kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Evaluation of transfer learning on three medical image classification tasks
KTH, School of Electrical Engineering and Computer Science (EECS), Computer Science.
KTH, School of Electrical Engineering and Computer Science (EECS), Computer Science.
2022 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Utvärdering av transfer learning på tre bildklassificeringsproblem (Swedish)
Abstract [en]

Medical image diagnosing is a domain of medicine which has proven suitable for automation through machine learning. Automation would reduce cost and increase efficiency, making healthcare more widely available and decreasing the time needed to arrive at a diagnosis. One limiting aspect of this automation is the amount of available data needed for training. Transfer learning is a technique used to mitigate the negative effects of limited data. In this thesis the effects of using transfer learning on performance was examined on three different datasets: APTOS2019, ISIC2019, and PatchCamelyon. Results showed that on two of the three datasets (APTOS2019, ISIC2019), pretraining had a significant positive impact on performance. On the PatchCamelyon dataset, pretraining had a small positive impact on performance. The relatively small performance gain on PatchCamelyon can be understood by the large size of the dataset.

Abstract [sv]

Medicinsk bilddiagnostik är en domän inom medicin som har visat sig lämplig för automatisering genom maskininlärning. Automatisering skulle minska kostnaderna och öka effektiviteten inom sjukvård, vilket i sin tur skulle göra sjukvård mer tillgänglig samt minska tiden som behövs för att komma fram till en diagnos. En begränsande aspekt av denna automatisering är mängden tillgänglig data som behövs för träning. Transfer learning är en teknik som används för att reducera de negativa effekterna av begränsad data. I detta arbete undersöktes effekterna på prestanda av att använda transfer learning med avseende på tre olika dataset: APTOS2019, ISIC2019 och PatchCamelyon. Resultaten visade att pretraining hade en signifikant positiv inverkan på två av tre dataset (APTOS2019, ISIC2019). På PatchCamelyon hade pretraining en knapp positiv inverkan på prestandan. Den relativt marginella prestandavinsten på PatchCamelyon kan förklaras av datasetets omfattande storlek.

Place, publisher, year, edition, pages
2022. , p. 18
Series
TRITA-EECS-EX ; 2022:496
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-320014OAI: oai:DiVA.org:kth-320014DiVA, id: diva2:1703301
Subject / course
Computer Science
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2022-10-13 Created: 2022-10-12 Last updated: 2022-10-13Bibliographically approved

Open Access in DiVA

fulltext(2692 kB)232 downloads
File information
File name FULLTEXT01.pdfFile size 2692 kBChecksum SHA-512
fe3e331b3101355453cdc49787f765ea247f41afff140b6d281eca5a85bde75215eccccc27ba71824fb8072912e0b40d723bdc00e522202b8ce67d7624722404
Type fulltextMimetype application/pdf

By organisation
Computer Science
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 232 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: 322 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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