FLARe: Forecasting by Learning Anticipated Representations
This work addresses patient-level forecasting for Alzheimer's disease, aiding early intervention and treatment planning, but it is incremental as it builds on existing latent representation methods.
The paper tackles forecasting Alzheimer's disease progression by proposing a model that generates sequences of latent representations across time horizons, outperforming baselines in accuracy and F1 score while handling missing visits robustly.
Computational models that forecast the progression of Alzheimer's disease at the patient level are extremely useful tools for identifying high risk cohorts for early intervention and treatment planning. The state-of-the-art work in this area proposes models that forecast by using latent representations extracted from the longitudinal data across multiple modalities, including volumetric information extracted from medical scans and demographic info. These models incorporate the time horizon, which is the amount of time between the last recorded visit and the future visit, by directly concatenating a representation of it to the data latent representation. In this paper, we present a model which generates a sequence of latent representations of the patient status across the time horizon, providing more informative modeling of the temporal relationships between the patient's history and future visits. Our proposed model outperforms the baseline in terms of forecasting accuracy and F1 score with the added benefit of robustly handling missing visits.