2026-10-14T06:08:33Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34048739 2026-10-01T17:36:23Z category_1 category_13 category_21 category_734 category_931 category_811 category_69 category_128 portal_5 item_type_6 month_year_10_2026

<p>Cross-dataset distribution alignment quality.</p> Ali Namvar (9279350) Sundaresh Ram (10428439) Wassim W. Labaki (10704446) Stefanie Galban (10428445) Njira L. Lugogo (13119569) Craig J. Galban (10428469) Biophysics Genetics Biotechnology Biological Sciences not elsewhere classified Information Systems not elsewhere classified Mathematical Sciences not elsewhere classified Inorganic Chemistry Hematology track physiological instability support clinical assessment provide limited insight multidimensional physiological space monitoring systems face method &# 8217 existing approaches rely distinct clinical signatures applies geometric analysis 6 %, 16 movement toward higher state trajectory representation physiological state monitoring linking state dynamics derived physiological states aware monitoring ), mortality prediction achieved evaluated stream using fold higher mortality icu stay within expected calibration error routine icu data fold higher expected state icu mortality state outliers risk states respectively ), remained within outcome prediction nearest state assigned states vital signs strong discrimination static thresholds spent less severity scores laboratory values interpretable patterns external validation excellent calibration driven framework discover data developed stream critical gap 8 hours 72 hours 002 ). <div><p>Intensive care unit (ICU) monitoring systems face a critical gap: translating continuous physiological data into interpretable patterns that support clinical assessment. Existing approaches rely on static thresholds or severity scores that fail to capture dynamic disease progression patterns and provide limited insight into why patients change over time. We developed STREAM (State Trajectory Representation and Evolution-Aware Monitoring), which models each patient as a point in a multidimensional physiological space and applies geometric analysis of routine ICU data to track physiological instability. STREAM analyzes 26 routinely collected clinical measurements using optimal transport theory to discover data-derived physiological states without predefined categories and maps each patient to their nearest state. We evaluated STREAM using the multicenter eICU Collaborative Research Database (N = 158,294) for development and MIMIC-IV (N = 84,517) for external validation. STREAM identified five reproducible data-derived physiological states with distinct clinical signatures. Patients who spent less than 10% of their ICU stay within their expected state (state outliers) had ICU mortality of 37.6%, 16-fold higher than those who remained within their assigned states (2.3%). Mortality prediction achieved an area under the receiver operating characteristic curve of 0.863 at 8 hours and 0.903 at 72 hours, with excellent calibration (expected calibration error of 0.002). External validation on MIMIC-IV maintained robust performance (0.798 and 0.857, respectively), with state outliers exhibiting 10-fold higher mortality (33.5% vs. 3.2%). Feature importance analysis identified which laboratory values and vital signs are associated with movement toward higher-risk states, providing interpretable clinical explanations. STREAM provides transparent monitoring across data-derived physiological states, linking state dynamics with outcome prediction. Strong discrimination, calibration, and reproducibility across multicenter datasets support the method’s potential for prospective evaluation.</p></div> 2026-10-01T17:36:17Z Text Journal contribution 10.1371/journal.pdig.0001753.s004 https://figshare.com/articles/journal_contribution/_p_Cross-dataset_distribution_alignment_quality_p_/34048739 CC BY 4.0