DeepTumorVQA is a new stage-wise 3D CT VQA benchmark showing that quantitative measurement is the main failure point for current medical VLMs and that tool augmentation substantially improves later reasoning stages.
hub
arXiv preprint arXiv:2406.19280 , year=
26 Pith papers cite this work, alongside 8 external citations. Polarity classification is still indexing.
26
Pith papers citing it
8
external citations · Pith
abstract
The rapid development of multimodal large language models (MLLMs), such as GPT-4V, has led to significant advancements. However, these models still face challenges in medical multimodal capabilities due to limitations in the quantity and quality of medical vision-text data, stemming from data privacy concerns and high annotation costs. While pioneering approaches utilize PubMed's large-scale, de-identified medical image-text pairs to address these limitations, they still fall short due to inherent data noise. To tackle this, we refined medical image-text pairs from PubMed and employed MLLMs (GPT-4V) in an 'unblinded' capacity to denoise and reformat the data, resulting in the creation of the PubMedVision dataset with 1.3 million medical VQA samples. Our validation demonstrates that: (1) PubMedVision can significantly enhance the medical multimodal capabilities of current MLLMs, showing significant improvement in benchmarks including the MMMU Health & Medicine track; (2) manual checks by medical experts and empirical results validate the superior data quality of our dataset compared to other data construction methods. Using PubMedVision, we train a 34B medical MLLM HuatuoGPT-Vision, which shows superior performance in medical multimodal scenarios among open-source MLLMs.
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
MedHorizon benchmark reveals current multimodal LLMs achieve only 41.1% accuracy on long medical videos due to failures in sparse evidence retrieval and procedural reasoning.
MedFlowBench evaluates VLM agents on full radiology and pathology studies by requiring both task answers and verifiable evidence like key slices and regions of interest, revealing that answer-only scores overestimate performance.
A large examination-level ultrasound dataset with long-form reports enables simple LVLM fine-tuning to outperform prior complex methods.
DermAgent orchestrates seven vision-language tools in a Plan-Execute-Reflect loop with dual-modality retrieval from 413k cases and a critic module to outperform GPT-4o by 17.6% in zero-shot dermatological diagnosis accuracy.
X-PCR is a new benchmark of 26,415 images and 177,868 expert VQA pairs that evaluates MLLMs on six-stage progressive reasoning and cross-modality integration in ophthalmology.
MEDSYN benchmark shows MLLMs match experts on differential diagnosis lists but have much larger gaps to final diagnosis selection than humans, due to text overreliance and cross-modal evidence gaps.
IBISAgent enables MLLMs to perform iterative pixel-level visual reasoning for biomedical object referring and segmentation via text-based clicks and agentic RL, outperforming prior SOTA methods without model modifications.
M3Bench is a clinically grounded benchmark showing that gradient-based VLM editors generalize but break locality, while memory-based editors preserve locality but fail on composition and temporal tasks, with failures traced to anisotropic latent representations.
ViToS uses dual-stream RL with cross-feedback optimization to prune medical image tokens to 77% length while reporting 108.27% and 104.16% relative performance on two 7B VLMs across seven benchmarks.
VLMs for synthetic medical image detection overweight text metadata, flipping authenticity judgments on the same image and dropping accuracy on authentic images by 61.1% on average when an explicit AI-origin tag is present.
ASAP introduces an anatomy-aware semantically-adaptive pre-training method for medical volumetric vision-language models and reports state-of-the-art results on a new benchmark spanning 15 datasets and 22 tasks.
DIYHealth Suite introduces a large home-care dataset, DIYHealthGPT model with Hybrid Hyper Low-Rank Adaptation, and DIYHealthBench, claiming SOTA results on 11 tasks over general and medical baselines.
GazeX uses radiologist gaze trajectories as a behavioral prior during pretraining to generate more accurate and expert-consistent results in chest X-ray report generation, disease grounding, and visual question answering.
MedRCube is a new fine-grained evaluation framework that benchmarks 33 MLLMs on medical imaging, ranks Lingshu-32B highest, and finds a significant positive link between shortcut behaviors and diagnostic performance.
HeartcareGPT proposes Dual Stream Projection Alignment (DSPA) on a structure-aware tokenizer for unified ECG signal-image modeling, supported by Heartcare-400K dataset and Heartcare-Bench.
HuatuoGPT-o1 achieves superior medical complex reasoning by using a verifier to curate reasoning trajectories for fine-tuning and then applying RL with verifier-based rewards.
Releases open medical LVLMs trained on a quality-filtered multimodal dataset, introduces CareQA-Vision benchmark from exams, reports performance gains over baselines, and flags adversarial vulnerabilities.
Cleaning and rebalancing biomedical figure-text data improves continued pretraining of a 4B medical vision-language model over raw BIOMEDICA, though headline gains in the abstract are not reproduced by the paper's tables.
VITAL adds visual-semantic dual supervision during training of medical MLLMs for latent reasoning, yielding SOTA results on 7 benchmarks with a new 61K multi-modality dataset while enabling post-hoc textual and visual explanations at zero inference overhead.
TIF-GRPO uses integral feedback on pseudo-temporal trajectories to regulate anatomy-aware rewards in RL for clinical faithfulness in volumetric CT analysis.
RoiMAM integrates a training-free ROI Generation Module with Semantic Selective Suppression and a Text Prompt Enhancer to produce a compact VLM that reports 2 percent and 4.6 percent accuracy gains on SLAKE and PMC-VQA at less than 20 percent the size of MedVInT-TD.
Single-agent LLM frameworks outperform naive multi-agent systems in multimodal clinical risk prediction tasks and are better calibrated.
The UPDP pipeline filters privacy terms and generates de-identified radiology images that preserve diagnostic pathology information, enabling models with competitive disease detection accuracy but reduced identity leakage and improved cross-hospital performance.
citing papers explorer
-
DeepTumorVQA: A Hierarchical 3D CT Benchmark for Stage-Wise Evaluation of Medical VLMs and Tool-Augmented Agents
DeepTumorVQA is a new stage-wise 3D CT VQA benchmark showing that quantitative measurement is the main failure point for current medical VLMs and that tool augmentation substantially improves later reasoning stages.
-
MedHorizon: Towards Long-context Medical Video Understanding in the Wild
MedHorizon benchmark reveals current multimodal LLMs achieve only 41.1% accuracy on long medical videos due to failures in sparse evidence retrieval and procedural reasoning.
-
MedOpenClaw and MedFlowBench: Auditing Medical Agents in Full-Study Workflows
MedFlowBench evaluates VLM agents on full radiology and pathology studies by requiring both task answers and verifiable evidence like key slices and regions of interest, revealing that answer-only scores overestimate performance.
-
Towards Real-World Ultrasound Understanding: Large Vision-Language Models from Multi-Image Examinations with Long-Form Reports
A large examination-level ultrasound dataset with long-form reports enables simple LVLM fine-tuning to outperform prior complex methods.
-
DermAgent: A Self-Reflective Agentic System for Dermatological Image Analysis with Multi-Tool Reasoning and Traceable Decision-Making
DermAgent orchestrates seven vision-language tools in a Plan-Execute-Reflect loop with dual-modality retrieval from 413k cases and a critic module to outperform GPT-4o by 17.6% in zero-shot dermatological diagnosis accuracy.
-
X-PCR: A Benchmark for Cross-modality Progressive Clinical Reasoning in Ophthalmic Diagnosis
X-PCR is a new benchmark of 26,415 images and 177,868 expert VQA pairs that evaluates MLLMs on six-stage progressive reasoning and cross-modality integration in ophthalmology.
-
MEDSYN: Benchmarking Multi-EviDence SYNthesis in Complex Clinical Cases for Multimodal Large Language Models
MEDSYN benchmark shows MLLMs match experts on differential diagnosis lists but have much larger gaps to final diagnosis selection than humans, due to text overreliance and cross-modal evidence gaps.
-
IBISAgent: Reinforcing Pixel-Level Visual Reasoning in MLLMs for Universal Biomedical Object Referring and Segmentation
IBISAgent enables MLLMs to perform iterative pixel-level visual reasoning for biomedical object referring and segmentation via text-based clicks and agentic RL, outperforming prior SOTA methods without model modifications.
-
Evaluating and Understanding Model Editing for Medical Vision Language Models
M3Bench is a clinically grounded benchmark showing that gradient-based VLM editors generalize but break locality, while memory-based editors preserve locality but fail on composition and temporal tasks, with failures traced to anisotropic latent representations.
-
Token-Sparse Medical Multimodal Reasoning via Dual-Stream Reinforcement Learning
ViToS uses dual-stream RL with cross-feedback optimization to prune medical image tokens to 77% length while reporting 108.27% and 104.16% relative performance on two 7B VLMs across seven benchmarks.
-
Text Over Image: Auditing Multimodal Robustness in Synthetic Medical Image Detection
VLMs for synthetic medical image detection overweight text metadata, flipping authenticity judgments on the same image and dropping accuracy on authentic images by 61.1% on average when an explicit AI-origin tag is present.
-
ASAP: Advancing Medical Volumetric Representation Learning with Anatomy-aware Semantically-adaptive Pre-training
ASAP introduces an anatomy-aware semantically-adaptive pre-training method for medical volumetric vision-language models and reports state-of-the-art results on a new benchmark spanning 15 datasets and 22 tasks.
-
DIYHealth Suite: Dataset, Model, and Benchmark for Health Management at Home
DIYHealth Suite introduces a large home-care dataset, DIYHealthGPT model with Hybrid Hyper Low-Rank Adaptation, and DIYHealthBench, claiming SOTA results on 11 tasks over general and medical baselines.
-
Seeing Through Experts Eyes A Foundational Vision Language Model Trained on Radiologists Gaze and Reasoning
GazeX uses radiologist gaze trajectories as a behavioral prior during pretraining to generate more accurate and expert-consistent results in chest X-ray report generation, disease grounding, and visual question answering.
-
MedRCube: A Multidimensional Framework for Fine-Grained and In-Depth Evaluation of MLLMs in Medical Imaging
MedRCube is a new fine-grained evaluation framework that benchmarks 33 MLLMs on medical imaging, ranks Lingshu-32B highest, and finds a significant positive link between shortcut behaviors and diagnostic performance.
-
HeartcareGPT: A Unified Multimodal ECG Suite for Dual Signal-Image Modeling and Understanding
HeartcareGPT proposes Dual Stream Projection Alignment (DSPA) on a structure-aware tokenizer for unified ECG signal-image modeling, supported by Heartcare-400K dataset and Heartcare-Bench.
-
HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs
HuatuoGPT-o1 achieves superior medical complex reasoning by using a verifier to curate reasoning trajectories for fine-tuning and then applying RL with verifier-based rewards.
-
Aloe-Vision: Robust Vision-Language Models for Healthcare
Releases open medical LVLMs trained on a quality-filtered multimodal dataset, introduces CareQA-Vision benchmark from exams, reports performance gains over baselines, and flags adversarial vulnerabilities.
-
Beyond Captions: Context-Grounded Reconstruction for Biomedical Multimodal Continued Pretraining
Cleaning and rebalancing biomedical figure-text data improves continued pretraining of a 4B medical vision-language model over raw BIOMEDICA, though headline gains in the abstract are not reproduced by the paper's tables.
-
VITAL: Visual-Semantic Dual Supervision for Enhanced and Interpretable Latent Reasoning in Medical MLLMs
VITAL adds visual-semantic dual supervision during training of medical MLLMs for latent reasoning, yielding SOTA results on 7 benchmarks with a new 61K multi-modality dataset while enabling post-hoc textual and visual explanations at zero inference overhead.
-
Regulating Anatomy-Aware Rewards via Trajectory-Integral Feedback for Volumetric Computed Tomography Analysis
TIF-GRPO uses integral feedback on pseudo-temporal trajectories to regulate anatomy-aware rewards in RL for clinical faithfulness in volumetric CT analysis.
-
RoiMAM: Region-of-Interest Medical Attention Model for Efficient Vision-Language Understanding
RoiMAM integrates a training-free ROI Generation Module with Semantic Selective Suppression and a Text Prompt Enhancer to produce a compact VLM that reports 2 percent and 4.6 percent accuracy gains on SLAKE and PMC-VQA at less than 20 percent the size of MedVInT-TD.
-
AgentRx: A Benchmark Study of LLM Agents for Multimodal Clinical Prediction Tasks
Single-agent LLM frameworks outperform naive multi-agent systems in multimodal clinical risk prediction tasks and are better calibrated.
-
A Utility-preserving De-identification Pipeline for Cross-hospital Radiology Data Sharing
The UPDP pipeline filters privacy terms and generates de-identified radiology images that preserve diagnostic pathology information, enabling models with competitive disease detection accuracy but reduced identity leakage and improved cross-hospital performance.
-
Thought Graph Traversal for Test-time Scaling in Chest X-ray VLLMs
A new prompting framework called Thought Graph Traversal combined with reasoning budget forcing improves test-time performance of frozen chest X-ray VLLMs on report generation benchmarks.
-
UniReason-Med: A Shared Grounded Reasoning Interface for 2D-to-3D Transfer in Medical VQA
UniReason-Med introduces a unified framework for 2D and 3D medical VQA with shared grounded reasoning, trained on a 220K dataset, claiming that joint 2D+3D supervision improves 3D performance over 3D-only training.