SVM-based glioma grading: optimization by feature reduction analysis
We investigated the predictive power of feature reduction analysis approaches in support vector machine (SVM)-based classification of glioma grade. In 101 untreated glioma patients, three analytic approaches were evaluated to derive an optimal reduction in features; (i) Pearson's correlation co...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
13 April 2012
|
| In: |
Zeitschrift für medizinische Physik
Year: 2012, Jahrgang: 22, Heft: 3, Pages: 205-214 |
| ISSN: | 1876-4436 |
| DOI: | 10.1016/j.zemedi.2012.03.007 |
| Online-Zugang: | Verlag, Volltext: http://dx.doi.org/10.1016/j.zemedi.2012.03.007 Verlag, Volltext: http://www.sciencedirect.com/science/article/pii/S0939388912000360 
|
| Verfasserangaben: | Frank G. Zöllner, Kyrre E. Emblem, Lothar R. Schad |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1580285511 | ||
| 003 | DE-627 | ||
| 005 | 20230427065653.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 180822s2012 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.zemedi.2012.03.007 |2 doi | |
| 035 | |a (DE-627)1580285511 | ||
| 035 | |a (DE-576)510285511 | ||
| 035 | |a (DE-599)BSZ510285511 | ||
| 035 | |a (OCoLC)1341017078 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Zöllner, Frank G. |d 1976- |e VerfasserIn |0 (DE-588)129580015 |0 (DE-627)473357054 |0 (DE-576)297732587 |4 aut | |
| 245 | 1 | 0 | |a SVM-based glioma grading |b optimization by feature reduction analysis |c Frank G. Zöllner, Kyrre E. Emblem, Lothar R. Schad |
| 264 | 1 | |c 13 April 2012 | |
| 300 | |a 10 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Gesehen am 22.08.2018 | ||
| 520 | |a We investigated the predictive power of feature reduction analysis approaches in support vector machine (SVM)-based classification of glioma grade. In 101 untreated glioma patients, three analytic approaches were evaluated to derive an optimal reduction in features; (i) Pearson's correlation coefficients (PCC), (ii) principal component analysis (PCA) and (iii) independent component analysis (ICA). Tumor grading was performed using a previously reported SVM approach including whole-tumor cerebral blood volume (CBV) histograms and patient age. Best classification accuracy was found using PCA at 85% (sensitivity=89%, specificity=84%) when reducing the feature vector from 101 (100-bins rCBV histogram+age) to 3 principal components. In comparison, classification accuracy by PCC was 82% (89%, 77%, 2 dimensions) and 79% by ICA (87%, 75%, 9 dimensions). For improved speed (up to 30%) and simplicity, feature reduction by all three methods provided similar classification accuracy to literature values (∼87%) while reducing the number of features by up to 98%. - Zusammenfassung - In dieser Studie wurde untersucht, ob Merkmalsreduktion die Klassifikation (Grading) von Gehirntumoren verbessert. Insgesamt umfasste unser Datensatz 101 Patienten mit unbehandelten Glioma. Drei Merkmalsextraktionsverfahren (Pearson's cross correlation, principal component analysis (PCA) und independent component analysis (ICA)) wurden angewendet. Tumorklassifikation erfolgte über eine Support-Vektor-Maschine (SVM). Als Merkmale dienten Histogramme über das cerebrale Blutvolumen des Tumors und Patientenalter. Die höchste Klassifikationsgüte von 85% (Sensitivität=89%, Spezifität=84%) konnte mit PCA erreicht werden. Die Anzahl der Merkmale konnte dabei von 101 (100-Bin Histogramm+Patientenalter) auf 3 Hauptkomponenten reduziert werden. Im Vergleich erreichte die PCC eine korrekte Klassifizierung von 82% (89%, 77%, 2 Dimensionen) und die ICA von 79% (87%, 75%, 9 Dimensionen). Bei gleichbleibender Klassifikationsgüte im Vergleich zur Literatur (∼ 87%), konnten eine Reduktion von bis 98% der Merkmalesdimension als auch eine Reduktion der Berechnungszeit von bis zu 30% erreicht werden. | ||
| 650 | 4 | |a CAD | |
| 650 | 4 | |a Computer aided diagnostics (CAD) | |
| 650 | 4 | |a DSC-MRI | |
| 650 | 4 | |a Feature reduction | |
| 650 | 4 | |a Glioma grading | |
| 650 | 4 | |a Glioma-Grading | |
| 650 | 4 | |a Machine learning | |
| 650 | 4 | |a Maschinelles Lernen | |
| 650 | 4 | |a Merkmalsextraktion | |
| 650 | 4 | |a Perfusion MRI | |
| 650 | 4 | |a Support vector machines | |
| 650 | 4 | |a Support-Vektor-Maschine | |
| 700 | 1 | |a Schad, Lothar R. |d 1956- |e VerfasserIn |0 (DE-588)1028817630 |0 (DE-627)731640241 |0 (DE-576)376271221 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Zeitschrift für medizinische Physik |d Amsterdam [u.a.] : Elsevier, 1990 |g 22(2012), 3, Seite 205-214 |w (DE-627)510617662 |w (DE-600)2231492-1 |w (DE-576)272350818 |x 1876-4436 |7 nnas |a SVM-based glioma grading optimization by feature reduction analysis |
| 773 | 1 | 8 | |g volume:22 |g year:2012 |g number:3 |g pages:205-214 |g extent:10 |a SVM-based glioma grading optimization by feature reduction analysis |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.zemedi.2012.03.007 |x Verlag |x Resolving-System |3 Volltext |
| 856 | 4 | 0 | |u http://www.sciencedirect.com/science/article/pii/S0939388912000360 |x Verlag |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20180822 | ||
| 993 | |a Article | ||
| 994 | |a 2012 | ||
| 998 | |g 1028817630 |a Schad, Lothar R. |m 1028817630:Schad, Lothar R. |p 3 |y j | ||
| 998 | |g 129580015 |a Zöllner, Frank G. |m 129580015:Zöllner, Frank G. |p 1 |x j | ||
| 999 | |a KXP-PPN1580285511 |e 3022942370 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"type":{"media":"Online-Ressource","bibl":"article-journal"},"title":[{"subtitle":"optimization by feature reduction analysis","title_sort":"SVM-based glioma grading","title":"SVM-based glioma grading"}],"note":["Gesehen am 22.08.2018"],"origin":[{"dateIssuedDisp":"13 April 2012","dateIssuedKey":"2012"}],"id":{"eki":["1580285511"],"doi":["10.1016/j.zemedi.2012.03.007"]},"recId":"1580285511","relHost":[{"title":[{"title":"Zeitschrift für medizinische Physik","title_sort":"Zeitschrift für medizinische Physik","subtitle":"offizielles Organ der Deutschen, Österreichischen und Schweizerischen Gesellschaft für Medizinische Physik"}],"origin":[{"publisher":"Elsevier","dateIssuedKey":"1990","dateIssuedDisp":"1990-","publisherPlace":"Amsterdam [u.a.]"}],"note":["Gesehen am 31.01.2022"],"pubHistory":["Volume 1, supplement (1990) ; volume 1, issue 1 (1991)-"],"disp":"SVM-based glioma grading optimization by feature reduction analysisZeitschrift für medizinische Physik","type":{"bibl":"periodical","media":"Online-Ressource"},"recId":"510617662","id":{"zdb":["2231492-1"],"issn":["1876-4436"],"eki":["510617662"]},"corporate":[{"display":"Deutsche Gesellschaft für Medizinische Physik","role":"isb"},{"display":"Österreichische Gesellschaft für Medizinische Physik","role":"isb"}],"language":["ger"],"part":{"extent":"10","volume":"22","year":"2012","pages":"205-214","issue":"3","text":"22(2012), 3, Seite 205-214"},"titleAlt":[{"title":"Medizinische Physik"}]}],"physDesc":[{"extent":"10 S."}],"name":{"displayForm":["Frank G. Zöllner, Kyrre E. Emblem, Lothar R. Schad"]},"person":[{"given":"Frank G.","role":"aut","family":"Zöllner","display":"Zöllner, Frank G."},{"role":"aut","given":"Lothar R.","family":"Schad","display":"Schad, Lothar R."}],"language":["eng"]} | ||
| SRT | |a ZOELLNERFRSVMBASEDGL1320 | ||