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

TL;DR

This paper demonstrates how human-guided agentic AI improves multimodal clinical prediction tasks by combining domain expertise with automated workflows, leading to better accuracy and practical clinical insights.

Contribution

It introduces a human-in-the-loop approach for agentic AI in healthcare, showing significant performance gains across multiple clinical prediction benchmarks.

Findings

Human guidance yields +0.065 F1 improvement over automated baselines.

Multimodal feature engineering significantly enhances model performance.

Task-specific human judgment is crucial for effective multimodal data integration.

Abstract

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.