Investigators and funding organizations desire knowledge on topics and trends in publicly funded research but current efforts for manual categorization have been limited in breadth and depth of understanding. We present a semiautomated analysis of 21 years of R-type National Cancer Institute (NCI) grants to departments of radiation oncology and radiology using natural language processing. We selected all noneducation R-type NCI grants from 2000 to 2020 awarded to departments of radiation oncology/radiology with affiliated schools of medicine. We used pretrained word embedding vectors to represent each grant abstract. A sequential clustering algorithm assigned each grant to 1 of 60 clusters representing research topics; we repeated the same workflow for 15 clusters for comparison. Each cluster was then manually named using the top words and closest documents to each cluster centroid. The interpretability of document embeddings was evaluated by projecting them onto 2 dimensions. Changes in clusters over time were used to examine temporal funding trends. We included 5874 grants totaling 1.9 billion dollars of NCI funding over 21 years. The human-model agreement was similar to the human interrater agreement. Two-dimensional projections of grant clusters showed 2 dominant axes: physics-biology and therapeutic-diagnostic. Therapeutic and physics clusters have grown faster over time than diagnostic and biology clusters. The 3 topics with largest funding increase were imaging biomarkers, informatics, and radiopharmaceuticals, which all had a mean annual growth of >$218,000. The 3 topics with largest funding decrease were cellular stress response, advanced imaging hardware technology, and improving performance of breast cancer computer-aided detection, which all had a mean decrease of >$110,000. We developed a semiautomated natural language processing approach to analyze research topics and funding trends. We applied this approach to NCI funding in the radiological sciences to extract both domains of research being funded and temporal trends.

Advancements of deep learning in medical imaging are often constrained by the limited availability of large, annotated datasets, resulting in underperforming models when deployed under real-world conditions. This study investigated a generative artificial intelligence (AI) approach to create synthetic medical images taking the example of bone scintigraphy scans, to increase the data diversity of small-scale datasets for more effective model training and improved generalization. We trained a generative model on ^99mTc-bone scintigraphy scans from 9,170 patients in one center to generate high-quality and fully anonymized annotated scans of patients representing two distinct disease patterns: abnormal uptake indicative of (i) bone metastases and (ii) cardiac uptake indicative of cardiac amyloidosis. A blinded reader study was performed to assess the clinical validity and quality of the generated data. We investigated the added value of the generated data by augmenting an independent small single-center dataset with synthetic data and by training a deep learning model to detect abnormal uptake in a downstream classification task. We tested this model on 7,472 scans from 6,448 patients across four external sites in a cross-tracer and cross-scanner setting and associated the resulting model predictions with clinical outcomes. The clinical value and high quality of the synthetic imaging data were confirmed by four readers, who were unable to distinguish synthetic scans from real scans (average accuracy: 0.48% [95% CI 0.46-0.51]), disagreeing in 239 (60%) of 400 cases (Fleiss' kappa: 0.18). Adding synthetic data to the training set improved model performance by a mean (± SD) of 33(± 10)% AUC (p < 0.0001) for detecting abnormal uptake indicative of bone metastases and by 5(± 4)% AUC (p < 0.0001) for detecting uptake indicative of cardiac amyloidosis across both internal and external testing cohorts, compared to models without synthetic training data. Patients with predicted abnormal uptake had adverse clinical outcomes (log-rank: p < 0.0001). Generative AI enables the targeted generation of bone scintigraphy images representing different clinical conditions. Our findings point to the potential of synthetic data to overcome challenges in data sharing and in developing reliable and prognostic deep learning models in data-limited environments.

Pathological grade is a critical determinant of clinical outcomes and decision-making of follicular lymphoma (FL). This study aimed to develop a deep learning model as a digital biopsy for the non-invasive identification of FL grade. This study retrospectively included 513 FL patients from five independent hospital centers, randomly divided into training, internal validation, and external validation cohorts. A multimodal fusion Transformer model was developed integrating 3D PET tumor images with tabular data to predict FL grade. Additionally, the model is equipped with explainable modules, including Gradient-weighted Class Activation Mapping (Grad-CAM) for PET images, SHapley Additive exPlanations analysis for tabular data, and the calculation of predictive contribution ratios for both modalities, to enhance clinical interpretability and reliability. The predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC) and accuracy, and its prognostic value was also assessed. The Transformer model demonstrated high accuracy in grading FL, with AUCs of 0.964-0.985 and accuracies of 90.2-96.7% in the training cohort, and similar performance in the validation cohorts (AUCs: 0.936-0.971, accuracies: 86.4-97.0%). Ablation studies confirmed that the fusion model outperformed single-modality models (AUCs: 0.974 - 0.956, accuracies: 89.8%-85.8%). Interpretability analysis revealed that PET images contributed 81-89% of the predictive value. Grad-CAM highlighted the tumor and peri-tumor regions. The model also effectively stratified patients by survival risk (P < 0.05), highlighting its prognostic value. Our study developed an explainable multimodal fusion Transformer model for accurate grading and prognosis of FL, with the potential to aid clinical decision-making.

Large language models (LLMs) show promise in radiological diagnosis, but their performance may be affected by the context of the cases presented. Our purpose is to investigate how providing information about prior probabilities influences the diagnostic performance of an LLM in radiological quiz cases. We analyzed 322 consecutive cases from Radiology's "Diagnosis Please" quiz using Claude 3.5 Sonnet under three conditions: without context (Condition 1), informed as quiz cases (Condition 2), and presented as primary care cases (Condition 3). Diagnostic accuracy was compared using McNemar's test. The overall accuracy rate significantly improved in Condition 2 compared to Condition 1 (70.2% vs. 64.9%, p = 0.029). Conversely, the accuracy rate significantly decreased in Condition 3 compared to Condition 1 (59.9% vs. 64.9%, p = 0.027). Providing information that may influence prior probabilities significantly affects the diagnostic performance of the LLM in radiological cases. This suggests that LLMs may incorporate Bayesian-like principles and adjust the weighting of their diagnostic responses based on prior information, highlighting the potential for optimizing LLM's performance in clinical settings by providing relevant contextual information.

In the healthcare field, lung disease detection techniques based on deep learning (DL) are widely used. However, achieving high stability while maintaining privacy remains a challenge. To address this, this research employs Federated Learning (FL), enabling doctors to train models without sharing patient data with unauthorized parties, preserving privacy in local models. The study introduces the Deep Kronecker Convolutional Neural Network (DKCN-Net) for lung disease detection. Input Computed Tomography (CT) images are sourced from the LIDC-IDRI database and denoised using the Adaptive Gaussian Filter (AGF). After that, the Lung lobe and nodule segmentation are performed using Deep Fuzzy Clustering (DFC) and a 3-Dimensional Fully Convolutional Neural Network (3D-FCN). During feature extraction, various features, including statistical, Convolutional Neural Networks (CNN), and Gray-Level Co-Occurrence Matrix (GLCM), are obtained. Lung diseases are then detected using DKCN-Net, which combines the Deep Kronecker Neural Network (DKN) and Parallel Convolutional Neural Network (PCNN). The DKCN-Net achieves an accuracy of 92.18 %, a loss of 7.82 %, a Mean Squared Error (MSE) of 0.858, a True Positive Rate (TPR) of 92.99 %, and a True Negative Rate (TNR) of 92.19 %, with a processing time of 50 s per timestamp.

Gallbladder (GB) adenomas, precancerous lesions for gallbladder carcinoma (GBC), lack reliable non-invasive tools for preoperative differentiation of neoplastic polyps from cholesterol polyps. This study aimed to evaluate an interpretable machine learning (ML) combined model for the precise differentiation of the pathological nature of gallbladder polyps (GPs). This study consecutively enrolled 744 patients from Xi'an Jiaotong University First Affiliated Hospital between January 2017 and December 2023 who were pathologically diagnosed postoperatively with cholesterol polyps, adenomas or T1-stage GBC. Radiomics features were extracted and selected, while clinical variables were subjected to univariate and multivariate logistic regression analyses to identify significant predictors of neoplastic polyps. A optimal ML-based radiomics model was developed, and separate clinical, US and combined models were constructed. Finally, SHapley Additive exPlanations (SHAP) was employed to visualize the classification process. The areas under the curves (AUCs) of the CatBoost-based radiomics model were 0.852 (95 % CI: 0.818-0.884) and 0.824 (95 % CI: 0.758-0.881) for the training and test sets, respectively. The combined model demonstrated the best performance with an improved AUC of 0.910 (95 % CI: 0.885-0.934) and 0.869 (95 % CI: 0.812-0.919), outperformed the clinical, radiomics, and US model (all P < 0.05), and reduced the rate of unnecessary cholecystectomies. SHAP analysis revealed that the polyp short diameter is a crucial independent risk factor in predicting the nature of the GPs. The ML-based combined model may be an effective non-invasive tool for improving the precision treatment of GPs, utilizing SHAP to visualize the classification process can enhance its clinical application.

Deep generative models have gained considerable attention in low-level vision tasks due to their powerful generative capabilities. Among these, diffusion model-based approaches, which employ a forward diffusion process to degrade an image and a reverse denoising process for image generation, have become particularly prominent for producing high-quality, diverse samples with intricate texture details. Despite their widespread success in low-level vision, there remains a lack of a comprehensive, insightful survey that synthesizes and organizes the advances in diffusion model-based techniques. To address this gap, this paper presents the first comprehensive review focused on denoising diffusion models applied to low-level vision tasks, covering both theoretical and practical contributions. We outline three general diffusion modeling frameworks and explore their connections with other popular deep generative models, establishing a solid theoretical foundation for subsequent analysis. We then categorize diffusion models used in low-level vision tasks from multiple perspectives, considering both the underlying framework and the target application. Beyond natural image processing, we also summarize diffusion models applied to other low-level vision domains, including medical imaging, remote sensing, and video processing. Additionally, we provide an overview of widely used benchmarks and evaluation metrics in low-level vision tasks. Our review includes an extensive evaluation of diffusion model-based techniques across six representative tasks, with both quantitative and qualitative analysis. Finally, we highlight the limitations of current diffusion models and propose four promising directions for future research. This comprehensive review aims to foster a deeper understanding of the role of denoising diffusion models in low-level vision.

This study aimed to compare the quality and readability of the responses generated by 3 publicly available artificial intelligence (AI) chatbots in answering frequently asked questions (FAQs) related to Oral and Maxillofacial Radiology (OMR) to assess their suitability for patient education. Fifteen OMR-related questions were selected from professional patient information websites. These questions were posed to ChatGPT-3.5 by OpenAI, Gemini 1.5 Pro by Google, and Copilot by Microsoft to generate responses. Three board-certified OMR specialists evaluated the responses regarding scientific adequacy, ease of understanding, and overall reader satisfaction. Readability was assessed using the Flesch-Kincaid Grade Level (FKGL) and Flesch Reading Ease (FRE) scores. The Wilcoxon signed-rank test was conducted to compare the scores assigned by the evaluators to the responses from the chatbots and professional websites. Interevaluator agreement was examined by calculating the Fleiss kappa coefficient. There were no significant differences between groups in terms of scientific adequacy. In terms of readability, chatbots had overall mean FKGL and FRE scores of 12.97 and 34.11, respectively. Interevaluator agreement level was generally high. Although chatbots are relatively good at responding to FAQs, validating AI-generated information using input from healthcare professionals can enhance patient care and safety. Readability of the text content in the chatbots and websites requires high reading levels.

The assessment of neoadjuvant treatment's response is critical for selecting the most suitable therapeutic options for patients with breast cancer to reduce the need for invasive local therapies. Breast magnetic resonance imaging (MRI) is so far one of the most accurate approaches for assessing pathological complete response, although this is limited by the qualitative and subjective nature of radiologists' assessment, often making it insufficient for deciding whether to forgo additional locoregional therapy measures. To increase the accuracy and prediction of radiomic MRI with the aid of machine learning models and deep learning methods, as part of artificial intelligence, have been used to analyse the different subtypes of breast cancer and the specific changes observed before and after therapy. This review discusses recent advancements in radiomic MRI models for presurgical response assessment for patients with early breast cancer receiving preoperative treatments, with a focus on their implications for clinical practice.

Magnetic Resonance Imaging (MRI) has become a pivotal tool in diagnosing brain diseases, with a wide array of computer-aided artificial intelligence methods being proposed to enhance diagnostic accuracy. However, early studies were often limited by small-scale datasets and a narrow range of disease types, which posed challenges in model generalization. This study presents UniBrain, a hierarchical knowledge-enhanced pre-training framework designed for universal brain MRI diagnosis. UniBrain leverages a large-scale dataset comprising 24,770 imaging-report pairs from routine diagnostics for pre-training. Unlike previous approaches that either focused solely on visual representation learning or used brute-force alignment between vision and language, the framework introduces a hierarchical alignment mechanism. This mechanism extracts structured knowledge from free-text clinical reports at multiple granularities, enabling vision-language alignment at both the sequence and case levels, thereby significantly improving feature learning efficiency. A coupled vision-language perception module is further employed for text-guided multi-label classification, which facilitates zero-shot evaluation and fine-tuning of downstream tasks without modifying the model architecture. UniBrain is validated on both in-domain and out-of-domain datasets, consistently surpassing existing state-of-the-art diagnostic models and demonstrating performance on par with radiologists in specific disease categories. It shows strong generalization capabilities across diverse tasks, highlighting its potential for broad clinical application. The code is available at https://github.com/ljy19970415/UniBrain.