Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gan, Lubin, Zhang, Jing, Qu, Linhao, Wang, Yijun, Wu, Siying, Sun, Xiaoyan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2508.01602
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909724503441408
author Gan, Lubin
Zhang, Jing
Qu, Linhao
Wang, Yijun
Wu, Siying
Sun, Xiaoyan
author_facet Gan, Lubin
Zhang, Jing
Qu, Linhao
Wang, Yijun
Wu, Siying
Sun, Xiaoyan
contents The fine-grained classification of brain tumor subtypes from histopathological whole slide images is highly challenging due to subtle morphological variations and the scarcity of annotated data. Although vision-language models have enabled promising zero-shot classification, their ability to capture fine-grained pathological features remains limited, resulting in suboptimal subtype discrimination. To address these challenges, we propose the Fine-Grained Patch Alignment Network (FG-PAN), a novel zero-shot framework tailored for digital pathology. FG-PAN consists of two key modules: (1) a local feature refinement module that enhances patch-level visual features by modeling spatial relationships among representative patches, and (2) a fine-grained text description generation module that leverages large language models to produce pathology-aware, class-specific semantic prototypes. By aligning refined visual features with LLM-generated fine-grained descriptions, FG-PAN effectively increases class separability in both visual and semantic spaces. Extensive experiments on multiple public pathology datasets, including EBRAINS and TCGA, demonstrate that FG-PAN achieves state-of-the-art performance and robust generalization in zero-shot brain tumor subtype classification.
format Preprint
id arxiv_https___arxiv_org_abs_2508_01602
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhancing Zero-Shot Brain Tumor Subtype Classification via Fine-Grained Patch-Text Alignment
Gan, Lubin
Zhang, Jing
Qu, Linhao
Wang, Yijun
Wu, Siying
Sun, Xiaoyan
Computer Vision and Pattern Recognition
The fine-grained classification of brain tumor subtypes from histopathological whole slide images is highly challenging due to subtle morphological variations and the scarcity of annotated data. Although vision-language models have enabled promising zero-shot classification, their ability to capture fine-grained pathological features remains limited, resulting in suboptimal subtype discrimination. To address these challenges, we propose the Fine-Grained Patch Alignment Network (FG-PAN), a novel zero-shot framework tailored for digital pathology. FG-PAN consists of two key modules: (1) a local feature refinement module that enhances patch-level visual features by modeling spatial relationships among representative patches, and (2) a fine-grained text description generation module that leverages large language models to produce pathology-aware, class-specific semantic prototypes. By aligning refined visual features with LLM-generated fine-grained descriptions, FG-PAN effectively increases class separability in both visual and semantic spaces. Extensive experiments on multiple public pathology datasets, including EBRAINS and TCGA, demonstrate that FG-PAN achieves state-of-the-art performance and robust generalization in zero-shot brain tumor subtype classification.
title Enhancing Zero-Shot Brain Tumor Subtype Classification via Fine-Grained Patch-Text Alignment
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.01602