Generating radiology reports from 3D volumetric data remains challenging due to the difficulty of grounding fine-grained pathologies within high-dimensional scans.While retrievalaugmented generation (RAG) offers a potential solution, standard approaches struggle with visual-semantic ambiguity and often introduce irrelevant "normal" context that dilutes pathological signals.To address this limitation, we introduce CPR-RAG, a model-agnostic RAG framework that enhances organ-level grounding by integrating clinical priors into the retrieval process.Specifically, we propose a clinical prior-regularized re-ranking module that leverages corpus-derived co-occurrence statistics to align retrieved candidates with latent disease distributions, ensuring clinical consistency beyond mere visual similarity.Furthermore, we employ clinical relevance context refinement to selectively filter out boilerplate normal descriptions, thereby maximizing the information density of the evidence provided to the generator.Extensive experiments on the RadGenome-ChestCT benchmark demonstrate that CPR-RAG significantly improves clinical efficacy across state-of-theart radiology report generation models.Human evaluation further confirms that our approach achieves superior factual correctness, completeness, and utility compared to the existing models.

Accurate tumor segmentation in whole slide image (WSI) is essential for histopathological diagnosis and research, but the traditional manual analysis is labor-intensive and prone to variability. Furthermore, many artificial models focus on specific magnification images, limiting the detailed information available for segmentation. To address these challenges, we propose MFTrans, a novel multi-resolution fusion transformer with a CNN-based architecture designed for efficient tumor segmentation in WSI. Inspired by the diagnostic procedures of expert pathologists, MFTrans integrates both high- and low-magnification images, capturing detailed local features and broader contextual relationships through a dual-branch architecture. The model employs a global token transformer and cross-attention mechanism to fuse hierarchical features from dual branches to improve segmentation performance. We evaluate MFTrans on three real-world WSI datasets: Camelyon16, PAIP2019, and Catholic Uijeongbu St. Mary's hospital dataset, demonstrating its superior segmentation performance over state-of-the-art methods in balanced and imbalanced setups. These results highlight MFTrans's effectiveness in medical image analysis and its generalizability across different datasets, making it a robust tool for automated cancer diagnostics. Our code is available at https://github.com/aimed-gist/MFTrans.

BACKGROUND: In Alzheimer's Diseases (AD) research, multimodal imaging analysis can unveil complementary information from multiple imaging modalities and further our understanding of the disease. One application is to discover disease subtypes using unsupervised clustering. However, existing clustering methods are often applied to input features directly, and could suffer from the curse of dimensionality with high-dimensional multimodal data. The purpose of our study is to identify multimodal imaging-driven subtypes in Mild Cognitive Impairment (MCI) participants using a multiview learning framework based on Deep Generalized Canonical Correlation Analysis (DGCCA), to learn shared latent representation with low dimensions from 3 neuroimaging modalities. RESULTS: DGCCA applies non-linear transformation to input views using neural networks and is able to learn correlated embeddings with low dimensions that capture more variance than its linear counterpart, generalized CCA (GCCA). We designed experiments to compare DGCCA embeddings with single modality features and GCCA embeddings by generating 2 subtypes from each feature set using unsupervised clustering. In our validation studies, we found that amyloid PET imaging has the most discriminative features compared with structural MRI and FDG PET which DGCCA learns from but not GCCA. DGCCA subtypes show differential measures in 5 cognitive assessments, 6 brain volume measures, and conversion to AD patterns. In addition, DGCCA MCI subtypes confirmed AD genetic markers with strong signals that existing late MCI group did not identify. CONCLUSION: Overall, DGCCA is able to learn effective low dimensional embeddings from multimodal data by learning non-linear projections. MCI subtypes generated from DGCCA embeddings are different from existing early and late MCI groups and show most similarity with those identified by amyloid PET features. In our validation studies, DGCCA subtypes show distinct patterns in cognitive measures, brain volumes, and are able to identify AD genetic markers. These findings indicate the promise of the imaging-driven subtypes and their power in revealing disease structures beyond early and late stage MCI.

Alzheimer's disease (AD) is a progressive neurodegenerative brain disorder characterized by memory loss and cognitive decline. Early detection and accurate prognosis of AD is an important research topic, and numerous machine learning methods have been proposed to solve this problem. However, traditional machine learning models are facing challenges in effectively integrating longitudinal neuroimaging data and biologically meaningful structure and knowledge to build accurate and interpretable prognostic predictors. To bridge this gap, we propose an interpretable graph neural network (GNN) model for AD prognostic prediction based on longitudinal neuroimaging data while embracing the valuable knowledge of structural brain connectivity. In our empirical study, we demonstrate that 1) the proposed model outperforms several competing models (i.e., DNN, SVM) in terms of prognostic prediction accuracy, and 2) our model can capture neuroanatomical contribution to the prognostic predictor and yield biologically meaningful interpretation to facilitate better mechanistic understanding of the Alzheimer's disease. Source code is available at https://github.com/JaesikKim/temporal-GNN.

Despite the great success of machine learning in various biomedical domains, applications to dental hard tissue conditions (primarily on dental Caries, Erosive Tooth Wear (ETW), and Fluorosis) are under-explored, in particular for analyzing photographic images. The clinical diagnostics of these dental hard-tissue conditions is routinely performed by visual examination but is often limited by its subjectivity. To bridge this gap, we apply four categories of machine learning strategies including nine different methods with two different feature representations to estimate the probability and severity of dental hard-tissue conditions from photographic tooth images. Our first empirical study is performed on the real dataset containing both controls and cases, and the best probability estimation results are achieved by Extra Trees Regression (RMSE: 0.030, Pearson correlation: 0.600) for Caries, Decision Tree (RMSE: 0.183, Pearson correlation: 0.581) for ETW, and Bayesian ARD Regression (RMSE: 0.191, Pearson correlation: 0.745) for Fluorosis. Our second empirical study is performed on the case only datasets, and the best severity estimation results are achieved by Extra Trees Regression (RMSE: 0.029, Pearson correlation: 0.687) for Caries, Bayesian ARD Regression and Linear Regression (RMSE: 0.192, Pearson correlation: 0.490) for ETW, and Bayesian ARD Regression (RMSE: 0.238, Pearson correlation: 0.537) for Fluorosis. These results indicate that machine learning models provide promising opportunities to help clinical evaluation and save resources in the management of these dental conditions.

We present an effective deep multiview learning framework to identify population structure using multimodal imaging data. Our approach is based on canonical correlation analysis (CCA). We propose to use deep generalized CCA (DGCCA) to learn a shared latent representation of non-linearly mapped and maximally correlated components from multiple imaging modalities with reduced dimensionality. In our empirical study, this representation is shown to effectively capture more variance in original data than conventional generalized CCA (GCCA) which applies only linear transformation to the multi-view data. Furthermore, subsequent cluster analysis on the new feature set learned from DGCCA is able to identify a promising population structure in an Alzheimer's disease (AD) cohort. Genetic association analyses of the clustering results demonstrate that the shared representation learned from DGCCA yields a population structure with a stronger genetic basis than several competing feature learning methods.

Imaging genetics is a method used to detect associations between imaging and genetic variables. Some researchers have used sparse canonical correlation analysis (SCCA) for imaging genetics. This study was conducted to improve the efficiency and interpretability of SCCA. We propose a connectivity-based penalty for incorporating biological prior information. Our proposed approach, named joint connectivity-based SCCA (JCB-SCCA), includes the proposed penalty and can handle multi-modal neuroimaging datasets. Different neuroimaging techniques provide distinct information on the brain and have been used to investigate various neurological disorders, including Parkinson's disease (PD). We applied our algorithm to simulated and real imaging genetics datasets for performance evaluation. Our algorithm was able to select important features in a more robust manner compared with other multivariate methods. The algorithm revealed promising features of single-nucleotide polymorphisms and brain regions related to PD by using a real imaging genetic dataset. The proposed imaging genetics model can be used to improve clinical diagnosis in the form of novel potential biomarkers. We hope to apply our algorithm to cohorts such as Alzheimer's patients or healthy subjects to determine the generalizability of our algorithm.