Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A robust multi-utility neural network technique integrated with discriminators for bone health decisioning to facilitate clinical-driven processes

A robust multi-utility neural network technique integrated with discriminators for bone health... PurposeOsteoporosis (OP) is a malformation of the bones caused by the loss of bone mass and its mineral density, and also deterioration in bone quality or structures, causing an increased risk of fractures. Apart from bone cracks, damage to hip, spine vertebrae and wrist are the most prevalent.MethodThis work proposes a hybrid strategy to categorise the normal and abnormal bone mineral density (BMD) values obtained from 140 patients. Owing to the smaller sample and the large number of variables in a high-dimensional data classification issue, classical linear discriminant analysis (LDA) performs poorly, resulting in the instability and singularity of the sample covariance matrix. To suppress this, a normalised radial basis function neural network (NRBFNN) coupled with a modified version of LDA (MLDA) is being proposed through this paper for BMD evaluations. In this study, we propose a novel modified version of LDA (MLDA) based on a robust estimator for high-dimensional covariance matrices, and has been proved to be asymptotically highly reliable than the sample covariance matrix.ResultsInstead of employing un-normalised RBF, NRBFNN is used for BMD classification. The suggested hybrid algorithm MLDA-NRBFNN achieved a classification accuracy of 97.5% on comparing with ANN, PAC-ANN and RBFN.ConclusionWith the intervention of the proposed method, the doctors could tend to have a precursor to assay the patients with varied bone ailments and to understand the recovery impact of knee replacement as well. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Research on Biomedical Engineering Springer Journals

A robust multi-utility neural network technique integrated with discriminators for bone health decisioning to facilitate clinical-driven processes

Loading next page...
 
/lp/springer-journals/a-robust-multi-utility-neural-network-technique-integrated-with-n2XWs7umRh

References (25)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to The Brazilian Society of Biomedical Engineering 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
ISSN
2446-4732
eISSN
2446-4740
DOI
10.1007/s42600-023-00259-x
Publisher site
See Article on Publisher Site

Abstract

PurposeOsteoporosis (OP) is a malformation of the bones caused by the loss of bone mass and its mineral density, and also deterioration in bone quality or structures, causing an increased risk of fractures. Apart from bone cracks, damage to hip, spine vertebrae and wrist are the most prevalent.MethodThis work proposes a hybrid strategy to categorise the normal and abnormal bone mineral density (BMD) values obtained from 140 patients. Owing to the smaller sample and the large number of variables in a high-dimensional data classification issue, classical linear discriminant analysis (LDA) performs poorly, resulting in the instability and singularity of the sample covariance matrix. To suppress this, a normalised radial basis function neural network (NRBFNN) coupled with a modified version of LDA (MLDA) is being proposed through this paper for BMD evaluations. In this study, we propose a novel modified version of LDA (MLDA) based on a robust estimator for high-dimensional covariance matrices, and has been proved to be asymptotically highly reliable than the sample covariance matrix.ResultsInstead of employing un-normalised RBF, NRBFNN is used for BMD classification. The suggested hybrid algorithm MLDA-NRBFNN achieved a classification accuracy of 97.5% on comparing with ANN, PAC-ANN and RBFN.ConclusionWith the intervention of the proposed method, the doctors could tend to have a precursor to assay the patients with varied bone ailments and to understand the recovery impact of knee replacement as well.

Journal

Research on Biomedical EngineeringSpringer Journals

Published: Mar 1, 2023

Keywords: Osteoporosis (OP); Bone mineral density (BMD); Modified linear discriminant analysis (MLDA); Normalised radial basis function neural network (NRBFNN); T-score; Z-score

There are no references for this article.