
Thanks to ongoing support from Alzheimer’s Research UK, our lab has been conducting ongoing research into the application of machine learning to Alzheimer’s—a disease that is too often overlooked, despite affecting roughly 1 in 14 people over the age of 65, and 1 in every 6 people over the age of 80 (according to the UK’s NHS).
It was very gratifying to see Bill Gates present a compelling summary, in a post on his GatesNotes blog earlier this week, of the enormous challenges facing researchers and clinical professionals in attempting to understand, mitigate, and treat Alzheimer’s.
We agree with Gates that data could hold the key to research breakthroughs for Alzheimer’s, and that we are finally moving beyond the widespread data availability issues that have crippled progress so far. We have Gates himself to thank for much of this recent progress, due to his role in the creation of the Alzheimer’s Disease Data Initiative (ADDI), which hosts the Alzheimer’s Disease workbench, a resource for collaboration and data sharing.
In providing the following overview of our lab’s key research projects, we hope to demonstrate that machine learning, driven by data, can offer powerful new tools in the fight against Alzheimer’s.
Side-note: all of the projects below made use of data provided through the open-access Alzheimer’s Disease Neuroimaging Initiative (ADNI) study, which tracks disease progression for over 1,700 patients.
Temporal Phenotyping using Deep Predicting Clustering of Disease Progression
Changhee Lee, Mihaela van der Schaar
ICML 2020
Abstract
Due to the wider availability of modern electronic health records, patient care data is often being stored in the form of time-series. Clustering such time-series data is crucial for patient phenotyping, anticipating patients’ prognoses by identifying “similar” patients, and designing treatment guidelines that are tailored to homogeneous patient subgroups.
In this paper, we develop a deep learning approach for clustering time-series data, where each cluster comprises patients who share similar future outcomes of interest (e.g., adverse events, the onset of comorbidities). To encourage each cluster to have homogeneous future outcomes, the clustering is carried out by learning discrete representations that best describe the future outcome distribution based on novel loss functions.
Experiments on two real-world datasets show that our model achieves superior clustering performance over state-of-the-art benchmarks and identifies meaningful clusters that can be translated into actionable information for clinical decision-making.
Target-Embedding Autoencoders for Supervised Representation Learning
Daniel Jarrett, Mihaela van der Schaar
ICLR 2020
Abstract
Autoencoder-based learning has emerged as a staple for disciplining representations in unsupervised and semi-supervised settings.
This paper analyzes a framework for improving generalization in a purely supervised setting, where the target space is high-dimensional. We motivate and formalize the general framework of targetembedding autoencoders (TEA) for supervised prediction, learning intermediate latent representations jointly optimized to be both predictable from features as well as predictive of targets—encoding the prior that variations in targets are driven by a compact set of underlying factors.
As our theoretical contribution, we provide a guarantee of generalization for linear TEAs by demonstrating uniform stability, interpreting the benefit of the auxiliary reconstruction task as a form of regularization.
As our empirical contribution, we extend validation of this approach beyond existing static classification applications to multivariate sequence forecasting, verifying their advantage on both linear and nonlinear recurrent architectures—thereby underscoring the further generality of this framework beyond feedforward instantiations.
ASAC: Active Sensing using Actor-Critic models
Jinsung Yoon, James Jordon, Mihaela van der Schaar
MLHC 2019
Abstract
Deciding what and when to observe is critical when making observations is costly. In a medical setting where observations can be made sequentially, making these observations (or not) should be an active choice. We refer to this as the active sensing problem.
In this paper, we propose a novel deep learning framework, which we call ASAC (Active Sensing using Actor-Critic models) to address this problem. ASAC consists of two networks: a selector network and a predictor network. The selector network uses previously selected observations to determine what should be observed in the future. The predictor network uses the observations selected by the selector network to predict a label, providing feedback to the selector network (well-selected variables should be predictive of the label). The goal of the selector network is then to select variables that balance the cost of observing the selected variables with their predictive power; we wish to preserve the conditional label distribution.
During training, we use the actor-critic models to allow the loss of the selector to be “back-propagated” through the sampling process. The selector network “acts” by selecting future observations to make. The predictor network acts as a “critic” by feeding predictive errors for the selected variables back to the selector network.
In our experiments, we show that ASAC significantly outperforms state-of-the-arts in two real-world medical datasets.
Dynamic Prediction in Clinical Survival Analysis Using Temporal Convolutional Networks
Daniel Jarrett, Jinsung Yoon, Mihaela van der Schaar
Published in IEEE journal of biomedical and health informatics, 2019
Abstract
Accurate prediction of disease trajectories is critical for early identification and timely treatment of patients at risk. Conventional methods in survival analysis are often constrained by strong parametric assumptions and limited in their ability to learn from high-dimensional data.
This paper develops a novel convolutional approach that addresses the drawbacks of both traditional statistical approaches as well as recent neural network models for survival. We present Match-Net: a missingness-aware temporal convolutional hitting-time network, designed to capture temporal dependencies and heterogeneous interactions in covariate trajectories and patterns of missingness.
To the best of our knowledge, this is the first investigation of temporal convolutions in the context of dynamic prediction for personalized risk prognosis. Using real-world data from the Alzheimer’s disease neuroimaging initiative, we demonstrate state-of-the-art performance without making any assumptions regarding underlying longitudinal or time-to-event processes-attesting to the model’s potential utility in clinical decision support.
MATCH-Net: Dynamic Prediction in Survival Analysis using Convolutional Neural Networks
Daniel Jarrett, Jinsung Yoon, Mihaela van der Schaar
NeurIPS Machine Learning for Health Workshop 2018
Abstract
Accurate prediction of disease trajectories is critical for early identification and timely treatment of patients at risk. Conventional methods in survival analysis are often constrained by strong parametric assumptions and limited in their ability to learn from high-dimensional data, while existing neural network models are not readily-adapted to the longitudinal setting.
This paper develops a novel convolutional approach that addresses these drawbacks. We present MATCH-Net: a Missingness-Aware Temporal Convolutional Hitting-time Network, designed to capture temporal dependencies and heterogeneous interactions in covariate trajectories and patterns of missingness.
To the best of our knowledge, this is the first investigation of temporal convolutions in the context of dynamic prediction for personalized risk prognosis. Using real-world data from the Alzheimer’s Disease Neuroimaging Initiative, we demonstrate state-of-the-art performance without making any assumptions regarding underlying longitudinal or time-to-event processes attesting to the model’s potential utility in clinical decision support.
Disease-Atlas: Navigating Disease Trajectories using Deep Learning
Bryan Lim, Mihaela van der Schaar
MLHC 2018
Abstract
Joint models for longitudinal and time-to-event data are commonly used in longitudinal studies to forecast disease trajectories over time. While there are many advantages to joint modeling, the standard forms suffer from limitations that arise from a fixed model specification, and computational difficulties when applied to high-dimensional datasets.
In this paper, we propose a deep learning approach to address these limitations, enhancing existing methods with the inherent flexibility and scalability of deep neural networks, while retaining the benefits of joint modeling.
Using longitudinal data from a real-world medical dataset, we demonstrate improvements in performance and scalability, as well as robustness in the presence of irregularly sampled data.
Forecasting Disease Trajectories in Alzheimer’s Disease Using Deep Learning
Bryan Lim, Mihaela van der Schaar
2018 KDD Workshop on Machine Learning for Medicine and Healthcare
Abstract
Joint models for longitudinal and time-to-event data are commonly used in longitudinal studies to forecast disease trajectories over time. Despite the many advantages of joint modeling, the standard forms suffer from limitations that arise from a fixed model specification and computational difficulties when applied to large datasets.
We adopt a deep learning approach to address these limitations, enhancing existing methods with the flexibility and scalability of deep neural networks while retaining the benefits of joint modeling.
Using data from the Alzheimer’s Disease Neuroimaging Institute, we show improvements in performance and scalability compared to traditional methods.
Again, the above projects would not have been possible without the generous support of Alzheimer’s Research UK and their sponsorship of Dan Jarrett’s Ph.D. studentship.
If you are a clinician and would like to learn more about how machine learning can be applied to real-world healthcare problems, please sign up for our Revolutionizing Healthcare online engagement sessions (no machine learning knowledge required).