Democratizing EHR analyses with FIDDLE: a flexible data-driven preprocessing pipeline for structured clinical data

TLDR

Applying machine learning to electronic health record data requires many preprocessing decisions that are labor‑intensive, and as ML’s role in health care grows, systematic and reproducible preprocessing techniques are increasingly needed. We developed FIDDLE, an open‑source framework that streamlines preprocessing of structured EHR data, reducing the number of decisions a user must make. We applied FIDDLE to the MIMIC‑III and eICU ICU datasets, training models for in‑hospital mortality, acute respiratory failure, and shock, extracting 2,528–7,403 features, and evaluating performance via AUROC against several baselines. FIDDLE‑based models achieved AUROCs of 0.757–0.886, comparable to the MIMIC‑Extract pipeline, and demonstrated generalizability across prediction times, algorithms, and datasets, proving robust to user‑defined settings and accelerating standardized preprocessing for clinical ML.

Abstract

In applying machine learning (ML) to electronic health record (EHR) data, many decisions must be made before any ML is applied; such preprocessing requires substantial effort and can be labor-intensive. As the role of ML in health care grows, there is an increasing need for systematic and reproducible preprocessing techniques for EHR data. Thus, we developed FIDDLE (Flexible Data-Driven Pipeline), an open-source framework that streamlines the preprocessing of data extracted from the EHR.Largely data-driven, FIDDLE systematically transforms structured EHR data into feature vectors, limiting the number of decisions a user must make while incorporating good practices from the literature. To demonstrate its utility and flexibility, we conducted a proof-of-concept experiment in which we applied FIDDLE to 2 publicly available EHR data sets collected from intensive care units: MIMIC-III and the eICU Collaborative Research Database. We trained different ML models to predict 3 clinically important outcomes: in-hospital mortality, acute respiratory failure, and shock. We evaluated models using the area under the receiver operating characteristics curve (AUROC), and compared it to several baselines.Across tasks, FIDDLE extracted 2,528 to 7,403 features from MIMIC-III and eICU, respectively. On all tasks, FIDDLE-based models achieved good discriminative performance, with AUROCs of 0.757-0.886, comparable to the performance of MIMIC-Extract, a preprocessing pipeline designed specifically for MIMIC-III. Furthermore, our results showed that FIDDLE is generalizable across different prediction times, ML algorithms, and data sets, while being relatively robust to different settings of user-defined arguments.FIDDLE, an open-source preprocessing pipeline, facilitates applying ML to structured EHR data. By accelerating and standardizing labor-intensive preprocessing, FIDDLE can help stimulate progress in building clinically useful ML tools for EHR data.