CTIS-QA: Clinical Template-Informed Slide-Level Question Answering for Pathology

2025-12-15

IEEE International Conference on Bioinformatics and Biomedicine

AI Summary

This work addresses the lack of clinical grounding, standardized data, and diagnostic workflow alignment in existing pathological visual question answering (VQA) methods. To bridge this gap, the authors construct a Clinical Pathology Report Template (CPRT) based on the College of American Pathologists (CAP) cancer protocols, enabling systematic extraction of pathological features. They introduce CTIS-Bench—the first slide-level VQA benchmark clinically aligned with real-world diagnostic practices—and curate the CTIS-Align dataset to facilitate vision–language alignment training. Furthermore, they propose CTIS-QA, a dual-stream architecture that mimics pathologists’ “global screening–local focusing” visual strategy. Experiments demonstrate that CTIS-QA significantly outperforms state-of-the-art methods on WSI-VQA, CTIS-Bench, and multiple downstream diagnostic tasks.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Multimodal large language models (MLLMs) have demonstrated strong performance in patch-level pathological image analysis; however, they often lack the holistic perceptual capability necessary for comprehensive Whole Slide Image (WSI) interpretation. Recent approaches have explored constructing slide-level MLLMs using VQA datasets that are entirely generated from pathology reports by large language models (LLMs). However, these datasets suffer from critical limitations: hallucinated content, information leakage in question stems, clinically irrelevant or visual independent questions, and the omission of essential diagnostic features-issues that undermine both data quality and clinical validity. In this paper, we introduce a clinical diagnosis template-based pipeline to collect pathological information. In collaboration with pathologists and guided by the the College of American Pathologists (CAP) Cancer Protocols, we design a Clinical Pathology Report Template (CPRT) that ensures comprehensive and standardized extraction of diagnostic elements from pathology reports. We validate the effectiveness of our pipeline on TCGA-BRCA. First, we extract pathological features from reports using CPRT. These features are then used to build CTIS-Align, a dataset of 80k slide-description pairs from 804 WSIs for vision-language alignment training, and CTISBench, a rigorously curated VQA benchmark comprising 977 WSIs and 14,879 question-answer pairs. CTIS-Bench emphasizes clinically grounded, closed-ended questions (e.g., tumor grade, receptor status) that reflect real diagnostic workflows, minimize non-visual reasoning, and require genuine slide understanding. We further propose CTIS-QA, a Slide-level Question Answering model, featuring a dual-stream architecture that mimics pathologists' diagnostic approach. One stream captures global slidelevel context via clustering-based feature aggregation, while the other focuses on salient local regions through attention-guided patch perception module. Extensive experiments on WSI-VQA, CTIS-Bench, and slide-level diagnostic tasks show that CTIS-QA consistently outperforms existing state-of-the-art models across multiple metrics. We will fully release both CTIS-Bench and CTIS-QA as open-source resources.

Problem

• slide-level question answering
• pathology
• vision-language alignment
• clinical diagnosis
• whole-slide image

Innovation

• Clinical Template
• Slide-level Question Answering
• Vision-Language Alignment
• Dual-stream Architecture
• Pathology Report Structuring