Diffusion models have emerged as powerful tools for 3D medical image generation, yet bridging the gap between standard training objectives and clinical relevance remains a challenge. This paper presents a method to enhance 3D diffusion models using Reinforcement Learning (RL) with multi-scale feedback. We first pretrain a 3D diffusion model on MRI volumes to establish a robust generative prior. Subsequently, we fine-tune the model using Proximal Policy Optimization (PPO), guided by a novel reward system that integrates both 2D slice-wise assessments and 3D volumetric analysis. This combination allows the model to simultaneously optimize for local texture details and global structural coherence. We validate our framework on the BraTS 2019 and OASIS-1 datasets. Our results indicate that incorporating RL feedback effectively steers the generation process toward higher quality distributions. Quantitative analysis reveals significant improvements in Fréchet Inception Distance (FID) and, crucially, the synthetic data demonstrates enhanced utility in downstream tumor and disease classification tasks compared to non-optimized baselines.
Compositional Reward Models for Conditional Medical Image Generation
Yardi School of Artificial Intelligence（亚尔迪人工智能学院）；Indian Institute Of Technology, Delhi（德里印度理工学院）；Indian Institute of Science（印度科学学院）；Department of Computer Science and Engineering（计算机科学与工程系）
Optimizing Multi-Round Enhanced Training in Diffusion Models for Improved Preference Understanding
Xiamen University（厦门大学）；University of Electronic Science and Technology of China（电子科技大学）；University of Minnesota—Twin Cities（明尼苏达大学—双城分校）；Emory University（埃默里大学）；Tsinghua University（清华大学）；University of Warwick（沃里克大学）；University of the Chinese Academy of Sciences（中国科学院大学）；George Washington University（乔治华盛顿大学）；University of Toronto（多伦多大学）；Peking University（北京大学）
Adaptive Reinforcement for Open-ended Medical Reasoning via Semantic-Guided Reward Collapse Mitigation
College of Intelligent Robotics and Advanced Manufacturing, Fudan University（复旦大学智能机器人与先进制造学院）；Fysics Intelligence Technologies Co., Ltd. (Fysics AI)（菲斯克智能科技有限公司（菲斯克AI））；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）
MediX-R1: Open Ended Medical Reinforcement Learning
Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)（迈赫迈德·本·扎耶德人工智能大学）；Jubilee Mission Medical College（jubilee mission 医学院）；Research Institute（研究院）；JJM Medical College（JJM 医学院）
Optimizing Neural Network Architecture for Medical Image Segmentation Using Monte Carlo Tree Search
Xi'an Kedagaoxin University, Faculty of Science（西安科大新大学科学学院）；Xi'an Kedagaoxin University, School of International Education（西安科大新大学国际教育学院）；Saint Petersburg State University（圣彼得堡国立大学）
Med-RewardBench: Benchmarking Reward Models and Judges for Medical Multimodal Large Language Models
College of Computer Science and Software Engineering, Shenzhen University（深圳大学计算机科学与软件工程学院）；School of Artificial Intelligence, Shenzhen University（深圳大学人工智能学院）；Guangdong Provincial Key Laboratory of Intelligent Information Processing（广东省智能信息处理重点实验室）；The Hong Kong University of Science and Technology（香港科学与技术大学）；Renmin University of China（中国人民大学）；National Yang Ming Chiao Tung University；Taipei Veterans General Hospital（台北荣民总医院）；School of Biomedical Engineering, Shenzhen University（深圳大学生物医学工程学院）
Optimizing Visual Generative Models via Distribution-wise Rewards
University of Science and Technology of China（中国科学技术大学）；Shanghai Innovation Institute（上海创新研究院）；Hunyuan Frontier Lab, Tencent（腾讯混元前沿实验室）；National University of Singapore（新加坡国立大学）
EEG Emotion Copilot: Optimizing Lightweight LLMs for Emotional EEG Interpretation with Assisted Medical Record Generation
Laboratory of Digital Image and Intelligent Computation, Shanghai Maritime University, Shanghai 201306, China（数字图像与智能计算实验室，上海 Maritime 大学）；Department of Neurology, Affiliated Lianyungang Hospital of Xuzhou Medical University, Lianyungang 222002, China（神经内科，徐州医学院附属连云港医院）；Department of Chinese and Bilingual Studies, The Hong Kong Polytechnic University, Hong Kong, SAR, China（中文学与双语研究部，香港理工大学）
University of Maryland（马里兰大学）；Johns Hopkins University（约翰斯·霍普金斯大学）
Numerous large-scale chest x-ray datasets have spearheaded expert-level detection of abnormalities using deep learning. However, these datasets focus on detecting a subset of disease labels that could be present, thus making them distributed and non-iid with partial labels. Recent literature has indicated the impact of batch normalization layers on the convergence of federated learning due to domain shift associated with non-iid data with partial labels. To that end, we propose FedFBN, a federated learning framework that draws inspiration from transfer learning by using pretrained networks as …
The Dice score and Jaccard index are commonly used metrics for the evaluation of segmentation tasks in medical imaging. Convolutional neural networks trained for image segmentation tasks are usually optimized for (weighted) cross-entropy. This introduces an adverse discrepancy between the learning optimization objective (the loss) and the end target metric. Recent works in computer vision have proposed soft surrogates to alleviate this discrepancy and directly optimize the desired metric, either through relaxations (soft-Dice, soft-Jaccard) or submodular optimization (Lovász-softmax). The aim …
ETH Zurich（苏黎世联邦理工学院）；University of Zurich（苏黎世大学）；University Hospital Balgrist（巴尔格里斯特大学医院）
Segmentation is essential for medical image analysis tasks such as intervention planning, therapy guidance, diagnosis, treatment decisions. Deep learning is becoming increasingly prominent for segmentation, where the lack of annotations, however, often becomes the main limitation. Due to privacy concerns and ethical considerations, most medical datasets are created, curated, and allow access only locally. Furthermore, current deep learning methods are often suboptimal in translating anatomical knowledge between different medical imaging modalities. Active learning can be used to select an info…
Hebei University of Technology（河北工业大学）；Shenzhen University（深圳大学）；University of Oxford（牛津大学）
To reduce doctors' workload, deep-learning-based automatic medical report generation has recently attracted more and more research efforts, where deep convolutional neural networks (CNNs) are employed to encode the input images, and recurrent neural networks (RNNs) are used to decode the visual features into medical reports automatically. However, these state-of-the-art methods mainly suffer from three shortcomings: (i) incomprehensive optimization, (ii) low-order and unidimensional attention mechanisms, and (iii) repeated generation. In this article, we propose a hybrid reinforced medical rep…
A Bio-Medical Snake Optimizer System Driven by Logarithmic Surviving Global Search for Optimizing Feature Selection and its application for Disorder Recognition
Middle East University（中东大学）；Al-Balqa Applied University（巴尔卡应用大学）；Helwan University（赫勒万大学）；Guru Nanak Dev University（古鲁纳纳克德夫大学）；University of Granada（格拉纳达大学）
IBM T.J. Watson Research Center（IBM T.J. Watson 研究中心）
Recently it has been shown that policy-gradient methods for reinforcement learning can be utilized to train deep end-to-end systems directly on non-differentiable metrics for the task at hand. In this paper we consider the problem of optimizing image captioning systems using reinforcement learning, and show that by carefully optimizing our systems using the test metrics of the MSCOCO task, significant gains in performance can be realized. Our systems are built using a new optimization approach that we call self-critical sequence training (SCST). SCST is a form of the popular REINFORCE algorith…
KIT — The Research University in the Helmholtz Association（卡尔斯鲁厄理工学院）
We investigate the "Visual Pushing for Grasping" (VPG) system by Zeng et al. and the "Hourglass" system by Ewerton et al., an evolution of the former. The focus of our work is the investigation of the capabilities of both systems to learn long-term rewards and policies. Zeng et al. original task only needs a limited amount of foresight. Ewerton et al. attain their best performance using an agent which only takes the most immediate action under consideration. We are interested in the ability of their models and training algorithms to accurately predict long-term Q-Values. To evaluate this abili…
