Spectators sit stacked one row behind another, all craning their necks to look down at the miniaturized soccer pitch below them. With bated breath, they watch as a tiny player gently bumps the ball up the pitch toward its opponent. The goal in sight, the attacker has two options -- evade its opponent by expertly moving the ball around it, or send a safer pass to a teammate outside the fray. Like any soccer star, the player chooses glory and begins to putter its feet back and forth to move the ball -- when it begins to lose balance. And down it falls, like a felled tree.

The recent release of large-scale healthcare datasets has greatly propelled the research of data-driven deep learning models for healthcare applications. However, due to the nature of such deep black-boxed models, concerns about interpretability, fairness, and biases in healthcare scenarios where human lives are at stake call for a careful and thorough examinations of both datasets and models. In this work, we focus on MIMIC-IV (Medical Information Mart for Intensive Care, version IV), the largest publicly available healthcare dataset, and conduct comprehensive analyses of dataset representation bias as well as interpretability and prediction fairness of deep learning models for in-hospital mortality prediction. In terms of interpretabilty, we observe that (1) the best performing interpretability method successfully identifies critical features for mortality prediction on various prediction models; (2) demographic features are important for prediction. In terms of fairness, we observe that (1) there exists disparate treatment in prescribing mechanical ventilation among patient groups across ethnicity, gender and age; (2) all of the studied mortality predictors are generally fair while the IMV-LSTM (Interpretable Multi-Variable Long Short-Term Memory) model provides the most accurate and unbiased predictions across all protected groups. We further draw concrete connections between interpretability methods and fairness metrics by showing how feature importance from interpretability methods can be beneficial in quantifying potential disparities in mortality predictors.

The lack of interpretability remains a key barrier to the adoption of deep models in many applications. In this work, we explicitly regularize deep models so human users might step through the process behind their predictions in little time. Specifically, we train deep time-series models so their class-probability predictions have high accuracy while being closely modeled by decision trees with few nodes. Using intuitive toy examples as well as medical tasks for treating sepsis and HIV, we demonstrate that this new tree regularization yields models that are easier for humans to simulate than simpler L1 or L2 penalties without sacrificing predictive power.

The lack of interpretability remains a key barrier to the adoption of deep models in many applications. In this work, we explicitly regularize deep models so human users might step through the process behind their predictions in little time. Specifically, we train deep time-series models so their class-probability predictions have high accuracy while being closely modeled by decision trees with few nodes. Using intuitive toy examples as well as medical tasks for treating sepsis and HIV, we demonstrate that this new tree regularization yields models that are easier for humans to simulate than simpler L1 or L2 penalties without sacrificing predictive power.