Meta-Weighted Gaussian Process Experts for Personalized Forecasting of AD Cognitive Changes

Ognjen (Oggi) Rudovic, Yuria Utsumi, Ricardo Guerrero, Kelly Peterson, Daniel Rueckert, Rosalind W. Picard
; Proceedings of the 4th Machine Learning for Healthcare Conference, PMLR 106:181-196, 2019.

Abstract

We introduce a novel personalized Gaussian Process Experts (pGPE) model for predicting per-subject ADAS-Cog13 cognitive scores – a significant predictor of Alzheimer’s Disease (AD) in the cognitive domain – over the future 6, 12, 18, and 24 months. We start by training a population-level model using multi-modal data from previously seen subjects using a base Gaussian Process (GP) regression. Then, we personalize this model by adapting the base GP sequentially over time to a new (target) subject using domain adaptive GPs, and also by training subject-specific GP. While we show that these models achieve improved performance when selectively applied to the forecasting task (one performs better than the other on different subjects/visits), the average performance per model is suboptimal. To this end, we used the notion of meta learning in the proposed pGPE to design a regression-based weighting of these expert models, where the expert weights are optimized for each subject and his/her future visit. The results on a cohort of subjects from the ADNI dataset show that this newly introduced personalized weighting of the expert models leads to large improvements in accurately forecasting future ADAS-Cog13 scores and their fine-grained changes associated with the AD progression. This approach has potential to help identify at-risk patients early and improve the construction of clinical trials for AD.

Cite this Paper


BibTeX
@InProceedings{pmlr-v106-rudovic19a, title = {Meta-Weighted Gaussian Process Experts for Personalized Forecasting of AD Cognitive Changes}, author = {Rudovic, Ognjen (Oggi) and Utsumi, Yuria and Guerrero, Ricardo and Peterson, Kelly and Rueckert, Daniel and Picard, Rosalind W.}, pages = {181--196}, year = {2019}, editor = {Finale Doshi-Velez and Jim Fackler and Ken Jung and David Kale and Rajesh Ranganath and Byron Wallace and Jenna Wiens}, volume = {106}, series = {Proceedings of Machine Learning Research}, address = {Ann Arbor, Michigan}, month = {09--10 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v106/rudovic19a/rudovic19a.pdf}, url = {http://proceedings.mlr.press/v106/rudovic19a.html}, abstract = {We introduce a novel personalized Gaussian Process Experts (pGPE) model for predicting per-subject ADAS-Cog13 cognitive scores – a significant predictor of Alzheimer’s Disease (AD) in the cognitive domain – over the future 6, 12, 18, and 24 months. We start by training a population-level model using multi-modal data from previously seen subjects using a base Gaussian Process (GP) regression. Then, we personalize this model by adapting the base GP sequentially over time to a new (target) subject using domain adaptive GPs, and also by training subject-specific GP. While we show that these models achieve improved performance when selectively applied to the forecasting task (one performs better than the other on different subjects/visits), the average performance per model is suboptimal. To this end, we used the notion of meta learning in the proposed pGPE to design a regression-based weighting of these expert models, where the expert weights are optimized for each subject and his/her future visit. The results on a cohort of subjects from the ADNI dataset show that this newly introduced personalized weighting of the expert models leads to large improvements in accurately forecasting future ADAS-Cog13 scores and their fine-grained changes associated with the AD progression. This approach has potential to help identify at-risk patients early and improve the construction of clinical trials for AD.} }
Endnote
%0 Conference Paper %T Meta-Weighted Gaussian Process Experts for Personalized Forecasting of AD Cognitive Changes %A Ognjen (Oggi) Rudovic %A Yuria Utsumi %A Ricardo Guerrero %A Kelly Peterson %A Daniel Rueckert %A Rosalind W. Picard %B Proceedings of the 4th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2019 %E Finale Doshi-Velez %E Jim Fackler %E Ken Jung %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v106-rudovic19a %I PMLR %J Proceedings of Machine Learning Research %P 181--196 %U http://proceedings.mlr.press %V 106 %W PMLR %X We introduce a novel personalized Gaussian Process Experts (pGPE) model for predicting per-subject ADAS-Cog13 cognitive scores – a significant predictor of Alzheimer’s Disease (AD) in the cognitive domain – over the future 6, 12, 18, and 24 months. We start by training a population-level model using multi-modal data from previously seen subjects using a base Gaussian Process (GP) regression. Then, we personalize this model by adapting the base GP sequentially over time to a new (target) subject using domain adaptive GPs, and also by training subject-specific GP. While we show that these models achieve improved performance when selectively applied to the forecasting task (one performs better than the other on different subjects/visits), the average performance per model is suboptimal. To this end, we used the notion of meta learning in the proposed pGPE to design a regression-based weighting of these expert models, where the expert weights are optimized for each subject and his/her future visit. The results on a cohort of subjects from the ADNI dataset show that this newly introduced personalized weighting of the expert models leads to large improvements in accurately forecasting future ADAS-Cog13 scores and their fine-grained changes associated with the AD progression. This approach has potential to help identify at-risk patients early and improve the construction of clinical trials for AD.
APA
Rudovic, O.(., Utsumi, Y., Guerrero, R., Peterson, K., Rueckert, D. & Picard, R.W.. (2019). Meta-Weighted Gaussian Process Experts for Personalized Forecasting of AD Cognitive Changes. Proceedings of the 4th Machine Learning for Healthcare Conference, in PMLR 106:181-196

Related Material