Xiaofan Zhang

Qianhan Feng, Zhongzhen Huang, Yakun Zhu et al.

Vision-Language Models (VLMs) show promise in medical diagnosis, yet suffer from reasoning detachment, where linguistically fluent explanations drift from verifiable image evidence, undermining clinical trust. Recent multi-agent frameworks simulate Multidisciplinary Team (MDT) debates to mitigate single-model bias, but open-ended discussions amplify textual noise and computational cost while failing to anchor reasoning to visual evidence, the cornerstone of medical decision-making. We propose UCAgents, a hierarchical multi-agent framework enforcing unidirectional convergence through structured evidence auditing. Inspired by clinical workflows, UCAgents forbids position changes and limits agent interactions to targeted evidence verification, suppressing rhetorical drift while amplifying visual signal extraction. In UCAgents, a one-round inquiry discussion is introduced to uncover potential risks of visual-textual misalignment. This design jointly constrains visual ambiguity and textual noise, a dual-noise bottleneck that we formalize via information theory. Extensive experiments on four medical VQA benchmarks show UCAgents achieves superior accuracy (71.3% on PathVQA, +6.0% over state-of-the-art) with 87.7% lower token cost, the evaluation results further confirm that UCAgents strikes a balance between uncovering more visual evidence and avoiding confusing textual interference. These results demonstrate that UCAgents exhibits both diagnostic reliability and computational efficiency critical for real-world clinical deployment. Code is available at https://github.com/fqhank/UCAgents.

Yakun Zhu, Zhongzhen Huang, Linjie Mu et al.

The emergence of groundbreaking large language models capable of performing complex reasoning tasks holds significant promise for addressing various scientific challenges, including those arising in complex clinical scenarios. To enable their safe and effective deployment in real-world healthcare settings, it is urgently necessary to benchmark the diagnostic capabilities of current models systematically. Given the limitations of existing medical benchmarks in evaluating advanced diagnostic reasoning, we present DiagnosisArena, a comprehensive and challenging benchmark designed to rigorously assess professional-level diagnostic competence. DiagnosisArena consists of 1,113 pairs of segmented patient cases and corresponding diagnoses, spanning 28 medical specialties, deriving from clinical case reports published in 10 top-tier medical journals. The benchmark is developed through a meticulous construction pipeline, involving multiple rounds of screening and review by both AI systems and human experts, with thorough checks conducted to prevent data leakage. Our study reveals that even the most advanced reasoning models, o3, o1, and DeepSeek-R1, achieve only 51.12%, 31.09%, and 17.79% accuracy, respectively. This finding highlights a significant generalization bottleneck in current large language models when faced with clinical diagnostic reasoning challenges. Through DiagnosisArena, we aim to drive further advancements in AI's diagnostic reasoning capabilities, enabling more effective solutions for real-world clinical diagnostic challenges. We provide the benchmark and evaluation tools for further research and development https://github.com/SPIRAL-MED/DiagnosisArena.

Linjie Mu, Zhongzhen Huang, Shengqian Qin et al.

Large vision-language models (LVLMs) have shown great promise in medical applications, particularly in visual question answering (MedVQA) and diagnosis from medical images. However, existing datasets and models often fail to consider critical aspects of medical diagnostics, such as the integration of historical records and the analysis of disease progression over time. In this paper, we introduce MMXU (Multimodal and MultiX-ray Understanding), a novel dataset for MedVQA that focuses on identifying changes in specific regions between two patient visits. Unlike previous datasets that primarily address single-image questions, MMXU enables multi-image questions, incorporating both current and historical patient data. We demonstrate the limitations of current LVLMs in identifying disease progression on MMXU-\textit{test}, even those that perform well on traditional benchmarks. To address this, we propose a MedRecord-Augmented Generation (MAG) approach, incorporating both global and regional historical records. Our experiments show that integrating historical records significantly enhances diagnostic accuracy by at least 20\%, bridging the gap between current LVLMs and human expert performance. Additionally, we fine-tune models with MAG on MMXU-\textit{dev}, which demonstrates notable improvements. We hope this work could illuminate the avenue of advancing the use of LVLMs in medical diagnostics by emphasizing the importance of historical context in interpreting medical images. Our dataset is released at github: https://github.com/linjiemu/MMXU.

Yakun Zhu, Zhongzhen Huang, Qianhan Feng et al.

Medical care follows complex clinical pathways that extend beyond isolated physician-patient encounters, emphasizing decision-making and transitions between different stages. Current benchmarks focusing on static exams or isolated dialogues inadequately evaluate large language models (LLMs) in dynamic clinical scenarios. We introduce CP-Env, a controllable agentic hospital environment designed to evaluate LLMs across end-to-end clinical pathways. CP-Env simulates a hospital ecosystem with patient and physician agents, constructing scenarios ranging from triage and specialist consultation to diagnostic testing and multidisciplinary team meetings for agent interaction. Following real hospital adaptive flow of healthcare, it enables branching, long-horizon task execution. We propose a three-tiered evaluation framework encompassing Clinical Efficacy, Process Competency, and Professional Ethics. Results reveal that most models struggle with pathway complexity, exhibiting hallucinations and losing critical diagnostic details. Interestingly, excessive reasoning steps can sometimes prove counterproductive, while top models tend to exhibit reduced tool dependency through internalized knowledge. CP-Env advances medical AI agents development through comprehensive end-to-end clinical evaluation. We provide the benchmark and evaluation tools for further research and development at https://github.com/SPIRAL-MED/CP_ENV.

Wenhui Lei, Anqi Li, Yusheng Tan et al.

Foundation Models (FMs) in computational pathology (CPath) have significantly advanced the extraction of meaningful features from histopathology image datasets, achieving strong performance across various clinical tasks. Despite their impressive performance, these models often exhibit variability when applied to different tasks, prompting the need for a unified framework capable of consistently excelling across various applications. In this work, we propose Shazam, a novel framework designed to efficiently combine multiple CPath models. Unlike previous approaches that train a fixed-parameter FM, Shazam dynamically extracts and refines information from diverse FMs for each specific task. To ensure that each FM contributes effectively without dominance, a novel distillation strategy is applied, guiding the student model with features from all teacher models, which enhances its generalization ability. Experimental results on two pathology patch classification datasets demonstrate that Shazam outperforms existing CPath models and other fusion methods. Its lightweight, flexible design makes it a promising solution for improving CPath analysis in real-world settings. Code will be available at https://github.com/Tuner12/Shazam.

Zhongzhen Huang, Linjie Mu, Yakun Zhu et al.

Effective clinical decision-making depends on iterative, multimodal reasoning across diverse sources of evidence. The recent emergence of multimodal reasoning models has significantly transformed the landscape of solving complex tasks. Although such models have achieved notable success in mathematics and science, their application to medical domains remains underexplored. In this work, we propose \textit{MedE$^2$}, a two-stage post-training pipeline that elicits and then enhances multimodal reasoning for medical domains. In Stage-I, we fine-tune models using 2,000 text-only data samples containing precisely orchestrated reasoning demonstrations to elicit reasoning behaviors. In Stage-II, we further enhance the model's reasoning capabilities using 1,500 rigorously curated multimodal medical cases, aligning model reasoning outputs with our proposed multimodal medical reasoning preference. Extensive experiments demonstrate the efficacy and reliability of \textit{MedE$^2$} in improving the reasoning performance of medical multimodal models. Notably, models trained with \textit{MedE$^2$} consistently outperform baselines across multiple medical multimodal benchmarks. Additional validation on larger models and under inference-time scaling further confirms the robustness and practical utility of our approach.

Mengya Xu, Zhongzhen Huang, Jie Zhang et al.

In robot-assisted minimally invasive surgery, we introduce the Surgical Action Planning (SAP) task, which generates future action plans from visual inputs to address the absence of intraoperative predictive planning in current intelligent applications. SAP shows great potential for enhancing intraoperative guidance and automating procedures. However, it faces challenges such as understanding instrument-action relationships and tracking surgical progress. Large Language Models (LLMs) show promise in understanding surgical video content but remain underexplored for predictive decision-making in SAP, as they focus mainly on retrospective analysis. Challenges like data privacy, computational demands, and modality-specific constraints further highlight significant research gaps. To tackle these challenges, we introduce LLM-SAP, a Large Language Models-based Surgical Action Planning framework that predicts future actions and generates text responses by interpreting natural language prompts of surgical goals. The text responses potentially support surgical education, intraoperative decision-making, procedure documentation, and skill analysis. LLM-SAP integrates two novel modules: the Near-History Focus Memory Module (NHF-MM) for modeling historical states and the prompts factory for action planning. We evaluate LLM-SAP on our constructed CholecT50-SAP dataset using models like Qwen2.5 and Qwen2-VL, demonstrating its effectiveness in next-action prediction. Pre-trained LLMs are tested in a zero-shot setting, and supervised fine-tuning (SFT) with LoRA is implemented. Our experiments show that Qwen2.5-72B-SFT surpasses Qwen2.5-72B with a 19.3% higher accuracy.

Mengya Xu, Zhongzhen Huang, Dillan Imans et al.

Effective evaluation is critical for driving advancements in MLLM research. The surgical action planning (SAP) task, which aims to generate future action sequences from visual inputs, demands precise and sophisticated analytical capabilities. Unlike mathematical reasoning, surgical decision-making operates in life-critical domains and requires meticulous, verifiable processes to ensure reliability and patient safety. This task demands the ability to distinguish between atomic visual actions and coordinate complex, long-horizon procedures, capabilities that are inadequately evaluated by current benchmarks. To address this gap, we introduce SAP-Bench, a large-scale, high-quality dataset designed to enable multimodal large language models (MLLMs) to perform interpretable surgical action planning. Our SAP-Bench benchmark, derived from the cholecystectomy procedures context with the mean duration of 1137.5s, and introduces temporally-grounded surgical action annotations, comprising the 1,226 clinically validated action clips (mean duration: 68.7s) capturing five fundamental surgical actions across 74 procedures. The dataset provides 1,152 strategically sampled current frames, each paired with the corresponding next action as multimodal analysis anchors. We propose the MLLM-SAP framework that leverages MLLMs to generate next action recommendations from the current surgical scene and natural language instructions, enhanced with injected surgical domain knowledge. To assess our dataset's effectiveness and the broader capabilities of current models, we evaluate seven state-of-the-art MLLMs (e.g., OpenAI-o1, GPT-4o, QwenVL2.5-72B, Claude-3.5-Sonnet, GeminiPro2.5, Step-1o, and GLM-4v) and reveal critical gaps in next action prediction performance.

Mianxin Liu, Jinru Ding, Jie Xu et al.

Ensuring the general efficacy and goodness for human beings from medical large language models (LLM) before real-world deployment is crucial. However, a widely accepted and accessible evaluation process for medical LLM, especially in the Chinese context, remains to be established. In this work, we introduce "MedBench", a comprehensive, standardized, and reliable benchmarking system for Chinese medical LLM. First, MedBench assembles the currently largest evaluation dataset (300,901 questions) to cover 43 clinical specialties and performs multi-facet evaluation on medical LLM. Second, MedBench provides a standardized and fully automatic cloud-based evaluation infrastructure, with physical separations for question and ground truth. Third, MedBench implements dynamic evaluation mechanisms to prevent shortcut learning and answer remembering. Applying MedBench to popular general and medical LLMs, we observe unbiased, reproducible evaluation results largely aligning with medical professionals' perspectives. This study establishes a significant foundation for preparing the practical applications of Chinese medical LLMs. MedBench is publicly accessible at https://medbench.opencompass.org.cn.

Zhongzhen Huang, Linda Wei, Shaoting Zhang et al.

Combining images from multi-modalities is beneficial to explore various information in computer vision, especially in the medical domain. As an essential part of clinical diagnosis, multi-modal brain tumor segmentation aims to delineate the malignant entity involving multiple modalities. Although existing methods have shown remarkable performance in the task, the information exchange for cross-scale and high-level representations fusion in spatial and modality are limited in these methods. In this paper, we present a novel Modality Aware and Shift Mixer that integrates intra-modality and inter-modality dependencies of multi-modal images for effective and robust brain tumor segmentation. Specifically, we introduce a Modality-Aware module according to neuroimaging studies for modeling the specific modality pair relationships at low levels, and a Modality-Shift module with specific mosaic patterns is developed to explore the complex relationships across modalities at high levels via the self-attention. Experimentally, we outperform previous state-of-the-art approaches on the public Brain Tumor Segmentation (BraTS 2021 segmentation) dataset. Further qualitative experiments demonstrate the efficacy and robustness of MASM.