Hathaway, Quincy A.

Graph neural networks (GNNs) are increasingly used for medical imaging analysis and biological data modeling, where the integration of radiomics, topology, geometry, and generative artificial intelligence (AI) may improve representation learning from medical images and related biomedical data. Across the reviewed literature, GNNs show particular value for modeling spatial relationships, multimodal interactions, graph-structured biological networks, and non-Euclidean imaging features that are difficult to capture using conventional convolutional architectures alone. Topology- and geometry-aware approaches further expand this capability by encoding multi-scale structure, higher-order relationships, curvature, geodesic organization, and equivariant spatial priors. Hybrid graph–transformer models and generative graph methods represent emerging directions for modeling long-range dependencies, augmenting scarce datasets, supporting synthetic pretraining, and improving representation learning in low-label or heterogeneous biomedical settings. However, clinical translation remains limited by variability in graph construction, limited external validation, computational cost, scalability constraints, interpretability challenges, and uncertainty regarding the biological realism of synthetic data. Overall, this review highlights that GNN-based medical imaging analysis is most likely to advance when graph construction is biologically justified, model performance is evaluated across diverse clinical cohorts, and technical gains are paired with transparent validation, interpretability, and implementation strategies.

Background/Objectives: Juxtapleural and perifissural nodules are commonly encountered on thoracic imaging, are typically presumed benign, and are often considered to represent intrapulmonary lymph nodes. We evaluated the prevalence, clinical context, and management of juxtapleural nodules over a 10-year period and to assess factors associated with concern for malignancy. Methods: In this retrospective, single-center study, radiology reports from 1 January 2015 to 31 December 2024, were queried using natural language processing to identify examinations containing terminology consistent with juxtapleural nodules. After exclusion of duplicate encounters, 85,435 unique patient encounters were analyzed. Reports were further filtered to identify cases with imaging concern for malignancy and recommendations for follow-up or intervention (N = 659). Clinical characteristics, imaging modality, and prior malignancy history were compared using appropriate statistical tests. Results: Among 85,435 unique encounters, 659 patients (<1%) had juxtapleural nodules considered concerning for malignancy. These patients were older (68 vs. 64 years, p < 0.001), more frequently imaged with PET/CT (8% vs. 5%, p < 0.001), and more likely to be evaluated in inpatient settings (16% vs. 12%, p < 0.001). Of the cohort with a juxtapleural nodule considered concerning for malignancy by imaging criteria, 360 patients (55%) had a prior history of malignancy. In these patients, nodules were more commonly identified in outpatient settings (70% vs. 43%, p < 0.001) and on routine imaging (94% vs. 86%, p < 0.001). Representative cases demonstrated that malignant nodules may mimic benign perifissural morphology on initial imaging. Conclusions: Although the vast majority of reported juxtapleural nodules did not prompt malignancy-directed evaluation, a small subset of cases (approximately 0.8%), particularly in patients with prior malignancy, were interpreted as sufficiently concerning to warrant follow-up imaging or intervention. Because pathologic confirmation and standardized long-term outcome adjudication were not available for all patients, these findings should be interpreted as reflecting real-world radiologist concern and management patterns rather than a definitive malignancy rate.

The field of radiology is experiencing rapid adoption of large language models (LLMs), yet their tendency to generate hallucinations (plausible but incorrect information) remains a significant barrier to trust. This comprehensive review evaluates emerging agentic artificial intelligence (AI) approaches, including multi-agent role-based systems, retrieval-augmented generation (RAG), and uncertainty quantification, to assess their potential for reducing hallucinations in radiology workflows. Evidence from 2024 to 2025 demonstrates that agentic AI can improve diagnostic accuracy and reduce error rates, though these methods remain computationally demanding and lack comprehensive clinical validation. Multi-agent frameworks enable cross-validation through role-based specialization and systematic workflow orchestration, while RAG strategies enhance accuracy by grounding responses in verified medical literature. Within multi-agent systems, uncertainty quantification enables agents to communicate confidence levels to one another, allowing them to appropriately weigh each other’s contributions during collaborative analysis. While multi-agent frameworks and RAG strategies show significant promise, practical deployment will require careful integration with human oversight, robust evaluation metrics tailored to medical imaging tasks, and regulatory adaptation to ensure safe clinical use in diverse patient populations and imaging modalities.

This review explores how multimodal foundation models (MFMs) are transforming biliary tract cancer (BTC) research. BTCs are aggressive malignancies with poor prognosis, presenting unique challenges due to difficult diagnostic methods, molecular complexity, and rarity. Importantly, intrahepatic cholangiocarcinoma (iCCA), perihilar cholangiocarcinoma (pCCA), and distal bile duct cholangiocarcinoma (dCCA) represent fundamentally distinct clinical entities, with iCCA presenting as mass-forming lesions amenable to biopsy and targeted therapies, while pCCA manifests as infiltrative bile duct lesions with challenging diagnosis and primarily palliative management approaches. MFMs offer potential to advance research by integrating radiological images, histopathology, multi-omics profiles, and clinical data into unified computational frameworks, with applications tailored to these distinct BTC subtypes. Key applications include enhanced biomarker discovery that identifies previously unrecognizable cross-modal patterns, potential for improving currently limited diagnostic accuracy—though validation in BTC-specific cohorts remains essential—accelerated drug repurposing, and advanced patient stratification for personalized treatment. Despite promising results, challenges such as data scarcity, high computational demands, and clinical workflow integration remain to be addressed. Future research should focus on standardized data protocols, architectural innovations, and prospective validation studies. The integration of artificial intelligence (AI)-based methodologies offers new solutions for these historically challenging malignancies. However, current evidence for BTC-specific applications remains largely theoretical, with most studies limited to proof-of-concept designs or related cancer types. Comprehensive clinical validation studies and prospective trials demonstrating patient benefit are essential prerequisites for clinical implementation. The timeline for evidence-based clinical adoption likely extends 7–10 years, contingent on successful completion of validation studies addressing current evidence gaps.

The integration of artificial intelligence (AI) in medical imaging has revolutionized diagnostic capabilities, yet the black-box nature of deep learning models poses significant challenges for clinical adoption. Current explainable AI (XAI) approaches, including SHAP, LIME, and Grad-CAM, predominantly focus on post hoc explanations that may inadvertently undermine clinical decision-making by providing misleading confidence in AI outputs. This paper presents a systematic review and meta-analysis of 67 studies (covering 23 radiology, 19 pathology, and 25 ophthalmology applications) evaluating XAI fidelity, stability, and performance trade-offs across medical imaging modalities. Our meta-analysis of 847 initially identified studies reveals that LIME achieves superior fidelity (0.81, 95% CI: 0.78–0.84) compared to SHAP (0.38, 95% CI: 0.35–0.41) and Grad-CAM (0.54, 95% CI: 0.51–0.57) across all modalities. Post hoc explanations demonstrated poor stability under noise perturbation, with SHAP showing 53% degradation in ophthalmology applications (ρ = 0.42 at 10% noise) compared to 11% in radiology (ρ = 0.89). We demonstrate a consistent 5–7% AUC performance penalty for interpretable models but identify modality-specific stability patterns suggesting that tailored XAI approaches are necessary. Based on these empirical findings, we propose a comprehensive three-pillar accountability framework that prioritizes transparency in model development, interpretability in architecture design, and a cautious deployment of post hoc explanations with explicit uncertainty quantification. This approach offers a pathway toward genuinely accountable AI systems that enhance rather than compromise clinical decision-making quality and patient safety.

Abstract Breast cancer metastases are most commonly found in bone, an indication of poor prognosis. Pathway-based biomarkers identification may help elucidate the cellular signature of breast cancer metastasis in bone, further characterizing the etiology and promoting new therapeutic approaches. We extracted gene expression profiles from mouse macrophages from the GEO dataset, GSE152795 using the GEO2R webtool. The differentially expressed genes (DEGs) were filtered by log2 fold-change with threshold 1.5 (FDR < 0.05). STRING database and Enrichr were used for GO-term analysis, miRNA and TF analysis associated with DEGs. Autodock Vienna was exploited to investigate interaction of anti-cancer drugs, Actinomycin-D and Adriamycin. Sensitivity and specificity of DEGs was assessed using receiver operating characteristic (ROC) analyses. A total of 61 DEGs, included 27 down-regulated and 34 up-regulated, were found to be significant in breast cancer bone metastasis. Major DEGs were associated with lipid metabolism and immunological response of tumor tissue. Crucial DEGs, Bcl3, ADGRG7, FABP4, VCAN, and IRF4 were regulated by miRNAs, miR-497, miR-574, miR-138 and TFs, CCDN1, STAT6, IRF8. Docking analysis showed that these genes possessed strong binding with the drugs. ROC analysis demonstrated Bcl3 is specific to metastasis. DEGs Bcl3, ADGRG7, FABP4, IRF4, their regulating miRNAs and TFs have strong impact on proliferation and metastasis of breast cancer in bone tissues. In conclusion, present study revealed that DEGs are directly involved in of breast tumor metastasis in bone tissues. Identified genes, miRNAs, and TFs can be possible drug targets that may be used for the therapeutics. However, further experimental validation is necessary.