PBSBench: A Multi-Level Vision-Language Framework and Benchmark for Hematopathology Whole Slide Image Interpretation

TL;DR

This paper introduces PBSBench, a comprehensive vision-language framework and benchmark tailored for interpreting hematopathology whole slide images, focusing on peripheral blood smears, with new datasets, models, and evaluation tasks.

Contribution

The authors create PBSInstr, a novel PBS-specific dataset, and develop PBS-VL, a specialized vision-language model, along with PBSBench, a benchmark for multi-level PBS interpretation.

Findings

PBS-VL outperforms existing models on PBSBench tasks.

The dataset includes 353 PBS WSIs, 29k cell crops, and 27k QA pairs.

The framework supports decision-making in hematopathology.

Abstract

Peripheral Blood Smear (PBS) is a critical microscopic examination in hematopathology that yields whole-slide imaging (WSI). Unlike solid tissue pathology, PBS interpretation focuses on individual cell morphologies rather than tissue architecture, making it distinct in both visual characteristics and diagnostic reasoning. However, current multimodal large language models (MLLMs) for pathology are primarily developed on solid-tissue WSIs and struggle to generalize to PBS. To bridge this gap, we construct PBSInstr, the first vision-language dataset for PBS interpretation, comprising 353 PBS WSIs paired with microscopic impression paragraphs and 29k cell-level image crops annotated with cell type labels and morphological descriptions. To facilitate instruction tuning, PBSInstr further includes 27k question-answer (QA) pairs for cell crops and 1,286 QA pairs for PBS slides. Building upon PBSInstr, we develop PBS-VL, a hematopathology-tailored vision-language model for multi-level PBS interpretation at both cell and slide levels. To comprehensively evaluate PBS understanding, we construct PBSBench, a visual question answering (VQA) benchmark featuring four question categories and six PBS interpretation tasks. Experiments show that PBS-VL outperforms existing general-purpose and pathology MLLMs, underscoring the value of PBS-specific data. We release our code, datasets, and model weights to facilitate future research. Our proposed framework lays the foundation for developing practical AI assistants supporting decision-making in hematopathology.