umu.sePublikationer
Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Predictive support recovery with TV-Elastic Net penalty and logistic regression: An application to structural MRI
NeuroSpin, I2BM, CEA.
NeuroSpin, I2BM, CEA.
NeuroSpin, I2BM, CEA.ORCID-id: 0000-0001-7119-7646
Centre d'Acquisition et de Traitement des Images (CATI).
Visa övriga samt affilieringar
2014 (Engelska)Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

The use of machine-learning in neuroimaging offers new perspectives in early diagnosis and prognosis of brain diseases. Although such multivariate methods can capture complex relationships in the data, traditional approaches provide irregular (l12 penalty) or scattered (l1 penalty) predictive pattern with a very limited relevance. A penalty like Total Variation (TV) that exploits the natural 3D structure of the images can increase the spatial coherence of the weight map. However, TV penalization leads to non-smooth optimization problems that are hard to minimize. We propose an optimization framework that minimizes any combination of l1, l2, and TV penalties while preserving the exact l1 penalty. This algorithm uses Nesterov's smoothing technique to approximate the TV penalty with a smooth function such that the loss and the penalties are minimized with an exact accelerated proximal gradient algorithm. We propose an original continuation algorithm that uses successively smaller values of the smoothing parameter to reach a prescribed precision while achieving the best possible convergence rate. This algorithm can be used with other losses or penalties. The algorithm is applied on a classification problem on the ADNI dataset. We observe that the TV penalty does not necessarily improve the prediction but provides a major breakthrough in terms of support recovery of the predictive brain regions.

Ort, förlag, år, upplaga, sidor
IEEE conference proceedings, 2014.
Nyckelord [en]
Smoothing methods, Logistic regression, Approximation algorithms, Convergence, Vectors, Prediction algorithms, Neuroimaging
Nationell ämneskategori
Övrig annan medicin och hälsovetenskap Annan data- och informationsvetenskap
Forskningsämne
datoriserad bildanalys
Identifikatorer
URN: urn:nbn:se:umu:diva-144027DOI: 10.1109/PRNI.2014.6858517ISBN: 978-1-4799-4149-0 (digital)OAI: oai:DiVA.org:umu-144027DiVA, id: diva2:1175482
Konferens
International Workshop on Pattern Recognition in Neuroimaging, Tübingen, June 4-6, 2014
Tillgänglig från: 2018-01-18 Skapad: 2018-01-18 Senast uppdaterad: 2018-06-09

Open Access i DiVA

Fulltext saknas i DiVA

Övriga länkar

Förlagets fulltexthttp://ieeexplore.ieee.org/document/6858517/

Personposter BETA

Löfstedt, Tommy

Sök vidare i DiVA

Av författaren/redaktören
Löfstedt, Tommy
Övrig annan medicin och hälsovetenskapAnnan data- och informationsvetenskap

Sök vidare utanför DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetricpoäng

doi
isbn
urn-nbn
Totalt: 185 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf