Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Morphable Brain Model for Monitoring Disease Related Brain Changes
KTH, School of Computer Science and Communication (CSC).
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Alzheimer's Disease is a devastating neurodegenerative disease that costs billions of dollars each year worldwide. Early diagnosis of the disease can greatly improve patients well-being. There are computer-aided methods that can detect the disease using high quality brain scans. However, these kinds of scans are very rare in the clinic so the methods are not applicable. This paper explores methods for detecting Alzheimer's Disease based on deformation fields of brain scans. This has the advantage of being robust to intensity error, and thus applicable on lower quality scans. The methods were tested using low quality MRI scans, with a slice thickness of 5.5mm. Using a morphable model, which captures the general shape of a brain, the result of classifying diseased and healthy brains had 94/97/92 accuracy/sensitivity/specificity, which is comparable to other methods which use high quality images. This result suggests it may be possible to use methods based on deformation fields in research using clinical data, and possibly for clinical use as well.

Abstract [sv]

Alzheimers sjukdom är en förödande neurodegenerativ sjukdom som kostar hela världen miljarder dollar årligen. Tidig diagnos av sjukdomen kan forbattra patienternas välbennande. Det finns datorstödda metoder som kan upptäcka sjukdomen med hjälp av högkvalitativa hjärnskanningar. Dessa typer av skanningar är mycket sällsynta i den kliniska vardagen och därfor ar de metoderna inte tillämpliga for kliniskt bruk. Denna rapport utforskar metoder for att upptäcka Alzheimers sjukdom baserade på deformationsfält av hjärnskanningar. Detta har fördelen att vara robust för intensitetfel och därmed tillämpligt på hjarnskanningar av lägre kvalitet. Metoderna testades med lågkvalitet MRI hjarnskanningar med en skivtjocklek pa 5,5 mm. Tillämpning av en "morphable" modell, som fångar den generella hjärnformen, gav 94/97/92 noggrannhet/känslighet/specificitet för klassifikation av sjuka och friska hjärnor. Detta ar jämförbart med andra metoder som använder högkvalitativa bilder. Detta resultat antyder att det kan vara möjligt att använda metoder baserade pa deformationsfält i forskning med kliniska data, och ger även möjligheten for klinisk användning.

Place, publisher, year, edition, pages
2015.
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-164967OAI: oai:DiVA.org:kth-164967DiVA: diva2:806699
Supervisors
Examiners
Available from: 2015-04-21 Created: 2015-04-21 Last updated: 2015-04-21Bibliographically approved

Open Access in DiVA

fulltext(1481 kB)63 downloads
File information
File name FULLTEXT01.pdfFile size 1481 kBChecksum SHA-512
c63871827167a9cddb3ef9c863be71e63e0c2383efbfc92280637508e4a21e98a34738d99d95f3369e202512886aa6541496892322445a6ea0ad447501c38e89
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 63 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 225 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf