Machine learning (ML) approaches have demonstrated promising results in a wide range of healthcare applications. Data plays a crucial role in developing ML-based healthcare systems that directly affect people's lives. Many of the ethical issues surrounding the use of ML in healthcare stem from structural inequalities underlying the way we collect, use, and handle data. Developing guidelines to improve documentation practices regarding the creation, use, and maintenance of ML healthcare datasets is therefore of critical importance. In this work, we introduce Healthsheet, a contextualized adaptation of the original datasheet questionnaire ~\cite{gebru2018datasheets} for health-specific applications. Through a series of semi-structured interviews, we adapt the datasheets for healthcare data documentation. As part of the Healthsheet development process and to understand the obstacles researchers face in creating datasheets, we worked with three publicly-available healthcare datasets as our case studies, each with different types of structured data: Electronic health Records (EHR), clinical trial study data, and smartphone-based performance outcome measures. Our findings from the interviewee study and case studies show 1) that datasheets should be contextualized for healthcare, 2) that despite incentives to adopt accountability practices such as datasheets, there is a lack of consistency in the broader use of these practices 3) how the ML for health community views datasheets and particularly \textit{Healthsheets} as diagnostic tool to surface the limitations and strength of datasets and 4) the relative importance of different fields in the datasheet to healthcare concerns.
Fairness and robustness are often considered as orthogonal dimensions when evaluating machine learning models. However, recent work has revealed interactions between fairness and robustness, showing that fairness properties are not necessarily maintained under distribution shift. In healthcare settings, this can result in e.g. a model that performs fairly according to a selected metric in "hospital A" showing unfairness when deployed in "hospital B". While a nascent field has emerged to develop provable fair and robust models, it typically relies on strong assumptions about the shift, limiting its impact for real-world applications. In this work, we explore the settings in which recently proposed mitigation strategies are applicable by referring to a causal framing. Using examples of predictive models in dermatology and electronic health records, we show that real-world applications are complex and often invalidate the assumptions of such methods. Our work hence highlights technical, practical, and engineering gaps that prevent the development of robustly fair machine learning models for real-world applications. Finally, we discuss potential remedies at each step of the machine learning pipeline.
Survival analysis is a challenging variation of regression modeling because of the presence of censoring, where the outcome measurement is only partially known, due to, for example, loss to follow up. Such problems come up frequently in medical applications, making survival analysis a key endeavor in biostatistics and machine learning for healthcare, with Cox regression models being amongst the most commonly employed models. We describe a new approach for survival analysis regression models, based on learning mixtures of Cox regressions to model individual survival distributions. We propose an approximation to the Expectation Maximization algorithm for this model that does hard assignments to mixture groups to make optimization efficient. In each group assignment, we fit the hazard ratios within each group using deep neural networks, and the baseline hazard for each mixture component non-parametrically. We perform experiments on multiple real world datasets, and look at the mortality rates of patients across ethnicity and gender. We emphasize the importance of calibration in healthcare settings and demonstrate that our approach outperforms classical and modern survival analysis baselines, both in terms of discriminative performance and calibration, with large gains in performance on the minority demographics.
ML models often exhibit unexpectedly poor behavior when they are deployed in real-world domains. We identify underspecification as a key reason for these failures. An ML pipeline is underspecified when it can return many predictors with equivalently strong held-out performance in the training domain. Underspecification is common in modern ML pipelines, such as those based on deep learning. Predictors returned by underspecified pipelines are often treated as equivalent based on their training domain performance, but we show here that such predictors can behave very differently in deployment domains. This ambiguity can lead to instability and poor model behavior in practice, and is a distinct failure mode from previously identified issues arising from structural mismatch between training and deployment domains. We show that this problem appears in a wide variety of practical ML pipelines, using examples from computer vision, medical imaging, natural language processing, clinical risk prediction based on electronic health records, and medical genomics. Our results show the need to explicitly account for underspecification in modeling pipelines that are intended for real-world deployment in any domain.
Bayesian neural networks (BNNs) demonstrate promising success in improving the robustness and uncertainty quantification of modern deep learning. However, they generally struggle with underfitting at scale and parameter efficiency. On the other hand, deep ensembles have emerged as alternatives for uncertainty quantification that, while outperforming BNNs on certain problems, also suffer from efficiency issues. It remains unclear how to combine the strengths of these two approaches and remediate their common issues. To tackle this challenge, we propose a rank-1 parameterization of BNNs, where each weight matrix involves only a distribution on a rank-1 subspace. We also revisit the use of mixture approximate posteriors to capture multiple modes, where unlike typical mixtures, this approach admits a significantly smaller memory increase (e.g., only a 0.4% increase for a ResNet-50 mixture of size 10). We perform a systematic empirical study on the choices of prior, variational posterior, and methods to improve training. For ResNet-50 on ImageNet, Wide ResNet 28-10 on CIFAR-10/100, and an RNN on MIMIC-III, rank-1 BNNs achieve state-of-the-art performance across log-likelihood, accuracy, and calibration on the test sets and out-of-distribution variants.
The use of collaborative and decentralized machine learning techniques such as federated learning have the potential to enable the development and deployment of clinical risk predictions models in low-resource settings without requiring sensitive data be shared or stored in a central repository. This process necessitates communication of model weights or updates between collaborating entities, but it is unclear to what extent patient privacy is compromised as a result. To gain insight into this question, we study the efficacy of centralized versus federated learning in both private and non-private settings. The clinical prediction tasks we consider are the prediction of prolonged length of stay and in-hospital mortality across thirty one hospitals in the eICU Collaborative Research Database. We find that while it is straightforward to apply differentially private stochastic gradient descent to achieve strong privacy bounds when training in a centralized setting, it is considerably more difficult to do so in the federated setting.
In medicine, both ethical and monetary costs of incorrect predictions can be significant, and the complexity of the problems often necessitates increasingly complex models. Recent work has shown that changing just the random seed is enough for otherwise well-tuned deep neural networks to vary in their individual predicted probabilities. In light of this, we investigate the role of model uncertainty methods in the medical domain. Using RNN ensembles and various Bayesian RNNs, we show that population-level metrics, such as AUC-PR, AUC-ROC, log-likelihood, and calibration error, do not capture model uncertainty. Meanwhile, the presence of significant variability in patient-specific predictions and optimal decisions motivates the need for capturing model uncertainty. Understanding the uncertainty for individual patients is an area with clear clinical impact, such as determining when a model decision is likely to be brittle. We further show that RNNs with only Bayesian embeddings can be a more efficient way to capture model uncertainty compared to ensembles, and we analyze how model uncertainty is impacted across individual input features and patient subgroups.
Learning-to-learn or meta-learning leverages data-driven inductive bias to increase the efficiency of learning on a novel task. This approach encounters difficulty when transfer is not mutually beneficial, for instance, when tasks are sufficiently dissimilar or change over time. Here, we use the connection between gradient-based meta-learning and hierarchical Bayes (Grant et al., 2018) to propose a mixture of hierarchical Bayesian models over the parameters of an arbitrary function approximator such as a neural network. Generalizing the model-agnostic meta-learning (MAML) algorithm (Finn et al., 2017), we present a stochastic expectation maximization procedure to jointly estimate parameter initializations for gradient descent as well as a latent assignment of tasks to initializations. This approach better captures the diversity of training tasks as opposed to consolidating inductive biases into a single set of hyperparameters. Our experiments demonstrate better generalization performance on the standard miniImageNet benchmark for 1-shot classification. We further derive a novel and scalable non-parametric variant of our method that captures the evolution of a task distribution over time as demonstrated on a set of few-shot regression tasks.
We propose a new method that uses deep learning techniques to solve the inverse problems. The inverse problem is cast in the form of learning an end-to-end mapping from observed data to the ground-truth. Inspired by the splitting strategy widely used in regularized iterative algorithm to tackle inverse problems, the mapping is decomposed into two networks, with one handling the inversion of the physical forward model associated with the data term and one handling the denoising of the output from the former network, i.e., the inverted version, associated with the prior/regularization term. The two networks are trained jointly to learn the end-to-end mapping, getting rid of a two-step training. The training is annealing as the intermediate variable between these two networks bridges the gap between the input (the degraded version of output) and output and progressively approaches to the ground-truth. The proposed network, referred to as InverseNet, is flexible in the sense that most of the existing end-to-end network structure can be leveraged in the first network and most of the existing denoising network structure can be used in the second one. Extensive experiments on both synthetic data and real datasets on the tasks, motion deblurring, super-resolution, and colorization, demonstrate the efficiency and accuracy of the proposed method compared with other image processing algorithms.
Sepsis is a poorly understood and potentially life-threatening complication that can occur as a result of infection. Early detection and treatment improves patient outcomes, and as such it poses an important challenge in medicine. In this work, we develop a flexible classifier that leverages streaming lab results, vitals, and medications to predict sepsis before it occurs. We model patient clinical time series with multi-output Gaussian processes, maintaining uncertainty about the physiological state of a patient while also imputing missing values. The mean function takes into account the effects of medications administered on the trajectories of the physiological variables. Latent function values from the Gaussian process are then fed into a deep recurrent neural network to classify patient encounters as septic or not, and the overall model is trained end-to-end using back-propagation. We train and validate our model on a large dataset of 18 months of heterogeneous inpatient stays from the Duke University Health System, and develop a new "real-time" validation scheme for simulating the performance of our model as it will actually be used. Our proposed method substantially outperforms clinical baselines, and improves on a previous related model for detecting sepsis. Our model's predictions will be displayed in a real-time analytics dashboard to be used by a sepsis rapid response team to help detect and improve treatment of sepsis.