arXiv subjects

Lalitha Pranathi Pulavarthy

Publications and source records attributed to Lalitha Pranathi Pulavarthy.

4 recordsLinked to original sources

arXiv2026

Machine Learning for Pre-Culture ESBL Risk Stratification to Guide Empiric Antibiotic Selection: A 12-Hospital Study of Enterobacteriaceae Cultures

Empiric antibiotic therapy for suspected ESBL-producing Enterobacteriaceae must be selected 48-72 hours before culture results, forcing clinicians to choose between undertreating resistant infections and overusing carbapenems that drive further resistance. We developed a cost-sensitive XGBoost model predicting an ESBL phenotype (resistance to ceftriaxone, ceftazidime, cefepime or piperacillin-tazobactam) at culture ordering using 45 pre-culture EHR features across 132,955 cultures from 72,217 patients at 12 hospitals (14.41% with the ESBL phenotype). Cultures were partitioned at the patient level. At 90% sensitivity, the model achieved 95.8% NPV, reducing post-test ESBL probability to 4.2%, a threshold that may support safe carbapenem-sparing in non-ICU settings, while sparing 307 of every 1,000 cultures an unnecessary broad-spectrum course at the cost of 14 missed ESBL cases per 1,000. SHAP analysis identified prior ESBL colonization as the dominant predictor, ahead of prior organism burden and neighborhood deprivation; removing deprivation features caused minimal performance loss ($Δ\text{AUROC} = -0.020$), enabling equitable bedside deployment. Discrimination was unchanged under a strict IDSA ESBL-E definition (AUROC 0.766), with specimen type added as a predictor (0.764) and without any class-imbalance correction (0.762), and ranged from 0.71 to 0.78 across organism strata.

cs.LG↗

arXiv2026

Predicting Early Functional Decline from Longitudinal Laboratory and Vital Sign Trajectories: A Large-Scale Study Using the All of Us Research Program

Functional decline in older adults is typically recognized only after falls or observable gait impairment, closing the window for prevention. We investigated whether temporal trajectories of routine biomarkers, already recorded but rarely analyzed longitudinally, can identify patients in the pre-clinical phase of mobility decline. Using the All of Us Research Program (N = 297,861; 11.1% cases), we derived trajectory features (slope, variability, delta, mean) for twelve biomarkers over a three-year pre-index window. LightGBM models incorporating trajectories significantly outperformed static laboratory summaries (AUROC 0.797 vs. 0.755; DeLong p < 0.001; AUPRC 0.380 vs. 0.304). A 1:1 age- and sex-matched analysis confirmed an independent trajectory signal (AUROC 0.727 vs. demographics-only 0.680). A horizon analysis demonstrated sustained prediction 3-12 months before decline onset (AUROC 0.768-0.740). Because the model uses only measurements already ordered in routine care, it supports passive, zero-burden EHR integration for early detection of pre-clinical functional decline.

cs.LG↗

arXiv2026

Multimodal EEG-IMU Fusion for Motor Assessment: Leveraging Task-Dependent Complementarity for Robustness

Movement disorders such as Parkinson's disease require comprehensive motor assessment, but reliable digital assessment pipelines integrating multiple sensing modalities across diverse motor tasks remain insufficiently characterized. We present a proof-of-concept study evaluating task-specific modality performance and multimodal fusion across ten motor activities. Synchronized EEG-IMU data were recorded from six participants (52 recording pairs). We evaluated an EEGNet + Transformer model for 16-channel EEG (125 Hz) and XGBoost on hand-crafted accelerometer and gyroscope features (25 Hz). Under 5-fold cross-validation in a subject-dependent setting, IMU achieved 94.41+/-0.58% accuracy and outperformed EEG on 7 of 10 activities, while EEG achieved 92.82+/-1.45% and showed lower error for rhythmic cycling (4.03% vs. 12.10%). Late fusion via logistic regression reached 98.68+/-0.32%, giving an 81.5% error reduction versus EEG alone and improving worst-task accuracy from approximately 87% for a single modality to 96.76%. Fusion also reduced cross-task performance variance from approximately 3% to 1.06% (paired t-test, p < 0.001, df = 4; p-values approximate given fold dependence), showing more uniform reliability across the assessment battery. Although the small sample limits generalizability, these results suggest that EEG and IMU provide asymmetric, task-dependent strengths and that late fusion can leverage this complementarity to improve assessment reliability. This study provides methodological and empirical motivation for larger-scale clinical validation in movement disorder populations.

eess.SP↗

arXiv2026

Human-Guided Agentic AI for Multimodal Clinical Prediction: Lessons from the AgentDS Healthcare Benchmark

Agentic AI systems are increasingly capable of autonomous data science workflows, yet clinical prediction tasks demand domain expertise that purely automated approaches struggle to provide. We investigate how human guidance of agentic AI can improve multimodal clinical prediction, presenting our approach to all three AgentDS Healthcare benchmark challenges: 30-day hospital readmission prediction (Macro-F1 = 0.8986), emergency department cost forecasting (MAE = $465.13), and discharge readiness assessment (Macro-F1 = 0.7939). Across these tasks, human analysts directed the agentic workflow at key decision points, multimodal feature engineering from clinical notes, scanned PDF billing receipts, and time-series vital signs; task-appropriate model selection; and clinically informed validation strategies. Our approach ranked 5th overall in the healthcare domain, with a 3rd-place finish on the discharge readiness task. Ablation studies reveal that human-guided decisions compounded to a cumulative gain of +0.065 F1 over automated baselines, with multimodal feature extraction contributing the largest single improvement (+0.041 F1). We distill three generalizable lessons: (1) domain-informed feature engineering at each pipeline stage yields compounding gains that outperform extensive automated search; (2) multimodal data integration requires task-specific human judgment that no single extraction strategy generalizes across clinical text, PDFs, and time-series; and (3) deliberate ensemble diversity with clinically motivated model configurations outperforms random hyperparameter search. These findings offer practical guidance for teams deploying agentic AI in healthcare settings where interpretability, reproducibility, and clinical validity are essential.

cs.AI↗