Improving Sepsis Treatment Strategies by Combining Deep and Kernel-Based Reinforcement Learning

Sepsis is the leading cause of mortality in the ICU. It is challenging to manage because individual patients respond differently to treatment. Thus, tailoring treatment to the individual patient is essential for the best outcomes. In this paper, we take steps toward this goal by applying a mixture-of-experts framework to personalize sepsis treatment. The mixture model selectively alternates between neighbor-based (kernel) and deep reinforcement learning (DRL) experts depending on patient's current history. On a large retrospective cohort, this mixture-based approach outperforms physician, kernel only, and DRL-only experts.

Evaluating Reinforcement Learning Algorithms in Observational Health Settings

Much attention has been devoted recently to the development of machine learning algorithms with the goal of improving treatment policies in healthcare. Reinforcement learning (RL) is a sub-field within machine learning that is concerned with learning how to make sequences of decisions so as to optimize long-term effects. Already, RL algorithms have been proposed to identify decision-making strategies for mechanical ventilation, sepsis management and treatment of schizophrenia. However, before implementing treatment policies learned by black-box algorithms in high-stakes clinical decision problems, special care must be taken in the evaluation of these policies. In this document, our goal is to expose some of the subtleties associated with evaluating RL algorithms in healthcare. We aim to provide a conceptual starting point for clinical and computational researchers to ask the right questions when designing and evaluating algorithms for new ways of treating patients. In the following, we describe how choices about how to summarize a history, variance of statistical estimators, and confounders in more ad-hoc measures can result in unreliable, even misleading estimates of the quality of a treatment policy. We also provide suggestions for mitigating these effects---for while there is much promise for mining observational health data to uncover better treatment policies, evaluation must be performed thoughtfully.

Sleep-deprived Fatigue Pattern Analysis using Large-Scale Selfies from Social Med

The complexities of fatigue have drawn much attention from researchers across various disciplines. Short-term fatigue may cause safety issue while driving; thus, dynamic systems were designed to track driver fatigue. Long-term fatigue could lead to chronic syndromes, and eventually affect individuals physical and psychological health. Traditional methodologies of evaluating fatigue not only require sophisticated equipment but also consume enormous time. In this paper, we attempt to develop a novel and efficient method to predict individual's fatigue rate by scrutinizing human facial cues. Our goal is to predict fatigue rate based on a selfie. To associate the fatigue rate with user behaviors, we have collected nearly 1-million timeline posts from 10,480 users on Instagram. We first detect all the faces and identify their demographics using automatic algorithms. Next, we investigate the fatigue distribution by weekday over different age, gender, and ethnic groups. This work represents a promising way to assess sleep-deprived fatigue, and our study provides a viable and efficient computational framework for user fatigue modeling in large-scale via social media.