Interpretable Trajectory-Based Feature Extraction from Longitudinal Electronic Health Records
- 1MGB
Abstract
Longitudinal electronic health records (EHRs) contain rich information about the evolution of a patients clinical state, but temporal measurements and events are often represented using summary statistics or complex learned representations that can be difficult to interpret clinically. We propose an interpretable trajectory-based feature representation that transforms heterogeneous longitudinal EHR data into compact, human-readable predictive features. Numerical temporal variables are represented by trajectory states such as INCREASING, DECREASING, and STABLE, while categorical clinical events are represented by interpretable states describing repeated occurrence, stability, or change. These trajectory features are combined with static patient and encounter characteristics and evaluated using L1-regularized logistic regression, which provides sparse feature selection and direct identification of positive and negative predictive features.
The approach is evaluated using MIMIC-IV in two clinically distinct prediction tasks: in-hospital mortality following ICU admission and hospital readmission within 30 days after discharge. For ICU mortality, multiple observation windows ranging from 1 to 48 hours are investigated to examine how the availability of longitudinal information affects prediction and the relative contribution of trajectory features. The highest AUROC of 0.8751 is obtained with a 32-hour observation window. For 30-day readmission, trajectories from the final 72 hours of the index hospitalization are evaluated using both three-state and five-state categorical representations. The two representations produce essentially identical discriminative performance (AUROC 0.6875 and 0.6876, respectively), indicating that increasing trajectory granularity provides little predictive benefit in this setting.
The results demonstrate that heterogeneous longitudinal EHR observations can be transformed into compact and directly interpretable temporal features while retaining useful predictive information. The proposed framework therefore provides a simple and model-compatible approach for identifying clinically interpretable patterns of patient evolution and can be extended with more detailed trajectory definitions or applied with other predictive models.
Related
- Hybrid feature pyramid attention and capsule network (FpaNet+CapsNet) for enhanced spatial and contextual feature extraction for improved brain tumor segmentation
- Scalable Causal-Interpretable Machine Learning for Cancer Prescreening Using Electronic Health Records
- Multi-task deep learning framework for brain tumor classification and segmentation using ensemble feature extraction
- HAFI-DTA: an interpretable multimodal framework for drug-target affinity prediction based on hierarchical attention guidance and feature iterative interaction
- Enhancing medical image segmentation with HCE-Net: A heterogeneous feature collaborative extraction network
BibTeX
@misc{bulgakov2026interpretable,
title = {{Interpretable Trajectory-Based Feature Extraction from Longitudinal Electronic Health Records}},
author = {V. Bulgakov and A. Turchin},
howpublished = {medRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.14.26363036},
url = {https://www.medrxiv.org/content/10.64898/2026.09.14.26363036v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Interpretable Trajectory-Based Feature Extraction from Longitudinal Electronic Health Records
AU - V. Bulgakov
AU - A. Turchin
PY - 2026
DA - 2026/09/14
PB - medRxiv
DO - 10.64898/2026.09.14.26363036
UR - https://www.medrxiv.org/content/10.64898/2026.09.14.26363036v1
AB - Longitudinal electronic health records (EHRs) contain rich information about the evolution of a patients clinical state, but temporal measurements and events are often represented using summary statistics or complex learned representations that can be difficult to interpret clinically. We propose an interpretable trajectory-based feature representation that transforms heterogeneous longitudinal EHR data into compact, human-readable predictive features. Numerical temporal variables are represented by trajectory states such as INCREASING, DECREASING, and STABLE, while categorical clinical events are represented by interpretable states describing repeated occurrence, stability, or change. These trajectory features are combined with static patient and encounter characteristics and evaluated using L1-regularized logistic regression, which provides sparse feature selection and direct identification of positive and negative predictive features. The approach is evaluated using MIMIC-IV in two clinically distinct prediction tasks: in-hospital mortality following ICU admission and hospital readmission within 30 days after discharge. For ICU mortality, multiple observation windows ranging from 1 to 48 hours are investigated to examine how the availability of longitudinal information affects prediction and the relative contribution of trajectory features. The highest AUROC of 0.8751 is obtained with a 32-hour observation window. For 30-day readmission, trajectories from the final 72 hours of the index hospitalization are evaluated using both three-state and five-state categorical representations. The two representations produce essentially identical discriminative performance (AUROC 0.6875 and 0.6876, respectively), indicating that increasing trajectory granularity provides little predictive benefit in this setting. The results demonstrate that heterogeneous longitudinal EHR observations can be transformed into compact and directly interpretable temporal features while retaining useful predictive information. The proposed framework therefore provides a simple and model-compatible approach for identifying clinically interpretable patterns of patient evolution and can be extended with more detailed trajectory definitions or applied with other predictive models.
ER -