Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Human-in-the-Loop Interactive Report Generation for Chronic Disease Adherence

Xiaotian Zhang, Jinhong Yu, Pengwei Yan, Le Jiang, Xingyi Shen, Mumo Cheng, Xiaozhong Liu

TL;DR

The paper addresses the burden of generating timely, personalized adherence feedback for chronic disease management and the tension between scalable AI generation and clinical trust. It introduces a clinician-in-the-loop interface that confines AI to data organization, uses a single-page editor with chart-linked visuals, and applies automated urgency flags with fail-safe escalation. In a pilot with 3 physicians reviewing 24 reports, AI drafts achieved near-baseline quality (mean 4.86/10 vs 5.0 baseline) with minimal edits (8.3%), and no safety-critical issues. However, time savings did not materialize due to an accountability paradox: complete physician verification remains mandatory in high-stakes care, underscoring the need for selective verification mechanisms and governance to harness AI effectively.

Abstract

Chronic disease management requires regular adherence feedback to prevent avoidable hospitalizations, yet clinicians lack time to produce personalized patient communications. Manual authoring preserves clinical accuracy but does not scale; AI generation scales but can undermine trust in patient-facing contexts. We present a clinician-in-the-loop interface that constrains AI to data organization and preserves physician oversight through recognition-based review. A single-page editor pairs AI-generated section drafts with time-aligned visualizations, enabling inline editing with visual evidence for each claim. This division of labor (AI organizes, clinician decides) targets both efficiency and accountability. In a pilot with three physicians reviewing 24 cases, AI successfully generated clinically personalized drafts matching physicians' manual authoring practice (overall mean 4.86/10 vs. 5.0/10 baseline), requiring minimal physician editing (mean 8.3\% content modification) with zero safety-critical issues, demonstrating effective automation of content generation. However, review time remained comparable to manual practice, revealing an accountability paradox: in high-stakes clinical contexts, professional responsibility requires complete verification regardless of AI accuracy. We contribute three interaction patterns for clinical AI collaboration: bounded generation with recognition-based review via chart-text pairing, automated urgency flagging that analyzes vital trends and adherence patterns with fail-safe escalation for missed critical monitoring tasks, and progressive disclosure controls that reduce cognitive load while maintaining oversight. These patterns indicate that clinical AI efficiency requires not only accurate models, but also mechanisms for selective verification that preserve accountability.

Human-in-the-Loop Interactive Report Generation for Chronic Disease Adherence

TL;DR

The paper addresses the burden of generating timely, personalized adherence feedback for chronic disease management and the tension between scalable AI generation and clinical trust. It introduces a clinician-in-the-loop interface that confines AI to data organization, uses a single-page editor with chart-linked visuals, and applies automated urgency flags with fail-safe escalation. In a pilot with 3 physicians reviewing 24 reports, AI drafts achieved near-baseline quality (mean 4.86/10 vs 5.0 baseline) with minimal edits (8.3%), and no safety-critical issues. However, time savings did not materialize due to an accountability paradox: complete physician verification remains mandatory in high-stakes care, underscoring the need for selective verification mechanisms and governance to harness AI effectively.

Abstract

Chronic disease management requires regular adherence feedback to prevent avoidable hospitalizations, yet clinicians lack time to produce personalized patient communications. Manual authoring preserves clinical accuracy but does not scale; AI generation scales but can undermine trust in patient-facing contexts. We present a clinician-in-the-loop interface that constrains AI to data organization and preserves physician oversight through recognition-based review. A single-page editor pairs AI-generated section drafts with time-aligned visualizations, enabling inline editing with visual evidence for each claim. This division of labor (AI organizes, clinician decides) targets both efficiency and accountability. In a pilot with three physicians reviewing 24 cases, AI successfully generated clinically personalized drafts matching physicians' manual authoring practice (overall mean 4.86/10 vs. 5.0/10 baseline), requiring minimal physician editing (mean 8.3\% content modification) with zero safety-critical issues, demonstrating effective automation of content generation. However, review time remained comparable to manual practice, revealing an accountability paradox: in high-stakes clinical contexts, professional responsibility requires complete verification regardless of AI accuracy. We contribute three interaction patterns for clinical AI collaboration: bounded generation with recognition-based review via chart-text pairing, automated urgency flagging that analyzes vital trends and adherence patterns with fail-safe escalation for missed critical monitoring tasks, and progressive disclosure controls that reduce cognitive load while maintaining oversight. These patterns indicate that clinical AI efficiency requires not only accurate models, but also mechanisms for selective verification that preserve accountability.
Paper Structure (8 sections, 3 figures, 1 table)

This paper contains 8 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method and Approach
  4. Evaluation
  5. Setup
  6. Results
  7. Discussion and Implications
  8. Conclusion

Figures (3)

  • Figure 1: System architecture and feedback loop. Data flows from patient through unified memory to parallel AI processing: content generation creates sectioned drafts, while risk assessment assigns urgency flags. The single-page HTML editor integrates both for physician review. Left panel shows four design principles. Right side shows the doctor-AI-patient feedback cycle.
  • Figure 2: Physician interface for chronic disease management report.
  • Figure 3: Structured review workflow. One short pass: check essentials, make focused edits, approve and export.