Improved Prediction of Cognitive Outcomes via Globally Aligned Imaging Biomarker Enrichments Over Progressions

Lyujian Lu, Saad Elbeleidy, Lauren Baker, Hua Wang, Heng Huang, Li Shen

MICCAI - 2019

Alzheimer's Disease (AD) is a severe progressive neurodegenerative disorder, threatening the health of millions of people. Many longitudinal prediction models have been proposed to study clinical scores for automatic AD diagnosis. For accurate clinical score prediction, one major challenge is missing patient temporal neuroimaging data over AD progressions. To address this issue, we present an unsupervised method to learn enriched imaging biomarker representation that can simultaneously capture all the baseline neuroimaging measures and progressive variations of the available follow-up measurements of each participant. Our experiments on Alzheimer's Disease Neuroimaging Initiative (ADNI) demonstrate that the proposed approach not only achieves improved regression performance, but also successfully identifies disease relevant biomarkers.

Links

Cite this paper

MLA Copied to clipboard!
Lu, Lyujian, et al. "Improved Prediction of Cognitive Outcomes via Globally Aligned Imaging Biomarker Enrichments over Progressions." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2019.
BibTeX Copied to clipboard!
@inproceedings{lu2019improved,
  title={Improved Prediction of Cognitive Outcomes via Globally Aligned Imaging Biomarker Enrichments over Progressions},
  author={Lu, Lyujian and Elbeleidy, Saad and Baker, Lauren and Wang, Hua and Huang, Heng and Shen, Li and others},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={140--148},
  year={2019},
  organization={Springer}
}