995–2008). This paper compares the
14 algorithm and validates its accuracy and efficiency with different
15 platforms. Nearly 80%accuracy and 88%positive predictive value
16 and sensitivity were recorded for the algorithm. The results were
17 encouraging; 40–50% of negative predictive value and specificity
18 warrant further work. Preliminary results were promising and
19 provided ample amount of reasons for testing the algorithm on a
20 larger scale.
Original language | English |
---|---|
Pages (from-to) | 251-259 |
Journal | IEEE Transactions on Information Technology in Biomedicine |
Volume | 15 |
Issue number | 2 |
DOIs | |
Publication status | Published - Mar 2011 |
Bibliographical note
© 2011 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Keywords
- breast screening
- cancer
- machine learning
- neural networks
- prediction
- screening attendance