Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

QCAgent: An agentic framework for quality-controllable pathology report generation from whole slide image

Rundong Wang, Wei Ba, Ying Zhou, Yingtai Li, Bowen Liu, Baizhi Wang, Yuhao Wang, Zhidong Yang, Kun Zhang, Rui Yan, S. Kevin Zhou

TL;DR

Experiments demonstrate that by making report requirements explicitly prompt-defined, constraint-aware, and verifiable through evidence-grounded refinement, QCAgent enables controllable generation of clinically meaningful and high-coverage pathology reports from WSI.

Abstract

Recent methods for pathology report generation from whole-slide image (WSI) are capable of producing slide-level diagnostic descriptions but fail to ground fine-grained statements in localized visual evidence. Furthermore, they lack control over which diagnostic details to include and how to verify them. Inspired by emerging agentic analysis paradigms and the diagnostic workflow of pathologists,who selectively examine multiple fields of view, we propose QCAgent, an agentic framework for quality-controllable WSI report generation. The core innovations of this framework are as follows: (i) it incorporates a customized critique mechanism guided by a user-defined checklist specifying required diagnostic details and constraints; (ii) it re-identifies informative regions in the WSI based on the critique feedback and text-patch semantic retrieval, a process that iteratively enriches and reconciles the report. Experiments demonstrate that by making report requirements explicitly prompt-defined, constraint-aware, and verifiable through evidence-grounded refinement, QCAgent enables controllable generation of clinically meaningful and high-coverage pathology reports from WSI.

QCAgent: An agentic framework for quality-controllable pathology report generation from whole slide image

TL;DR

Experiments demonstrate that by making report requirements explicitly prompt-defined, constraint-aware, and verifiable through evidence-grounded refinement, QCAgent enables controllable generation of clinically meaningful and high-coverage pathology reports from WSI.

Abstract

Recent methods for pathology report generation from whole-slide image (WSI) are capable of producing slide-level diagnostic descriptions but fail to ground fine-grained statements in localized visual evidence. Furthermore, they lack control over which diagnostic details to include and how to verify them. Inspired by emerging agentic analysis paradigms and the diagnostic workflow of pathologists,who selectively examine multiple fields of view, we propose QCAgent, an agentic framework for quality-controllable WSI report generation. The core innovations of this framework are as follows: (i) it incorporates a customized critique mechanism guided by a user-defined checklist specifying required diagnostic details and constraints; (ii) it re-identifies informative regions in the WSI based on the critique feedback and text-patch semantic retrieval, a process that iteratively enriches and reconciles the report. Experiments demonstrate that by making report requirements explicitly prompt-defined, constraint-aware, and verifiable through evidence-grounded refinement, QCAgent enables controllable generation of clinically meaningful and high-coverage pathology reports from WSI.
Paper Structure (13 sections, 1 equation, 3 figures, 2 tables, 1 algorithm)

This paper contains 13 sections, 1 equation, 3 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Method
  3. Problem Formulation.
  4. Initial Round: Global Draft and Coverage-Oriented Evidence Extraction.
  5. Quality-Control Rounds: Feedback-Guided Evidence Seeking and Evidence-Driven Revision.
  6. Experiments and Results
  7. Datasets and Evaluation Metrics
  8. Datasets.
  9. Evaluation metrics.
  10. Baseline and implementation details.
  11. Experimental Results
  12. Ablation Study
  13. Conclusion

Figures (3)

  • Figure 1: Pipeline of QCAgent: (a) initial-round report generation using PRISM and cluster-based patch selection (Tool-1) with Patho-R1 patch reports; (b) QC-round iterative refinement using feedback-guided CONCH retrieval (Tool-2) and Patho-R1 evidence until QC criteria are met.
  • Figure 2: Qualitative comparison of pathology reports on TCGA-STAD. For each case, we show the WSI thumbnail, the ground-truth clinical report, the one-pass draft generated by PRISM, and our QC-refined report.
  • Figure 3: Case study of the iterative audit--retrieve--revise workflow. Left: WSI thumbnail and patch clustering, with PRISM providing the initial global context. Middle/Right: across QC rounds, QCAgent issues missing-field queries, retrieves supporting patches, and revises the report, improving completeness while marking non-image-verifiable items as undetermined.