Non-Binary Approaches for Classification of Amyloid Brain PET

Katherine Zukotvnski, Vincent C. Gaudet, Phillip Kuo, Sabrina Adamo, Maged Goubran, Christian Bocti, Michael Borrie, Howard Chertkow, Richard Frayne, Robin Hsiung, Robert Laforce, Michael D. Noseworthy, Frank S. Prato, Jim D. Sahlas, Christopher Scott, Eric E. Smith, Vesna Sossi, Alex Thiel, Jean Paul Soucy, Jean Claude TardifSandra E. Black

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Machine learning (ML) is increasingly used in medical imaging. This paper provides pilot data of decision trees and random forests (RFs) to predict if a 18F-florbetapir brain positron emission tomography (PET) is positive or negative for amyloid deposition based on quantitative data analysis. The dataset included 55 18F-florbetapir brain PETs in participants with severe white matter disease and mild cognitive impairment (MCI), early Alzheimer's disease (AD) or transient ischemic events. The Montreal Cognitive Assessment (MoCA) score was known for each participant. All PET images were processed using the MINC toolkit to extract standardized uptake value ratios (SUVRs) for 59 regions of interest (features). Each PET was clinically read by 2 dual certified radiology/nuclear medicine physicians with final interpretation based on consensus. An initial study of RFs using conventional binary decision trees and PET quantitation suggests this is a powerful algorithm for PET classification as positive or negative for amyloid deposition. Preliminary data did not show improved results when a ternary RF approach was used. Finally, a soft-decision approach may be helpful to predict the MoCA score.

Original languageEnglish (US)
Title of host publicationProceedings - 2019 IEEE 49th International Symposium on Multiple-Valued Logic, ISMVL 2019
PublisherIEEE Computer Society
Pages206-211
Number of pages6
ISBN (Electronic)9781728100913
DOIs
StatePublished - May 2019
Event49th IEEE International Symposium on Multiple-Valued Logic, ISMVL 2019 - Fredericton, Canada
Duration: May 21 2019May 23 2019

Publication series

NameProceedings of The International Symposium on Multiple-Valued Logic
Volume2019-May
ISSN (Print)0195-623X

Conference

Conference49th IEEE International Symposium on Multiple-Valued Logic, ISMVL 2019
CountryCanada
CityFredericton
Period5/21/195/23/19

Keywords

  • brain imaging
  • decision tree
  • dementia
  • machine learning
  • multiple-valued logic
  • nuclear medicine
  • positron emission tomography
  • random forest

ASJC Scopus subject areas

  • Computer Science(all)
  • Mathematics(all)

Fingerprint Dive into the research topics of 'Non-Binary Approaches for Classification of Amyloid Brain PET'. Together they form a unique fingerprint.

  • Cite this

    Zukotvnski, K., Gaudet, V. C., Kuo, P., Adamo, S., Goubran, M., Bocti, C., Borrie, M., Chertkow, H., Frayne, R., Hsiung, R., Laforce, R., Noseworthy, M. D., Prato, F. S., Sahlas, J. D., Scott, C., Smith, E. E., Sossi, V., Thiel, A., Soucy, J. P., ... Black, S. E. (2019). Non-Binary Approaches for Classification of Amyloid Brain PET. In Proceedings - 2019 IEEE 49th International Symposium on Multiple-Valued Logic, ISMVL 2019 (pp. 206-211). [8758715] (Proceedings of The International Symposium on Multiple-Valued Logic; Vol. 2019-May). IEEE Computer Society. https://doi.org/10.1109/ISMVL.2019.00043