WindowSHAP: An efficient framework for explaining time-series classifiers based on Shapley values.

Unpacking and comprehending how black-box machine learning algorithms (such as deep learning models) make decisions has been a persistent challenge for researchers and end-users. Explaining time-series predictive models is useful for clinical applications with high stakes to understand the behavior of prediction models, e.g., to determine how different variables and time points influence the clinical outcome. However, existing approaches to explain such models are frequently unique to architectures and data where the features do not have a time-varying component. In this paper, we introduce WindowSHAP, a model-agnostic framework for explaining time-series classifiers using Shapley values. We intend for WindowSHAP to mitigate the computational complexity of calculating Shapley values for long time-series data as well as improve the quality of explanations. WindowSHAP is based on partitioning a sequence into time windows. Under this framework, we present three distinct algorithms of Stationary, Sliding and Dynamic WindowSHAP, each evaluated against baseline approaches, KernelSHAP and TimeSHAP, using perturbation and sequence analyses metrics. We applied our framework to clinical time-series data from both a specialized clinical domain (Traumatic Brain Injury - TBI) as well as a broad clinical domain (critical care medicine). The experimental results demonstrate that, based on the two quantitative metrics, our framework is superior at explaining clinical time-series classifiers, while also reducing the complexity of computations. We show that for time-series data with 120 time steps (hours), merging 10 adjacent time points can reduce the CPU time of WindowSHAP by 80 % compared to KernelSHAP. We also show that our Dynamic WindowSHAP algorithm focuses more on the most important time steps and provides more understandable explanations. As a result, WindowSHAP not only accelerates the calculation of Shapley values for time-series data, but also delivers more understandable explanations with higher quality.

A self-supervised learning-based approach to clustering multivariate time-series data with missing values (SLAC-Time): An application to TBI phenotyping.

Self-supervised learning approaches provide a promising direction for clustering multivariate time-series data. However, real-world time-series data often include missing values, and the existing approaches require imputing missing values before clustering, which may cause extensive computations and noise and result in invalid interpretations. To address these challenges, we present a Self-supervised Learning-based Approach to Clustering multivariate Time-series data with missing values (SLAC-Time). SLAC-Time is a Transformer-based clustering method that uses time-series forecasting as a proxy task for leveraging unlabeled data and learning more robust time-series representations. This method jointly learns the neural network parameters and the cluster assignments of the learned representations. It iteratively clusters the learned representations with the K-means method and then utilizes the subsequent cluster assignments as pseudo-labels to update the model parameters. To evaluate our proposed approach, we applied it to clustering and phenotyping Traumatic Brain Injury (TBI) patients in the Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) study. Clinical data associated with TBI patients are often measured over time and represented as time-series variables characterized by missing values and irregular time intervals. Our experiments demonstrate that SLAC-Time outperforms the baseline K-means clustering algorithm in terms of silhouette coefficient, Calinski Harabasz index, Dunn index, and Davies Bouldin index. We identified three TBI phenotypes that are distinct from one another in terms of clinically significant variables as well as clinical outcomes, including the Extended Glasgow Outcome Scale (GOSE) score, Intensive Care Unit (ICU) length of stay, and mortality rate. The experiments show that the TBI phenotypes identified by SLAC-Time can be potentially used for developing targeted clinical trials and therapeutic strategies.

Predicting Failure of Noninvasive Respiratory Support Using Deep Recurrent Learning.

BACKGROUND: Noninvasive respiratory support (NRS) is increasingly used to support patients with acute respiratory failure. However, noninvasive support failure may worsen outcomes compared to primary support with invasive mechanical ventilation. Therefore, there is a need to identify patients where NRS is failing so that treatment can be reassessed and adjusted. The objective of this study was to develop and evaluate 3 recurrent neural network (RNN) models to predict NRS failure. METHODS: This was a cross-sectional observational study to evaluate the ability of deep RNN models (long short-term memory [LSTM], gated recurrent unit [GRU]), and GRU with trainable decay) to predict failure of NRS. Data were extracted from electronic health records from all adult (≥ 18 y) patient records requiring any type of oxygen therapy or mechanical ventilation between November 1, 2013-September 30, 2020, across 46 ICUs in the Southwest United States in a single health care network. Input variables for each model included serum chloride, creatinine, albumin, breathing frequency, heart rate, SpO2 , FIO2 , arterial oxygen saturation (SaO2 ), and 2 measurements each (point-of-care and laboratory measurement) of PaO2 and partial pressure of arterial oxygen from an arterial blood gas. RESULTS: Time series data from electronic health records were available for 22,075 subjects. The highest accuracy and area under the receiver operating characteristic curve were for the LSTM model (94.04% and 0.9636, respectively). Accurate predictions were made 12 h after ICU admission, and performance remained high well in advance of NRS failure. CONCLUSIONS: RNN models using routinely collected time series data can accurately predict NRS failure well before intubation. This lead time may provide an opportunity to intervene to optimize patient outcomes.

Identifying TBI Physiological States by Clustering Multivariate Clinical Time-Series Data.

Determining clinically relevant physiological states from multivariate time-series data with missing values is essential for providing appropriate treatment for acute conditions such as Traumatic Brain Injury (TBI), respiratory failure, and heart failure. Utilizing non-temporal clustering or data imputation and aggregation techniques may lead to loss of valuable information and biased analyses. In our study, we apply the SLAC-Time algorithm, an innovative self-supervision-based approach that maintains data integrity by avoiding imputation or aggregation, offering a more useful representation of acute patient states. By using SLAC-Time to cluster data in a large research dataset, we identified three distinct TBI physiological states and their specific feature profiles. We employed various clustering evaluation metrics and incorporated input from a clinical domain expert to validate and interpret the identified physiological states. Further, we discovered how specific clinical events and interventions can influence patient states and state transitions.

An Empirical Comparison of Explainable Artificial Intelligence Methods for Clinical Data: A Case Study on Traumatic Brain Injury.

A longstanding challenge surrounding deep learning algorithms is unpacking and understanding how they make their decisions. Explainable Artificial Intelligence (XAI) offers methods to provide explanations of internal functions of algorithms and reasons behind their decisions in ways that are interpretable and understandable to human users. . Numerous XAI approaches have been developed thus far, and a comparative analysis of these strategies seems necessary to discern their relevance to clinical prediction models. To this end, we first implemented two prediction models for short- and long-term outcomes of traumatic brain injury (TBI) utilizing structured tabular as well as time-series physiologic data, respectively. Six different interpretation techniques were used to describe both prediction models at the local and global levels. We then performed a critical analysis of merits and drawbacks of each strategy, highlighting the implications for researchers who are interested in applying these methodologies. The implemented methods were compared to one another in terms of several XAI characteristics such as understandability, fidelity, and stability. Our findings show that SHAP is the most stable with the highest fidelity but falls short of understandability. Anchors, on the other hand, is the most understandable approach, but it is only applicable to tabular data and not time series data.

Recurrent Neural Network based Time-Series Modeling for Long-term Prognosis Following Acute Traumatic Brain Injury.

We developed a prognostic model for longer-term outcome prediction in traumatic brain injury (TBI) using an attention-based recurrent neural network (RNN). The model was trained on admission and time series data obtained from a multi-site, longitudinal, observational study of TBI patients. We included 110 clinical variables as model input and Glasgow Outcome Score Extended (GOSE) at six months after injury as the outcome variable. Designed to handle missing values in time series data, the RNN model was compared to an existing TBI prognostic model using 10-fold cross validation. The area under receiver operating characteristic curve (AUC) for the RNN model is 0.86 (95% CI 0.83-0.89) for binary outcomes, whereas the AUC of the comparison model is 0.69 (95% CI 0.67-0.71). We demonstrated that including time series data into prognostic models for TBI can boost the discriminative ability of prediction models with either binary or ordinal outcomes.