Non-linear least squares features transformation for improving the performance of probabilistic neural networks in classifying human brain tumors on MRI

dc.contributor.author	Γεωργιάδης, Παντελής	el
dc.contributor.author	Κάβουρας, Διονύσης Α.	el
dc.contributor.author	Καλατζής, Ιωάννης	el
dc.contributor.author	Δασκαλάκης, Αντώνης	el
dc.contributor.author	Καγκάδης, Γεώργιος Χ.	el
dc.date.accessioned	2015-05-03T10:38:31Z
dc.date.available	2015-05-03T10:38:31Z
dc.date.issued	2015-05-03
dc.identifier.uri	http://hdl.handle.net/11400/9539
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ηνωμένες Πολιτείες	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/us/	*
dc.source	http://link.springer.com/chapter/10.1007%2F978-3-540-74484-9_21	en
dc.subject	Applications
dc.subject	Magnetic resonance imaging
dc.subject	Εφαρμογές
dc.subject	Μαγνητική τομογραφία
dc.title	Non-linear least squares features transformation for improving the performance of probabilistic neural networks in classifying human brain tumors on MRI	en
heal.type	conferenceItem
heal.generalDescription	Proceedings of the International conference Computational Science and Its Applications – ICCSA 2007. Kuala Lumpur, Malaysia, 26th-29th August 2007. Springer Berlin Heidelberg: 2007. vol. 4707. Part III. pp. 239-247.	en
heal.classification	Medicine
heal.classification	Neural computers
heal.classification	Ιατρική
heal.classification	Νευρωνικό δίκτυο
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh00006614
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh87008041
heal.classificationURI	N/A-Ιατρική
heal.classificationURI	N/A-Νευρωνικό δίκτυο
heal.keywordURI	http://id.loc.gov/authorities/subjects/sh85079741
heal.contributorName	Σηφάκη, Κοραλία	el
heal.contributorName	Μάλαμας, Μενέλαος	el
heal.contributorName	Νικηφορίδης, Γεώργιος Χ.	el
heal.contributorName	Σολωμού, Αικατερίνη	el
heal.identifier.secondary	DOI 10.1007/978-3-540-74484-9_21
heal.language	en
heal.access	campus
heal.recordProvider	Τ.Ε.Ι. Αθήνας. Σχολή Τεχνολογικών Εφαρμογών. Τμήμα Μηχανικών Βιοϊατρικής Τεχνολογίας Τ.Ε.	el
heal.publicationDate	2007
heal.bibliographicCitation	Georgiadis, P., Cavouras, D., Kalatzis, I., Daskalakis, A., Kagadis, G., et al. (2007). Non-linear least squares features transformation for improving the performance of probabilistic neural networks in classifying human brain tumors on MRI. In the International conference Computational Science and Its Applications – ICCSA 2007. Kuala Lumpur, Malaysia, 26th-29th August 2007. Springer Berlin Heidelberg: 2007.	en
heal.abstract	The aim of the present study was to design, implement, and evaluate a software system for discriminating between metastases, meningiomas, and gliomas on MRI. The proposed classifier is a modified probabilistic neural network (PNN), incorporating a second degree least squares features transformation (LSFT) into the PNN classifier. Thirty-six textural features were extracted from each one of 75 T1-weighted post-contrast MR images (24 metastases, 21 meningiomas, and 30 gliomas). Classification performance was evaluated employing the leave-one-out method and for all possible textural feature combinations. LSFT enhanced the performance of the PNN, achieving 93.33% in discriminating between the three major types of human brain tumors, against 89.33% scored by the PNN alone. Best feature combination for achieving highest discrimination power included the mean value and entropy, which reflect specific properties of texture, i.e. signal strength and inhomogeneity. LSFT improved PNN performance, increased class separability, and resulted in dimensionality reduction.	en
heal.publisher	Springer Berlin Heidelberg	en
heal.fullTextAvailability	true
heal.conferenceName	International conference Computational Science and Its Applications – ICCSA 2007	en
heal.conferenceItemType	full paper