Extracorporeal membrane oxygenation (ECMO) is an essential life-supporting modality for COVID-19 patients who are refractory to conventional therapies. However, the proper treatment decision has been the subject of significant debate and it remains controversial about who benefits from this scarcely available and technically complex treatment option. To support clinical decisions, it is a critical need to predict the treatment need and the potential treatment and no-treatment responses. Targeting this clinical challenge, we propose Treatment Variational AutoEncoder (TVAE), a novel approach for individualized treatment analysis. TVAE is specifically designed to address the modeling challenges like ECMO with strong treatment selection bias and scarce treatment cases. TVAE conceptualizes the treatment decision as a multi-scale problem. We model a patient's potential treatment assignment and the factual and counterfactual outcomes as part of their intrinsic characteristics that can be represented by a deep latent variable model. The factual and counterfactual prediction errors are alleviated via a reconstruction regularization scheme together with semi-supervision, and the selection bias and the scarcity of treatment cases are mitigated by the disentangled and distribution-matched latent space and the label-balancing generative strategy. We evaluate TVAE on two real-world COVID-19 datasets: an international dataset collected from 1651 hospitals across 63 countries, and a institutional dataset collected from 15 hospitals. The results show that TVAE outperforms state-of-the-art treatment effect models in predicting both the propensity scores and factual outcomes on heterogeneous COVID-19 datasets. Additional experiments also show TVAE outperforms the best existing models in individual treatment effect estimation on the synthesized IHDP benchmark dataset.

We present an end-to-end model using streaming physiological time series to predict near-term risk for hypoxemia, a rare, but life-threatening condition known to cause serious patient harm during surgery. Inspired by the fact that a hypoxemia event is defined based on a future sequence of low SpO2 (i.e., blood oxygen saturation) instances, we propose the hybrid inference network (hiNet) that makes hybrid inference on both future low SpO2 instances and hypoxemia outcomes. hiNet integrates 1) a joint sequence autoencoder that simultaneously optimizes a discriminative decoder for label prediction, and 2) two auxiliary decoders trained for data reconstruction and forecast, which seamlessly learn contextual latent representations that capture the transition from present states to future states. All decoders share a memory-based encoder that helps capture the global dynamics of patient measurement. For a large surgical cohort of 72,081 surgeries at a major academic medical center, our model outperforms strong baselines including the model used by the state-of-the-art hypoxemia prediction system. With its capability to make real-time predictions of near-term hypoxemic at clinically acceptable alarm rates, hiNet shows promise in improving clinical decision making and easing burden of perioperative care.

Major postoperative complications are devastating to surgical patients. Some of these complications are potentially preventable via early predictions based on intraoperative data. However, intraoperative data comprise long and fine-grained multivariate time series, prohibiting the effective learning of accurate models. The large gaps associated with clinical events and protocols are usually ignored. Moreover, deep models generally lack transparency. Nevertheless, the interpretability is crucial to assist clinicians in planning for and delivering postoperative care and timely interventions. Towards this end, we propose a hierarchical model combining the strength of both attention and recurrent models for intraoperative time series. We further develop an explanation module for the hierarchical model to interpret the predictions by providing contributions of intraoperative data in a fine-grained manner. Experiments on a large dataset of 111,888 surgeries with multiple outcomes and an external high-resolution ICU dataset show that our model can achieve strong predictive performance (i.e., high accuracy) and offer robust interpretations (i.e., high transparency) for predicted outcomes based on intraoperative time series.

Burnout is a significant public health concern affecting nearly half of the healthcare workforce. This paper presents the first end-to-end deep learning framework for predicting physician burnout based on electronic health record (EHR) activity logs, digital traces of physician work activities that are available in any EHR system. In contrast to prior approaches that exclusively relied on surveys for burnout measurement, our framework directly learns deep representations of physician behaviors from large-scale clinician activity logs to predict burnout. We propose the Hierarchical burnout Prediction based on Activity Logs (HiPAL), featuring a pre-trained time-dependent activity embedding mechanism tailored for activity logs and a hierarchical predictive model, which mirrors the natural hierarchical structure of clinician activity logs and captures physicians' evolving burnout risk at both short-term and long-term levels. To utilize the large amount of unlabeled activity logs, we propose a semi-supervised framework that learns to transfer knowledge extracted from unlabeled clinician activities to the HiPAL-based prediction model. The experiment on over 15 million clinician activity logs collected from the EHR at a large academic medical center demonstrates the advantages of our proposed framework in predictive performance of physician burnout and training efficiency over state-of-the-art approaches.

Given the risks and cost of hospitalization, there has been significant interest in exploiting machine learning models to improve perioperative care. However, due to the high dimensionality and noisiness of perioperative data, it remains a challenge to develop accurate and robust encoding for surgical predictions. Furthermore, it is important for the encoding to be interpretable by perioperative care practitioners to facilitate their decision making process. We proposeclinical variational autoencoder (cVAE), a deep latent variable model that addresses the challenges of surgical applications through two salient features. (1) To overcome performance limitations of traditional VAE, it isprediction-guided with explicit expression of predicted outcome in the latent representation. (2) Itdisentangles the latent space so that it can be interpreted in a clinically meaningful fashion. We apply cVAE to two real-world perioperative datasets to evaluate its efficacy and performance in predicting outcomes that are important to perioperative care, including postoperative complication and surgery duration. To demonstrate the generality and facilitate reproducibility, we also apply cVAE to the open MIMIC-III dataset for predicting ICU duration and mortality. Our results show that the latent representation provided by cVAE leads to superior performance in classification, regression and multi-task predictions. The two features of cVAE are mutually beneficial and eliminate the need of a predictor. We further demonstrate the interpretability of the disentangled representation and its capability to capture intrinsic characteristics of hospitalized patients. While this work is motivated by and evaluated in the context of clinical applications, the proposed approach may be generalized for other fields using high-dimensional and noisy data and valuing interpretable representations.

Machine learning techniques have shown promise in predicting clinical deterioration of hospitalized patients based on electronic health record (EHR). However, building accurate early warning systems (EWS) remains challenging in practice. EHRs are heterogeneous, comprising both static and time-series data. Moreover, missing values are prevalent in both static and time-series data, and the missingness of certain data can be correlated to clinical outcomes. This paper proposes a novel approach for integrating static and time-series clinical data in deep recurrent models through multi-modal fusion. Furthermore, we exploit the correlation of static and time-series data through cross-modal imputation in an integrated recurrent model. We apply the proposed approaches to a dataset extracted from the EHR of 20,700 hospitalizations of adult oncology patients in a research hospital. The experiments demonstrate the proposed approaches outperform the state-of-the-art models in terms of predictive accuracy in generating early warnings for clinical deterioration. A case study further establishes the efficacy of the predictive model for early warning systems under realistic clinical settings.

Clinical study found that early detection and intervention are essential for preventing clinical deterioration in patients, for patients both in intensive care units (ICU) as well as in general wards but under real-time data sensing (RDS). In this paper, we develop an integrated data mining approach to give early deterioration warnings for patients under real-time monitoring in ICU and RDS.