Recent reinforcement learning (RL) approaches have advanced radiology report generation (RRG), yet two core limitations persist: (1) report-level rewards offer limited evidence-grounded guidance for clinical faithfulness; and (2) current methods lack an explicit self-improving mechanism to align with clinical preference.We introduce clinically aligned Evidence-aware Self-Correcting Reinforcement Learning (ESC-RL), comprising two key components.First, a Group-wise Evidence-aware Alignment Reward (GEAR) delivers group-wise, evidence-aware feedback.GEAR reinforces consistent grounding for true positives, recovers missed findings for false negatives, and suppresses unsupported content for false positives.Second, a Self-correcting Preference Learning (SPL) strategy automatically constructs a reliable, disease-aware preference dataset from multiple noisy observations and leverages an LLM to synthesize refined reports without human supervision.ESC-RL promotes clinically faithful, disease-aligned reward and supports continual self-improvement during training.Extensive experiments on two public chest X-ray datasets demonstrate consistent gains and state-of-the-art performance.

Malignant brain tumors have become an aggressive and dangerous disease that leads to death worldwide. Multi-modal MRI data is crucial for accurate brain tumor segmentation, but missing modalities common in clinical practice can severely degrade the segmentation performance. While incomplete multi-modal learning methods attempt to address this, learning robust and discriminative features from arbitrary missing modalities remains challenging. To address this challenge, we propose a novel Semantic-guided Masked Mutual Learning (SMML) approach to distill robust and discriminative knowledge across diverse missing modality scenarios. Specifically, we propose a novel dual-branch masked mutual learning scheme guided by Hierarchical Consistency Constraints (HCC) to ensure multi-level consistency, thereby enhancing mutual learning in incomplete multi-modal scenarios. The HCC framework comprises a pixel-level constraint that selects and exchanges reliable knowledge to guide the mutual learning process. Additionally, it includes a feature-level constraint that uncovers robust inter-sample and inter-class relational knowledge within the latent feature space. To further enhance multi-modal learning from missing modality data, we integrate a refinement network into each student branch. This network leverages semantic priors from the Segment Anything Model (SAM) to provide supplementary information, effectively complementing the masked mutual learning strategy in capturing auxiliary discriminative knowledge. Extensive experiments on three challenging brain tumor segmentation datasets demonstrate that our method significantly improves performance over state-of-the-art methods in diverse missing modality settings.

Automated radiology report generation is essential for improving diagnostic efficiency and reducing the workload of medical professionals. However, existing methods face significant challenges, such as disease class imbalance and insufficient cross-modal fusion. To address these issues, we propose the learnable Retrieval Enhanced Visual-Text Alignment and Fusion (REVTAF) framework, which effectively tackles both class imbalance and visual-text fusion in report generation. REVTAF incorporates two core components: (1) a Learnable Retrieval Enhancer (LRE) that utilizes semantic hierarchies from hyperbolic space and intra-batch context through a ranking-based metric. LRE adaptively retrieves the most relevant reference reports, enhancing image representations, particularly for underrepresented (tail) class inputs; and (2) a fine-grained visual-text alignment and fusion strategy that ensures consistency across multi-source cross-attention maps for precise alignment. This component further employs an optimal transport-based cross-attention mechanism to dynamically integrate task-relevant textual knowledge for improved report generation. By combining adaptive retrieval with multi-source alignment and fusion, REVTAF achieves fine-grained visual-text integration under weak image-report level supervision while effectively mitigating data imbalance issues. The experiments demonstrate that REVTAF outperforms state-of-the-art methods, achieving an average improvement of 7.4% on the MIMIC-CXR dataset and 2.9% on the IU X-Ray dataset. Comparisons with mainstream multimodal LLMs (e.g., GPT-series models), further highlight its superiority in radiology report generation https://github.com/banbooliang/REVTAF-RRG.

Cervical spondylotic myelopathy (CSM) has a high incidence in the middle-aged and elderly people. According to clinical research, there is a connection between hand dexterity and cervical nerves. So the surgeon makes a preliminary assessment of the severity of CSM based on a 10-second grip and release (G&R) test. At present, the statistics of G&R test rely on the surgeon's manual counting. When a patient's hand motion speed is too fast, the surgeon's manual counting is prone to error, leading to potential misdiagnosis. On the other hand, in recent years, artificial intelligence has been developed rapidly, where three-dimensional convolutional neural networks (3D-CNNs) have been widely used in video analysis. This work proposes a hand motion analysis model using a 3D-CNN combined with a de-jittering mechanism to assess the severity of CSM on 10-second G&R videos. We collect 1500 10-second G&R videos recorded by 750 subjects to establish a dataset. The proposed model using 3D-MobileNetV2 as the classifier obtains a Levenshtein accuracy of 97.40% and an average GPU inference time of 3.31 seconds for each 10-second G&R video. Such accuracy and inference speed ensure that the proposed model can be used as a screening examination tool for CSM and a medical assistance tool to help decision making during CSM treatment planning.