LLM-Augmented Computational Phenotyping of Long Covid

Jing Wang; Jie Shen; Amar Sra; Qiaomin Xie; Jeremy C Weiss

cs.LGMar 18, 2026

LLM-Augmented Computational Phenotyping of Long Covid

Authors: Jing Wang, Jie Shen, Amar Sra, Qiaomin Xie, Jeremy C Weiss

Abstract

Phenotypic characterization is essential for understanding heterogeneity in chronic diseases and for guiding personalized interventions. Long COVID, a complex and persistent condition, yet its clinical subphenotypes remain poorly understood. In this work, we propose an LLM-augmented computational phenotyping framework ``Grace Cycle'' that iteratively integrates hypothesis generation, evidence extraction, and feature refinement to discover clinically meaningful subgroups from longitudinal patient data. The framework identifies three distinct clinical phenotypes, Protected, Responder, and Refractory, based on 13,511 Long Covid participants. These phenotypes exhibit pronounced separation in peak symptom severity, baseline disease burden, and longitudinal dose-response patterns, with strong statistical support across multiple independent dimensions. This study illustrates how large language models can be integrated into a principled, statistically grounded pipeline for phenotypic screening from complex longitudinal data. Note that the proposed framework is disease-agnostic and offers a general approach for discovering clinically interpretable subphenotypes.

Figures & tables

Figure 1: The pipeline of the LLM augmented phyenotyping for Long Covid, we called “Grace Circle”. We starts with a hypothesis, LLM looks for evidence from the data to support the hypothesis. With feedback about alignment with hypothesis and evidence by LLM, update the data and hypothesis. Repeat the process until the evidence are aligned with hypothesis.

Feature	Protected	Responder	Refractory	Test Statistic	p-value
Peak PASC Score	$4.98\pm 6.3$	$18.40\pm 7.1$	$22.80\pm 5.1$	$H=4215.2$	$<0.001$
Initial Score	$4.98$	$8.50$	$14.20$	$F=128.4$	$<0.001$
Avg. Vaccine Doses	$3.41$	$2.38$	$1.82$	$F=84.2$	$<0.001$

Table 1: Group comparisons across Protected, Responder, and Refractory cohorts.

Figure 2: The vaccine and wellness (PASC score) trajectory for Long Covid participants.

Figure 3: The steps in the pipeline. LLM performs pairwise comparison, fairness criteria in the tabular data reading. The hypothesis updates by choosing different hypothesis from the pool. Evidence is updated by collecting related features, replacing or removing unrelated features from the knowledge base. The process continues until hypothesis and evidence are aligned.

Figure 4: The update of hypothesis and feature space given the Long Covid cohort. The target variable is the PASC score which reflects the overall wellness of the participants of Long Covid.

Figure 5: Example of insights of LLM after reading trajectories of multiple participants.

Characteristic	Protected	Responder	Refractory
# Participants	9,544	3302	665
Age, median (IQR)	44.0 (33.0-60)	49.0 (37-60)	48.0 (39-57)
Age category at enrollment
18-45	32%	41%	48%
46-64	52%	44%	43%
$>$ 65	16%	15%	9%

Table 2: Demographic characteristics.

Panel of Figure 6 — Figure 6: Distribution of PASC severity across patient cohorts. A, Distribution of peak PASC severity and individual score variance. B, Distribution of initial PASC severity and individual score variance.

Figure 7: Longitudinal dose-response of PASC severity differs by clinical phenotype.

Figure 8: PASC severity shows weak but significant correlations with time and prior vaccination across phenotypes.

Large Language Model-Guided Scientific Discovery

Large language model-guided scientific discovery uses AI to propose, test, and refine hypotheses, models, experiments, and designs.

See all 274 papers

Large language models as synthetic clinical experts to inform longitudinal rare-disease modeling
Aug 17, 2026Clemens Schächter, Astrid Pechmann, Janbernd Kirschner +2Representation Learning Latent Variable Models
Institute of Medical Biometry and Statistics (IMBI), Faculty of Medicine and Medical Center – University of Freiburg, Freiburg, Germany · Freiburg Center for Data Analysis, Modeling and AI – University of Freiburg, Freiburg, Germany · Department of Neuropediatrics and Muscle Disorders, Faculty of Medicine and Medical Center – University of Freiburg, Freiburg, Germany +3
COTCAgent: Preventive Consultation via Probabilistic Chain-of-Thought Completion
May 14, 2026Zihan Deng, Xiaozhen Zhong, Chuanzhi XuHealthcare Clinical Decision Support
School of Computing and Data Science, The University of Hong Kong · 2Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China · School of Computer Science, The University of Sydney
Grounded in Consensus, In Step With Emerging Science: A Consensus-Anchored Multi-Corpus Clinical Chatbot for Long COVID
Jul 27, 2026Yining Wu, Philip DiGiacomo, Ying Ding +1Retrieval-Augmented Generation Clinical Decision Support
School of Information University of Texas at Austin Austin, TX, USA · Department of Computer Science University of Texas at Austin Austin, TX, USA · School of Information, Dell Medical School University of Texas at Austin Austin, TX, USA +1

Aug 17, 2026J/K move · Enter open · S save

Large language models as synthetic clinical experts to inform longitudinal rare-disease modeling

Clemens Schächter, Astrid Pechmann, Janbernd Kirschner, Jan Hasenauer+1

Institute of Medical Biometry and Statistics (IMBI), Faculty of Medicine and Medical Center – University of Freiburg, Freiburg, Germany · Freiburg Center for Data Analysis, Modeling and AI – University of Freiburg, Freiburg, Germany · Department of Neuropediatrics and Muscle Disorders, Faculty of Medicine and Medical Center – University of Freiburg, Freiburg, Germany+3

Representation Learning Latent Variable Models Longitudinal EHR Modeling

Open paper AskPDF arXiv

Due to the limited amount of information, modeling longitudinal rare-disease data can benefit from integrating clinical knowledge. Yet, elicitation of expert knowledge and formalization for model fitting is challenging, in particular due to limited time of clinical experts. To nevertheless make domain knowledge accessible during model fitting, we use large language models (LLMs) as synthetic clinical experts to supervise a variational-autoencoder-based approach that learns low-dimensional latent summaries of visit-level observations. Specifically, LLMs are queried offline on textual descriptions of patient observations to obtain judgments, e.g., the suspected clinical category. To improve the variational autoencoder fit, we train a differentiable surrogate model on these judgments and augment the loss function to encourage reconstructions that preserve the clinical-label distribution of their corresponding input profile. In an application to longitudinal motor-function assessments from children with spinal muscular atrophy, we map visit-level clinical profiles to low-dimensional representations that are linked by a multivariate mixed-effects model. The synthetic expert loss discourages reconstructions that remain numerically close in data space but alter the clinical interpretation of the reconstructed motor function profile, such as by crossing a disease-type boundary. We thus reduced disagreement between original and reconstructed SMA type labels from about 11 to 7 percent. Furthermore, informing the latent representation by the synthetic expert improved prediction of motor function milestones compared with unsupervised latent representations and a data-level baseline. These results suggest that incorporating LLMs into model fitting can make clinical knowledge available to representation learning and improve clinical faithfulness for longitudinal rare-disease data.

Abstract

Large Language Model-Guided Scientific Discovery

Explore similar work

Large language models as synthetic clinical experts to inform longitudinal rare-disease modeling

COTCAgent: Preventive Consultation via Probabilistic Chain-of-Thought Completion

Grounded in Consensus, In Step With Emerging Science: A Consensus-Anchored Multi-Corpus Clinical Chatbot for Long COVID