PULSE: Agentic Investigation with Passive Sensing for Proactive Intervention in Cancer Survivorship

2026-05-17

AI Summary

This study addresses the critical gap in timely psychological intervention for cancer survivors, who often miss opportunities due to insufficient proactive help-seeking. To overcome this, we propose a novel approach leveraging a large language model (LLM) agent that autonomously queries passively sensed smartphone data. By integrating personalized baselines with population-level retrieval-augmented reasoning, the agent emulates clinical hypothesis testing to dynamically identify emotional regulation needs and optimal intervention windows. Our method transcends the limitations of conventional fixed-feature pipelines, achieving a prediction accuracy of 0.713 for intervention availability using passive data alone. When augmented with diary entries, it attains a balanced accuracy of 0.743 in predicting willingness to engage in emotion regulation. This work pioneers the application of agent-driven, hypothesis-testing-style reasoning in passive sensing, substantially enhancing the timeliness and personalization of mental health interventions.

Abstract

Cancer survivors face elevated rates of depression, anxiety, and general emotional distress, yet the precise moments they most need support are often the moments when self-report is sparse, a phenomenon we term the diary paradox. Passive smartphone sensing offers a continuous, unobtrusive alternative, but prior sensing-based affect prediction has been limited by an accuracy ceiling, suggesting a bottleneck not only in available data, but in how behavioral signals are interpreted. We present PULSE, a system that shifts from fixed feature pipelines to agentic sensing investigation: LLM agents equipped with eight purpose-built tools autonomously query smartphone sensing data, compare current behavior against personalized baselines, and calibrate inferences through retrieval-augmented population-level comparisons. Rather than receiving pre-formatted feature summaries, agents decide which modalities to inspect, how far back to look, and how deeply to investigate, mirroring hypothesis-driven clinical reasoning. We evaluate PULSE through a 2*2 factorial design crossing reasoning architecture (structured vs. agentic) with data modality (sensing-only vs. with diary) on 50 cancer survivors from a longitudinal study of cancer survivors. Agentic reasoning is the primary driver of performance: agentic multimodal agent achieves balanced accuracy of 0.743 for emotion regulation desire with diary and sensing data, while agentic agents predict intervention availability at 0.713 with passive sensing data only. These results suggest that agentic investigation may be a cornerstone for unlocking the clinical value of passive sensing, advancing the feasibility of proactive just-in-time mental health support.

Problem

• cancer survivorship
• emotional distress
• passive sensing
• diary paradox
• proactive intervention

Innovation

• agentic reasoning
• passive sensing
• LLM agents
• emotion regulation prediction
• just-in-time intervention