LePaX enables high-resolution chest X-ray report generation by learning to allocate resolution to diagnostically relevant regions and fusing high-res patches back into global features without increasing token count.
hub
Gener- ating Radiology Reports via Memory-driven Transformer
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
abstract
Medical imaging is frequently used in clinical practice and trials for diagnosis and treatment. Writing imaging reports is time-consuming and can be error-prone for inexperienced radiologists. Therefore, automatically generating radiology reports is highly desired to lighten the workload of radiologists and accordingly promote clinical automation, which is an essential task to apply artificial intelligence to the medical domain. In this paper, we propose to generate radiology reports with memory-driven Transformer, where a relational memory is designed to record key information of the generation process and a memory-driven conditional layer normalization is applied to incorporating the memory into the decoder of Transformer. Experimental results on two prevailing radiology report datasets, IU X-Ray and MIMIC-CXR, show that our proposed approach outperforms previous models with respect to both language generation metrics and clinical evaluations. Particularly, this is the first work reporting the generation results on MIMIC-CXR to the best of our knowledge. Further analyses also demonstrate that our approach is able to generate long reports with necessary medical terms as well as meaningful image-text attention mappings.
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
LLMs generate deterministic code to convert scar images into low-dimensional clinical features for classification, claimed to outperform end-to-end deep learning when training data is scarce.
MedScribe reformulates CT radiology reporting as an agentic evidence-acquisition workflow using LLM-invoked diagnostic tools and pathology-aligned retrieval, yielding higher clinical accuracy and consistency than standard VLMs on CT-RATE and RadChestCT.
RIHA proposes a hierarchical alignment transformer that uses multi-scale visual and textual feature pyramids plus optimal transport to generate more accurate radiology reports from medical images.
ESC-RL improves RL for radiology reports via group-wise evidence-aware rewards (GEAR) and LLM-driven self-correcting preference learning (SPL), reaching state-of-the-art on two chest X-ray datasets.
Transferring a 2D MLLM to 3D CT inputs via parameter reuse, a Text-Guided Hierarchical MoE framework, and two-stage training yields better performance than prior 3D medical MLLMs on medical report generation and visual question answering.
SemEnrich enriches radiology reports with positive/neutral findings via self-supervised semantic clustering, yielding average gains of 5-7% on COMET, BERT score, Sentence BLEU, CheXbert-F1 and RadGraph-F1 after fine-tuning, plus further gains when cluster info is added to GRPO rewards.
View-PNDF detects and selectively fine-tunes view-specific neurons for consistent multi-view chest X-ray report generation, followed by LLM consolidation of reports.
Reinforcement learning with a tunable control parameter and clinical reward enables precision-recall controllable radiology report generation that outperforms prior methods on MIMIC-CXR.
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.
HTSC-CIF applies hierarchical task decomposition and cross-modal causal intervention to generate medical reports from images while addressing domain knowledge, alignment, and bias challenges.
M4CXR is a multi-modal large language model that performs multiple tasks in chest X-ray analysis including report generation with claimed SOTA clinical accuracy using chain-of-thought prompting.
SCOUT achieves state-of-the-art BLEU-1 to BLEU-4 and METEOR scores on TCGA-BRCA, MICCAI REG, and HistAI by fusing local histology, slide-level context, and semantic concepts in a context-aware transformer.
Radiologists provided input on explainable ML needs and beneficial clinical tasks, resulting in guidelines for developing clinically aligned models in radiology.
RL-ACRGNet applies reinforcement learning to an encoder-decoder architecture for radiology report generation and reports modest metric gains on IU-Xray and MIMIC-CXR.
LLaMA-XR fine-tunes LLaMA 3.1 with QLoRA on DenseNet-121 embeddings to generate radiology reports from chest X-rays, reporting ROUGE-L of 0.433 and METEOR of 0.336 on the IU X-ray benchmark.
CA-TriNet combines co-attention transformers with a triple-LSTM module for medical report generation and reports outperforming prior models on three public datasets.
citing papers explorer
-
Seeing What Matters: Lesion-Aware High-Resolution Patch Discovery and Fusion for Chest X-ray Report Generation
LePaX enables high-resolution chest X-ray report generation by learning to allocate resolution to diagnostically relevant regions and fusing high-res patches back into global features without increasing token count.
-
When LLMs Analyze Scars: From Images to Clinically-Meaningful Features
LLMs generate deterministic code to convert scar images into low-dimensional clinical features for classification, claimed to outperform end-to-end deep learning when training data is scarce.
-
MedScribe: Clinically Grounded CT Reporting through Agentic Workflows
MedScribe reformulates CT radiology reporting as an agentic evidence-acquisition workflow using LLM-invoked diagnostic tools and pathology-aligned retrieval, yielding higher clinical accuracy and consistency than standard VLMs on CT-RATE and RadChestCT.
-
RIHA: Report-Image Hierarchical Alignment for Radiology Report Generation
RIHA proposes a hierarchical alignment transformer that uses multi-scale visual and textual feature pyramids plus optimal transport to generate more accurate radiology reports from medical images.
-
Enhancing Reinforcement Learning for Radiology Report Generation with Evidence-aware Rewards and Self-correcting Preference Learning
ESC-RL improves RL for radiology reports via group-wise evidence-aware rewards (GEAR) and LLM-driven self-correcting preference learning (SPL), reaching state-of-the-art on two chest X-ray datasets.
-
Adapting 2D Multi-Modal Large Language Model for 3D CT Image Analysis
Transferring a 2D MLLM to 3D CT inputs via parameter reuse, a Text-Guided Hierarchical MoE framework, and two-stage training yields better performance than prior 3D medical MLLMs on medical report generation and visual question answering.
-
SemEnrich: Self-Supervised Semantic Enrichment of Radiology Reports for Vision-Language Learning
SemEnrich enriches radiology reports with positive/neutral findings via self-supervised semantic clustering, yielding average gains of 5-7% on COMET, BERT score, Sentence BLEU, CheXbert-F1 and RadGraph-F1 after fine-tuning, plus further gains when cluster info is added to GRPO rewards.
-
Seeing Through Multiple Views: Parameter-Efficient Fine-Tuning via Selective Neurons for Consistent Radiology Report Generation
View-PNDF detects and selectively fine-tunes view-specific neurons for consistent multi-view chest X-ray report generation, followed by LLM consolidation of reports.
-
Precision Recall Controllable Radiology Report Generation via Hybrid Natural Language and Clinical Reward Learning
Reinforcement learning with a tunable control parameter and clinical reward enables precision-recall controllable radiology report generation that outperforms prior methods on MIMIC-CXR.
-
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.
-
Medical Report Generation: A Hierarchical Task Structure-Based Cross-Modal Causal Intervention Framework
HTSC-CIF applies hierarchical task decomposition and cross-modal causal intervention to generate medical reports from images while addressing domain knowledge, alignment, and bias challenges.
-
M4CXR: Exploring Multi-task Potentials of Multi-modal Large Language Models for Chest X-ray Interpretation
M4CXR is a multi-modal large language model that performs multiple tasks in chest X-ray analysis including report generation with claimed SOTA clinical accuracy using chain-of-thought prompting.
-
Semantic Context-aware mOdality fUsion Transformer (SCOUT): A Context-Aware Multimodal Transformer for Concept-Grounded Pathology Report Generation
SCOUT achieves state-of-the-art BLEU-1 to BLEU-4 and METEOR scores on TCGA-BRCA, MICCAI REG, and HistAI by fusing local histology, slide-level context, and semantic concepts in a context-aware transformer.
-
Exploring Radiologists' Expectations of Explainable Machine Learning Models in Medical Image Analysis
Radiologists provided input on explainable ML needs and beneficial clinical tasks, resulting in guidelines for developing clinically aligned models in radiology.
-
RL-ACRGNet: Reinforcement Learning-Based Chest Radiology Report Generation Network
RL-ACRGNet applies reinforcement learning to an encoder-decoder architecture for radiology report generation and reports modest metric gains on IU-Xray and MIMIC-CXR.
-
LLaMA-XR: A Novel Framework for Radiology Report Generation using LLaMA and QLoRA Fine Tuning
LLaMA-XR fine-tunes LLaMA 3.1 with QLoRA on DenseNet-121 embeddings to generate radiology reports from chest X-rays, reporting ROUGE-L of 0.433 and METEOR of 0.336 on the IU X-ray benchmark.
-
Image-to-Text for Medical Reports Using Adaptive Co-Attention and Triple-LSTM Module
CA-TriNet combines co-attention transformers with a triple-LSTM module for medical report generation and reports outperforming prior models on three public datasets.