UNITE: Uncertainty-based Health Risk Prediction Leveraging Multi-sourced Data

Successful health risk prediction demands accuracy and reliability of the model. Existing predictive models mainly depend on mining electronic health records (EHR) with advanced deep learning techniques to improve model accuracy. However, they all ignore the importance of publicly available online health data, especially socioeconomic status, environmental factors, and detailed demographic information for each location, which are all strong predictive signals and can definitely augment precision medicine. To achieve model reliability, the model needs to provide accurate prediction and uncertainty score of the prediction. However, existing uncertainty estimation approaches often failed in handling high-dimensional data, which are present in multi-sourced data. To fill the gap, we propose UNcertaInTy-based hEalth risk prediction (UNITE) model. Building upon an adaptive multimodal deep kernel and a stochastic variational inference module, UNITE provides accurate disease risk prediction and uncertainty estimation leveraging multi-sourced health data including EHR data, patient demographics, and public health data collected from the web. We evaluate UNITE on real-world disease risk prediction tasks: nonalcoholic fatty liver disease (NASH) and Alzheimer's disease (AD). UNITE achieves up to 0.841 in F1 score for AD detection, up to 0.609 in PR-AUC for NASH detection, and outperforms various state-of-the-art baselines by up to $19\%$ over the best baseline. We also show UNITE can model meaningful uncertainties and can provide evidence-based clinical support by clustering similar patients.

SWIFT: Scalable Wasserstein Factorization for Sparse Nonnegative Tensors

Existing tensor factorization methods assume that the input tensor follows some specific distribution (i.e. Poisson, Bernoulli and Gaussian), and solve the factorization by minimizing some empirical loss functions defined based on the corresponding distribution. However, it suffers from several drawbacks: 1) In reality, the underlying distributions are complicated and unknown, making it infeasible to be approximated by a simple distribution. 2) The correlation across dimensions of the input tensor is not well utilized, leading to sub-optimal performance. Although heuristics were proposed to incorporate such correlation as side information under Gaussian distribution, they can not easily be generalized to other distributions. Thus, a more principled way of utilizing the correlation in tensor factorization models is still an open challenge. Without assuming any explicit distribution, we formulate the tensor factorization as an optimal transport problem with Wasserstein distance, which can handle non-negative inputs. We introduce SWIFT, which minimizes the Wasserstein distance that measures the distance between the input tensor and that of the reconstruction. In particular, we define the N-th order tensor Wasserstein loss for the widely used tensor CP factorization, and derive the optimization algorithm that minimizes it. By leveraging sparsity structure and different equivalent formulations for optimizing computational efficiency, SWIFT is as scalable as other well-known CP algorithms. Using the factor matrices as features, SWIFT achieves up to 9.65% and 11.31% relative improvement over baselines for downstream prediction tasks. Under the noisy conditions, SWIFT achieves up to 15% and 17% relative improvements over the best competitors for the prediction tasks.

MolDesigner: Interactive Design of Efficacious Drugs with Deep Learning

The efficacy of a drug depends on its binding affinity to the therapeutic target and pharmacokinetics. Deep learning (DL) has demonstrated remarkable progress in predicting drug efficacy. We develop MolDesigner, a human-in-the-loop web user-interface (UI), to assist drug developers leverage DL predictions to design more effective drugs. A developer can draw a drug molecule in the interface. In the backend, more than 17 state-of-the-art DL models generate predictions on important indices that are crucial for a drug's efficacy. Based on these predictions, drug developers can edit the drug molecule and reiterate until satisfaction. MolDesigner can make predictions in real-time with a latency of less than a second.

MIMOSA: Multi-constraint Molecule Sampling for Molecule Optimization

Molecule optimization is a fundamental task for accelerating drug discovery, with the goal of generating new valid molecules that maximize multiple drug properties while maintaining similarity to the input molecule. Existing generative models and reinforcement learning approaches made initial success, but still face difficulties in simultaneously optimizing multiple drug properties. To address such challenges, we propose the MultI-constraint MOlecule SAmpling (MIMOSA) approach, a sampling framework to use input molecule as an initial guess and sample molecules from the target distribution. MIMOSA first pretrains two property agnostic graph neural networks (GNNs) for molecule topology and substructure-type prediction, where a substructure can be either atom or single ring. For each iteration, MIMOSA uses the GNNs' prediction and employs three basic substructure operations (add, replace, delete) to generate new molecules and associated weights. The weights can encode multiple constraints including similarity and drug property constraints, upon which we select promising molecules for next iteration. MIMOSA enables flexible encoding of multiple property- and similarity-constraints and can efficiently generate new molecules that satisfy various property constraints and achieved up to 49.6% relative improvement over the best baseline in terms of success rate.

SumGNN: Multi-typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization

Thanks to the increasing availability of drug-drug interactions (DDI) datasets and large biomedical knowledge graphs (KGs), accurate detection of adverse DDI using machine learning models becomes possible. However, it remains largely an open problem how to effectively utilize large and noisy biomedical KG for DDI detection. Due to its sheer size and amount of noise in KGs, it is often less beneficial to directly integrate KGs with other smaller but higher quality data (e.g., experimental data). Most of existing approaches ignore KGs altogether. Some tries to directly integrate KGs with other data via graph neural networks with limited success. Furthermore most previous works focus on binary DDI prediction whereas the multi-typed DDI pharmacological effect prediction is more meaningful but harder task. To fill the gaps, we propose a new method SumGNN:~{\it knowledge summarization graph neural network}, which is enabled by a subgraph extraction module that can efficiently anchor on relevant subgraphs from a KG, a self-attention based subgraph summarization scheme to generate reasoning path within the subgraph, and a multi-channel knowledge and data integration module that utilizes massive external biomedical knowledge for significantly improved multi-typed DDI predictions. SumGNN outperforms the best baseline by up to 5.54\%, and performance gain is particularly significant in low data relation types. In addition, SumGNN provides interpretable prediction via the generated reasoning paths for each prediction.

Pay Attention to Evolution: Time Series Forecasting with Deep Graph-Evolution Learning

Time-series forecasting is one of the most active research topics in predictive analysis. A still open gap in that literature is that statistical and ensemble learning approaches systematically present lower predictive performance than deep learning methods as they generally disregard the data sequence aspect entangled with multivariate data represented in more than one time series. Conversely, this work presents a novel neural network architecture for time-series forecasting that combines the power of graph evolution with deep recurrent learning on distinct data distributions; we named our method Recurrent Graph Evolution Neural Network (ReGENN). The idea is to infer multiple multivariate relationships between co-occurring time-series by assuming that the temporal data depends not only on inner variables and intra-temporal relationships (i.e., observations from itself) but also on outer variables and inter-temporal relationships (i.e., observations from other-selves). An extensive set of experiments was conducted comparing ReGENN with dozens of ensemble methods and classical statistical ones, showing sound improvement of up to 64.87% over the competing algorithms. Furthermore, we present an analysis of the intermediate weights arising from ReGENN, showing that by looking at inter and intra-temporal relationships simultaneously, time-series forecasting is majorly improved if paying attention to how multiple multivariate data synchronously evolve.

SDE-Net: Equipping Deep Neural Networks with Uncertainty Estimates

Uncertainty quantification is a fundamental yet unsolved problem for deep learning. The Bayesian framework provides a principled way of uncertainty estimation but is often not scalable to modern deep neural nets (DNNs) that have a large number of parameters. Non-Bayesian methods are simple to implement but often conflate different sources of uncertainties and require huge computing resources. We propose a new method for quantifying uncertainties of DNNs from a dynamical system perspective. The core of our method is to view DNN transformations as state evolution of a stochastic dynamical system and introduce a Brownian motion term for capturing epistemic uncertainty. Based on this perspective, we propose a neural stochastic differential equation model (SDE-Net) which consists of (1) a drift net that controls the system to fit the predictive function; and (2) a diffusion net that captures epistemic uncertainty. We theoretically analyze the existence and uniqueness of the solution to SDE-Net. Our experiments demonstrate that the SDE-Net model can outperform existing uncertainty estimation methods across a series of tasks where uncertainty plays a fundamental role.

HOLMES: Health OnLine Model Ensemble Serving for Deep Learning Models in Intensive Care Units

Deep learning models have achieved expert-level performance in healthcare with an exclusive focus on training accurate models. However, in many clinical environments such as intensive care unit (ICU), real-time model serving is equally if not more important than accuracy, because in ICU patient care is simultaneously more urgent and more expensive. Clinical decisions and their timeliness, therefore, directly affect both the patient outcome and the cost of care. To make timely decisions, we argue the underlying serving system must be latency-aware. To compound the challenge, health analytic applications often require a combination of models instead of a single model, to better specialize individual models for different targets, multi-modal data, different prediction windows, and potentially personalized predictions. To address these challenges, we propose HOLMES-an online model ensemble serving framework for healthcare applications. HOLMES dynamically identifies the best performing set of models to ensemble for highest accuracy, while also satisfying sub-second latency constraints on end-to-end prediction. We demonstrate that HOLMES is able to navigate the accuracy/latency tradeoff efficiently, compose the ensemble, and serve the model ensemble pipeline, scaling to simultaneously streaming data from 100 patients, each producing waveform data at 250~Hz. HOLMES outperforms the conventional offline batch-processed inference for the same clinical task in terms of accuracy and latency (by order of magnitude). HOLMES is tested on risk prediction task on pediatric cardio ICU data with above 95% prediction accuracy and sub-second latency on 64-bed simulation.

EMIXER: End-to-end Multimodal X-ray Generation via Self-supervision

Deep generative models have enabled the automated synthesis of high-quality data for diverse applications. However, the most effective generative models are specialized to data from a single domain (e.g., images or text). Real-world applications such as healthcare require multi-modal data from multiple domains (e.g., both images and corresponding text), which are difficult to acquire due to limited availability and privacy concerns and are much harder to synthesize. To tackle this joint synthesis challenge, we propose an End-to-end MultImodal X-ray genERative model (EMIXER) for jointly synthesizing x-ray images and corresponding free-text reports, all conditional on diagnosis labels. EMIXER is an conditional generative adversarial model by 1) generating an image based on a label, 2) encoding the image to a hidden embedding, 3) producing the corresponding text via a hierarchical decoder from the image embedding, and 4) a joint discriminator for assessing both the image and the corresponding text. EMIXER also enables self-supervision to leverage vast amount of unlabeled data. Extensive experiments with real X-ray reports data illustrate how data augmentation using synthesized multimodal samples can improve the performance of a variety of supervised tasks including COVID-19 X-ray classification with very limited samples. The quality of generated images and reports are also confirmed by radiologists. We quantitatively show that EMIXER generated synthetic datasets can augment X-ray image classification, report generation models to achieve 5.94% and 6.9% improvement on models trained only on real data samples. Taken together, our results highlight the promise of state of generative models to advance clinical machine learning.

ELF: An Early-Exiting Framework for Long-Tailed Classification

The natural world often follows a long-tailed data distribution where only a few classes account for most of the examples. This long-tail causes classifiers to overfit to the majority class. To mitigate this, prior solutions commonly adopt class rebalancing strategies such as data resampling and loss reshaping. However, by treating each example within a class equally, these methods fail to account for the important notion of example hardness, i.e., within each class some examples are easier to classify than others. To incorporate this notion of hardness into the learning process, we propose the EarLy-exiting Framework(ELF). During training, ELF learns to early-exit easy examples through auxiliary branches attached to a backbone network. This offers a dual benefit-(1) the neural network increasingly focuses on hard examples, since they contribute more to the overall network loss; and (2) it frees up additional model capacity to distinguish difficult examples. Experimental results on two large-scale datasets, ImageNet LT and iNaturalist'18, demonstrate that ELF can improve state-of-the-art accuracy by more than 3 percent. This comes with the additional benefit of reducing up to 20 percent of inference time FLOPS. ELF is complementary to prior work and can naturally integrate with a variety of existing methods to tackle the challenge of long-tailed distributions.