aiqbench analysis
This research proposes a diffusion-based method to generate synthetic clinical time series that explicitly model the informative nature of missing laboratory data, treating observation patterns as integral to patient physiology rather than noise. By jointly learning continuous lab values and discrete sampling indicators on the MIMIC-III benchmark, the approach demonstrates how AI can capture complex dependencies between clinician behavior and underlying medical conditions under non-random missingness scenarios. These results lay groundwork for future clinical foundation models that leverage synthetic priors preserving realistic testing behaviors to enhance downstream diagnostic tasks.
Informative Missingness to Generate Irregular Clinical Time Series
arXiv:2606.17106v1 Announce Type: new Abstract: Laboratory tests in electronic health records are collected irregularly, and the absence of a test order can be as informative as the measurement itself. Such missingness reflects clinicians' decisions and patient physiology, making it important to model it directly rather than treat it as a preprocessing artifact. Here we present a diffusion-based approach for generating clinical time series that jointly models laboratory values and their observation patterns using the public Data Analytics Challenge on Missing Data Imputation (DACMI) benchmark derived from MIMIC-III. To preserve realistic sampling, we align chart times into 4-hour intervals and segment admissions into 7-day windows, producing trajectories that pair each lab value with a corresponding observation indicator. Standard transformations and normalization are applied to stabilize training. Our method extends the TimeDiff framework to learn continuous lab values and discrete missingness patterns through complementary diffusion objectives. Experiments show that the generated data closely match real patient trajectories across individual lab distributions and joint value-missingness embeddings, demonstrating that diffusion models can capture clinically meaningful dependencies between patient physiology and clinicians' testing behavior under MNAR-like (missing-not-at-random) missingness. These preliminary results indicate that our model can serve as an initial component toward developing clinical foundation models. By producing synthetic priors that preserve key physiology-missingness relationships, this work motivates the subsequent training of Prior-Data Fitted Networks capable of leveraging informative missingness, which we will investigate in the extended work.
Abstract:Laboratory tests in electronic health records are collected irregularly, and the absence of a test order can be as informative as the measurement itself. Such missingness reflects clinicians' decisions and patient physiology, making it important to model it directly rather than treat it as a preprocessing artifact. Here we present a diffusion-based approach for generating clinical time series that jointly models laboratory values and their observation patterns using the public Data Analytics Challenge on Missing Data Imputation (DACMI) benchmark derived from MIMIC-III. To preserve realistic sampling, we align chart times into 4-hour intervals and segment admissions into 7-day windows, producing trajectories that pair each lab value with a corresponding observation indicator. Standard transformations and normalization are applied to stabilize training. Our method extends the TimeDiff framework to learn continuous lab values and discrete missingness patterns through complementary diffusion objectives. Experiments show that the generated data closely match real patient trajectories across individual lab distributions and joint value-missingness embeddings, demonstrating that diffusion models can capture clinically meaningful dependencies between patient physiology and clinicians' testing behavior under MNAR-like (missing-not-at-random) missingness. These preliminary results indicate that our model can serve as an initial component toward developing clinical foundation models. By producing synthetic priors that preserve key physiology-missingness relationships, this work motivates the subsequent training of Prior-Data Fitted Networks capable of leveraging informative missingness, which we will investigate in the extended work.
Machine Learning (cs.LG); Computers and Society (cs.CY)
arXiv:2606.17106 [cs.LG]
(or arXiv:2606.17106v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.17106 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 324 words
Related coverage
Represent, Then Generate: Multimodal-Conditioned Time-Series Generation under Irregular Missingness
arXiv · Aug 19, 2026
Multimodal Prompt Learning with Irregular EHRs for Robust Monitoring of Critical Care Patients
arXiv · Aug 25, 2026
NOAH: Learning the Full Patient Journey. A Longitudinal Multimodal Time-Aware Model for Representation and Forecasting
arXiv · Sep 9, 2026
News
Quantum Computing Report · Jun 7, 2026