Author name cluster

Hong Bu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

AAAI Conference 2026 Conference Paper

Patho-AgenticRAG: Towards Multimodal Agentic Retrieval-Augmented Generation for Pathology VLMs via Reinforcement Learning

Wenchuan Zhang
Jingru Guo
Hengzhe Zhang
Penghao Zhang
Jie Chen
Shuwan Zhang
Zhang Zhang
Yuhao Yi

Although Vision Language Models (VLMs) have shown generalization in medical imaging, pathology presents unique challenges due to ultra-high resolution, complex tissue structures, and nuanced semantics. These factors make pathology VLMs prone to hallucinations, i.e., generating outputs inconsistent with visual evidence, which undermines clinical trust. Existing RAG approaches in this domain largely depend on text-based knowledge bases, limiting their ability to leverage diagnostic visual cues. To address this, we propose Patho-AgenticRAG, a multimodal RAG framework with a database built on page-level embeddings from authoritative pathology textbooks. Unlike traditional text-only retrieval systems, it supports joint text–image search, enabling retrieval of textbook pages that contain both the queried text and relevant visual cues, thus avoiding the loss of critical image-based information. Patho-AgenticRAG also supports reasoning, task decomposition, and multi-turn search interactions, improving accuracy in complex diagnostic scenarios. Experiments show that Patho-AgenticRAG significantly outperforms existing multimodal models in complex pathology tasks like multiple-choice diagnosis and visual question answering.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Patho-R1: A Multimodal Reinforcement Learning-Based Pathology Expert Reasoner

Wenchuan Zhang
Penghao Zhang
Jingru Guo
Tao Cheng
Jie Chen
Shuwan Zhang
Zhang Zhang
Yuhao Yi

Recent advances in vision-language models (VLMs) have enabled broad progress in the general medical field. However, pathology still remains a more challenging sub-domain, with current pathology-specific VLMs exhibiting limitations in both diagnostic accuracy and reasoning plausibility. Such shortcomings are largely attributable to the nature of current pathology datasets, which are primarily composed of image–description pairs that lack the depth and structured diagnostic paradigms employed by real-world pathologists. In this study, we leverage pathology textbooks and real-world pathology experts to construct high-quality, reasoning-oriented datasets. Building on this, we introduce Patho-R1, a multimodal RL-based pathology Reasoner, trained through a three-stage pipeline: (1) continued pretraining on 3.5 million image-text pairs for knowledge infusion; (2) supervised fine-tuning on 500k high-quality Chain-of-Thought samples for reasoning incentivizing; (3) reinforcement learning using Group Relative Policy Optimization and Decoupled Clip and Dynamic sAmpling Policy Optimization strategies for multimodal reasoning quality refinement. To further assess the alignment quality of our dataset, we propose Patho-CLIP, trained on the same figure-caption corpus used for continued pretraining. Comprehensive experimental results demonstrate that both Patho-CLIP and Patho-R1 achieve robust performance across a wide range of pathology-related tasks, including zero-shot classification, cross-modal retrieval, Visual Question Answering, and Multiple Choice Question.

PDF Details DOI

EAAI Journal 2024 Journal Article

Exponential distance transform maps for cell localization

Bo Li
Jie Chen
Hang Yi
Min Feng
Yongquan Yang
Qikui Zhu
Hong Bu

Cell localization in medical image analysis aims for precise identification of cell positions. Existing methods involve predicting density maps from images, followed by post-processing to extract cell location and number details. The quality of generated density maps significantly impacts the model’s localization and counting performance. However, density maps produced with Gaussian kernels exhibit stacking in dense regions, resulting in inaccurate cell location information and suboptimal localization performance. In this study, we propose an exponential distance transform map that ensures accurate location information and provides well-defined gradient details for effective model learning, setting a new benchmark for high performance. Additionally, to address the challenge of substantial variations in cell color within images, we introduce a multi-scale gradient aggregation module that enhances the model’s color recognition robustness through gradient information utilization. Experimental results across diverse datasets showcase notable improvements, establishing a novel benchmark for cell localization.

Details DOI

AAAI Conference 2024 Conference Paper

Point Transformer with Federated Learning for Predicting Breast Cancer HER2 Status from Hematoxylin and Eosin-Stained Whole Slide Images

Bao Li
Zhenyu Liu
Lizhi Shao
Bensheng Qiu
Hong Bu
Jie Tian

Directly predicting human epidermal growth factor receptor 2 (HER2) status from widely available hematoxylin and eosin (HE)-stained whole slide images (WSIs) can reduce technical costs and expedite treatment selection. Accurately predicting HER2 requires large collections of multi-site WSIs. Federated learning enables collaborative training of these WSIs without gigabyte-size WSIs transportation and data privacy concerns. However, federated learning encounters challenges in addressing label imbalance in multi-site WSIs from the real world. Moreover, existing WSI classification methods cannot simultaneously exploit local context information and long-range dependencies in the site-end feature representation of federated learning. To address these issues, we present a point transformer with federated learning for multi-site HER2 status prediction from HE-stained WSIs. Our approach incorporates two novel designs. We propose a dynamic label distribution strategy and an auxiliary classifier, which helps to establish a well-initialized model and mitigate label distribution variations across sites. Additionally, we propose a farthest cosine sampling based on cosine distance. It can sample the most distinctive features and capture the long-range dependencies. Extensive experiments and analysis show that our method achieves state-of-the-art performance at four sites with a total of 2687 WSIs. Furthermore, we demonstrate that our model can generalize to two unseen sites with 229 WSIs. Code is available at: https://github.com/boyden/PointTransformerFL

PDF Details DOI