Physiology as Topology: Intervenable Organ Graphs for Longitudinal Patient Forecasting
Abstract
Do language models forecast patient trajectories by reasoning through physiology, or by fitting surface statistics? We introduce PhysioCircuit, a multi-agent agent system that represents each organ as an agent and each physiological relation as an editable mechanism card with direction, lag, and evidence. Its expert-defined graph is frozen at inference, enabling controlled edge interventions. We also introduce PhysioCircuit-Bench, a mechanism-controlled benchmark spanning three organ systems, treatment events, and real ICU admissions. We evaluate intervention-direction agreement (IDAR), which requires the affected organ to change in the expected direction within the expected lag window. The expert graph substantially outperforms all topology controls on edge-IDAR, including fully connected, random, and reversed graphs. Independent clinician adjudication supports the counterfactual results. PhysioCircuit transfers to new organ systems by configuration alone, without retraining, while preserving inspectable and editable relations.
Related
- Non-invasive forecasting of skin cancer evolution through longitudinal hair sampling
- Reliability-Centered Evaluation of Sparse Longitudinal CT Lesion-Size Forecasting with Conformal Interval Calibration and Gompertz-Inspired Regularization
- Fatigue burden and patient attribution across remission states in inflammatory bowel disease: a longitudinal cohort study.
- A longitudinal analysis of real world symptom tracking and patient-reported outcomes in Brisa, a German Digital Health App for multiple sclerosis
- Replacing EEG with peripheral physiological signals for driving behavior recognition: a physiology-informed multimodal approach
BibTeX
@misc{liang2026physiology,
title = {{Physiology as Topology: Intervenable Organ Graphs for Longitudinal Patient Forecasting}},
author = {Xinru Liang and Yongchao Chen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.33022},
eprint = {2610.33022},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.33022},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Physiology as Topology: Intervenable Organ Graphs for Longitudinal Patient Forecasting
AU - Xinru Liang
AU - Yongchao Chen
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2610.33022
UR - https://arxiv.org/abs/2610.33022
AB - Do language models forecast patient trajectories by reasoning through physiology, or by fitting surface statistics? We introduce PhysioCircuit, a multi-agent agent system that represents each organ as an agent and each physiological relation as an editable mechanism card with direction, lag, and evidence. Its expert-defined graph is frozen at inference, enabling controlled edge interventions. We also introduce PhysioCircuit-Bench, a mechanism-controlled benchmark spanning three organ systems, treatment events, and real ICU admissions. We evaluate intervention-direction agreement (IDAR), which requires the affected organ to change in the expected direction within the expected lag window. The expert graph substantially outperforms all topology controls on edge-IDAR, including fully connected, random, and reversed graphs. Independent clinician adjudication supports the counterfactual results. PhysioCircuit transfers to new organ systems by configuration alone, without retraining, while preserving inspectable and editable relations.
ER -