Renaissance of RNNs in Streaming Clinical Time Series: Compact Recurrence Remains Competitive with Transformers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tong, Ran, Liu, Jiaqi, Liu, Su, Hu, Xin, Wang, Lanruo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917026731130880
author Tong, Ran
Liu, Jiaqi
Liu, Su
Hu, Xin
Wang, Lanruo
author_facet Tong, Ran
Liu, Jiaqi
Liu, Su
Hu, Xin
Wang, Lanruo
contents We present a compact, strictly causal benchmark for streaming clinical time series on the MIT--BIH Arrhythmia Database using per-second heart rate. Two tasks are studied under record-level, non-overlapping splits: near-term tachycardia risk (next ten seconds) and one-step heart rate forecasting. We compare a GRU-D (RNN) and a Transformer under matched training budgets against strong non-learned baselines. Evaluation is calibration-aware for classification and proper for forecasting, with temperature scaling and grouped bootstrap confidence intervals. On MIT-BIH, GRU-D slightly surpasses the Transformer for tachycardia risk, while the Transformer clearly lowers forecasting error relative to GRU-D and persistence. Our results show that, in longitudinal monitoring, model choice is task-dependent: compact RNNs remain competitive for short-horizon risk scoring, whereas compact Transformers deliver clearer gains for point forecasting.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16677
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Renaissance of RNNs in Streaming Clinical Time Series: Compact Recurrence Remains Competitive with Transformers
Tong, Ran
Liu, Jiaqi
Liu, Su
Hu, Xin
Wang, Lanruo
Machine Learning
Artificial Intelligence
We present a compact, strictly causal benchmark for streaming clinical time series on the MIT--BIH Arrhythmia Database using per-second heart rate. Two tasks are studied under record-level, non-overlapping splits: near-term tachycardia risk (next ten seconds) and one-step heart rate forecasting. We compare a GRU-D (RNN) and a Transformer under matched training budgets against strong non-learned baselines. Evaluation is calibration-aware for classification and proper for forecasting, with temperature scaling and grouped bootstrap confidence intervals. On MIT-BIH, GRU-D slightly surpasses the Transformer for tachycardia risk, while the Transformer clearly lowers forecasting error relative to GRU-D and persistence. Our results show that, in longitudinal monitoring, model choice is task-dependent: compact RNNs remain competitive for short-horizon risk scoring, whereas compact Transformers deliver clearer gains for point forecasting.
title Renaissance of RNNs in Streaming Clinical Time Series: Compact Recurrence Remains Competitive with Transformers
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.16677