Early Prediction of Causes (not Effects) in Healthcare by Long-Term Clinical Time Series Forecasting

TL;DR

This paper introduces a method for early prediction of causes in healthcare by forecasting clinical time series data with Transformer models, enabling interpretable and flexible diagnosis based on predicted causes rather than effects.

Contribution

It proposes a novel approach that forecasts causes directly through long-term time series prediction, improving interpretability and flexibility over traditional effect-based predictions.

Findings

Best results achieved with standard dense encoders and iterative multi-step decoders.

Iterative decoding captures cross-variable dependencies effectively.

The method outperforms recent set function encoder approaches.

Abstract

Machine learning for early syndrome diagnosis aims to solve the intricate task of predicting a ground truth label that most often is the outcome (effect) of a medical consensus definition applied to observed clinical measurements (causes), given clinical measurements observed several hours before. Instead of focusing on the prediction of the future effect, we propose to directly predict the causes via time series forecasting (TSF) of clinical variables and determine the effect by applying the gold standard consensus definition to the forecasted values. This method has the invaluable advantage of being straightforwardly interpretable to clinical practitioners, and because model training does not rely on a particular label anymore, the forecasted data can be used to predict any consensus-based label. We exemplify our method by means of long-term TSF with Transformer models, with a focus on accurate prediction of sparse clinical variables involved in the SOFA-based Sepsis-3 definition and the new Simplified Acute Physiology Score (SAPS-II) definition. Our experiments are conducted on two datasets and show that contrary to recent proposals which advocate set function encoders for time series and direct multi-step decoders, best results are achieved by a combination of standard dense encoders with iterative multi-step decoders. The key for success of iterative multi-step decoding can be attributed to its ability to capture cross-variate dependencies and to a student forcing training strategy that teaches the model to rely on its own previous time step predictions for the next time step prediction.