This study aimed to develop and evaluate a deep convolutional neural network (CNN) model for the automatic segmentation of foreign bodies and ghost images in panoramic radiographs (PRs), which can complicate diagnostic interpretation. A dataset of 11,226 PRs from four devices was annotated by two radiologists using the Computer Vision Annotation Tool. A U-Net-based CNN model was trained and evaluated using Intersection over Union (IoU), Dice coefficient, accuracy, precision, recall, and F1 score. For foreign body segmentation, the model achieved validation Dice and IoU scores of 0.9439 and 0.9043, and test scores of 0.9657 and 0.9371. For ghost image segmentation, validation Dice and IoU were 0.8234 and 0.7388, with test scores of 0.8749 and 0.8145. Overall test accuracy exceeded 0.999. The AI model showed high accuracy in segmenting foreign bodies and ghost images in PRs, indicating its potential to assist radiologists. Further clinical validation is recommended.

Osteoporosis classification from X-ray images remains challenging due to the high visual similarity between scans of healthy individuals and osteoporotic patients. In this paper, we propose a novel framework that extracts a discriminative gradient-based map from each X-ray image, capturing subtle structural differences that are not readily apparent to the human eye. The method uses analytic Gabor filters to decompose the image into multi-scale, multi-orientation components. At each pixel, we construct a filter response matrix, from which second-order texture features are derived via covariance analysis, followed by eigenvalue decomposition to capture dominant local patterns. The resulting Gabor Eigen Map serves as a compact, information-rich representation that is both interpretable and lightweight, making it well-suited for deployment on edge devices. These feature maps are further processed using a convolutional neural network (CNN) to extract high-level descriptors, followed by classification using standard machine learning algorithms. Experimental results demonstrate that the proposed framework outperforms existing methods in identifying osteoporotic cases, while offering strong potential for real-time, privacy-preserving inference at the point of care.

Vision-language pretraining has advanced image-text alignment, yet progress in radiology remains constrained by the heterogeneity of clinical reports, including abbreviations, impression-only notes, and stylistic variability. Unlike general-domain settings where more data often leads to better performance, naively scaling to large collections of noisy reports can plateau or even degrade model learning. We ask whether large language model (LLM) encoders can provide robust clinical representations that transfer across diverse styles and better guide image-text alignment. We introduce LLM2VEC4CXR, a domain-adapted LLM encoder for chest X-ray reports, and LLM2CLIP4CXR, a dual-tower framework that couples this encoder with a vision backbone. LLM2VEC4CXR improves clinical text understanding over BERT-based baselines, handles abbreviations and style variation, and achieves strong clinical alignment on report-level metrics. LLM2CLIP4CXR leverages these embeddings to boost retrieval accuracy and clinically oriented scores, with stronger cross-dataset generalization than prior medical CLIP variants. Trained on 1.6M CXR studies from public and private sources with heterogeneous and noisy reports, our models demonstrate that robustness -- not scale alone -- is the key to effective multimodal learning. We release models to support further research in medical image-text representation learning.

In 2022, the WHO conditionally recommended the use of treatment decision algorithms (TDAs) for treatment decision-making in children <10 years with presumptive tuberculosis (TB), aiming to decrease the substantial case detection gap and improve treatment access in high TB-incidence settings. WHO also called for external validation of these TDAs. Within the Decide-TB project (PACT ID: PACTR202407866544155, 23 July 2024), we aim to generate an individual-participant dataset (IPD) from prospective TB diagnostic accuracy cohorts (RaPaed-TB, UMOYA and two cohorts from TB-Speed). Using the IPD, we aim to: (1) assess the diagnostic accuracy of published TDAs using a set of consensus case definitions produced by the National Institute of Health as reference standard (confirmed and unconfirmed vs unlikely TB); (2) evaluate the added value of novel tools (including biomarkers and artificial intelligence-interpreted radiology) in the existing TDAs; (3) generate an artificial population, modelling the target population of children eligible for WHO-endorsed TDAs presenting at primary and secondary healthcare levels and assess the diagnostic accuracy of published TDAs and (4) identify clinical predictors of radiological disease severity in children from the study population of children with presumptive TB. This study will externally validate the first data-driven WHO TDAs in a large, well-characterised and diverse paediatric IPD derived from four large paediatric cohorts of children investigated for TB. The study has received ethical clearance for sharing secondary deidentified data from the ethics committees of the parent studies (RaPaed-TB, UMOYA and TB Speed) and as the aims of this study were part of the parent studies' protocols, a separate approval was not necessary. Study findings will be published in peer-reviewed journals and disseminated at local, regional and international scientific meetings and conferences. This database will serve as a catalyst for the assessment of the inclusion of novel tools and the generation of an artificial population to simulate the impact of novel diagnostic pathways for TB in children at lower levels of healthcare. TDAs have the potential to close the diagnostic gap in childhood TB. Further finetuning of the currently available algorithms will facilitate this and improve access to care.

Foundation vision encoders such as CLIP and DINOv2, trained on web-scale data, exhibit strong transfer performance across tasks and datasets. However, medical imaging foundation models remain constrained by smaller datasets, limiting our understanding of how data scale and pretraining paradigms affect performance in this setting. In this work, we systematically study continual pretraining of two vision encoders, MedImageInsight (MI2) and RAD-DINO representing the two major encoder paradigms CLIP and DINOv2, on up to 3.5M chest x-rays from a single institution, holding compute and evaluation protocols constant. We evaluate on classification (radiology findings, lines and tubes), segmentation (lines and tubes), and radiology report generation. While prior work has primarily focused on tasks related to radiology findings, we include lines and tubes tasks to counterbalance this bias and evaluate a model's ability to extract features that preserve continuity along elongated structures. Our experiments show that MI2 scales more effectively for finding-related tasks, while RAD-DINO is stronger on tube-related tasks. Surprisingly, continually pretraining MI2 with both reports and structured labels using UniCL improves performance, underscoring the value of structured supervision at scale. We further show that for some tasks, as few as 30k in-domain samples are sufficient to surpass open-weights foundation models. These results highlight the utility of center-specific continual pretraining, enabling medical institutions to derive significant performance gains by utilizing in-domain data.

Multimodal large language models (LLMs) have potential for medical image analysis, yet their reliability for pediatric panoramic radiographs remains uncertain. This study evaluated two multimodal LLMs (OpenAI o1, Claude 3.5 Sonnet) for detecting and counting teeth (including tooth germs) on pediatric panoramic radiographs. Eighty-seven pediatric panoramic radiographs from an open-source data set were analyzed. Two pediatric dentists annotated the presence or absence of each potential tooth position. Each image was processed five times by the LLMs using identical prompts, and the results were compared with the expert annotations. Standard performance metrics and Fleiss' kappa were calculated. Detailed examination revealed that subtle developmental stages and minor tooth loss were consistently misidentified. Claude 3.5 Sonnet had higher sensitivity but significantly lower specificity (29.8% ± 21.5%), resulting in many false positives. OpenAI o1 demonstrated superior specificity compared to Claude 3.5 Sonnet, but still failed to correctly detect subtle defects in certain mixed dentition cases. Both models showed large variability in repeated runs. Both LLMs failed to achieve clinically acceptable performance and cannot reliably identify nuanced discrepancies critical for pediatric dentistry. Further refinements and consistency improvements are essential before routine clinical use.

Chest radiographic (CXR) screening is currently not recommended in the United States by any major guideline organization. Multiple randomized controlled trials done in the United States and also in Europe, with the largest being the Prostate, Lung, Colorectal and Ovarian (PLCO) trial, all failed to show a benefit and are used as evidence to support the current recommendation. Nevertheless, there is renewed interest in CXR screening, especially in low- and middle-resourced countries around the world. Reasons for this are multi-factorial, including the continued concern that those trials still may have missed a benefit, but perhaps more importantly, it is now established conclusively that finding smaller cancers is better than finding larger ones. This was the key finding in those large randomized controlled trials for CT screening. So, while CT finds cancers smaller than CXR, both clearly perform better than waiting for cancers to be larger and detected by symptom prompting. Without it being well understood that treating cancers found in the asymptomatic state by CXR, there would also be no basis for treating them when found incidentally. In addition, advances in artificial intelligence are allowing for nodules to be found earlier and more reliably with CXR than in those prior studies, and in many countries around the world, TB screening is already taking place on a large scale. This presents a major opportunity for integration with lung screening programs.

Existing deep learning models for chest radiology often neglect patient metadata, limiting diagnostic accuracy and fairness. To bridge this gap, we introduce MetaCheX, a novel multimodal framework that integrates chest X-ray images with structured patient metadata to replicate clinical decision-making. Our approach combines a convolutional neural network (CNN) backbone with metadata processed by a multilayer perceptron through a shared classifier. Evaluated on the CheXpert Plus dataset, MetaCheX consistently outperformed radiograph-only baseline models across multiple CNN architectures. By integrating metadata, the overall diagnostic accuracy was significantly improved, measured by an increase in AUROC. The results of this study demonstrate that metadata reduces algorithmic bias and enhances model generalizability across diverse patient populations. MetaCheX advances clinical artificial intelligence toward robust, context-aware radiographic disease detection.

Radiology report generation (RRG) methods often lack sufficient medical knowledge to produce clinically accurate reports. A scene graph provides comprehensive information for describing objects within an image. However, automatically generated radiology scene graphs (RSG) may contain noise annotations and highly overlapping regions, posing challenges in utilizing RSG to enhance RRG. To this end, we propose Scene Graph aided RRG (SGRRG), a framework that leverages an automatically generated RSG and copes with noisy supervision problems in the RSG with a transformer-based module, effectively distilling medical knowledge in an end-to-end manner. SGRRG is composed of a dedicated scene graph encoder responsible for translating the radiography into a RSG, and a scene graph-aided decoder that takes advantage of both patch-level and region-level visual information and mitigates the noisy annotation problem in the RSG. The incorporation of both patch-level and region-level features, alongside the integration of the essential RSG construction modules, enhances our framework's flexibility and robustness, enabling it to readily exploit prior advanced RRG techniques. A fine-grained, sentence-level attention method is designed to better distill the RSG information. Additionally, we introduce two proxy tasks to enhance the model's ability to produce clinically accurate reports. Extensive experiments demonstrate that SGRRG outperforms previous state-of-the-art methods in report generation and can better capture abnormal findings. Code is available at https://github.com/Markin-Wang/SGRRG.

X-ray imaging is a ubiquitous in radiology, yet most existing AI foundation models are limited to chest anatomy and fail to generalize across broader clinical tasks. In this work, we introduce XR-0, the multi-anatomy X-ray foundation model using self-supervised learning on a large, private dataset of 1.15 million images spanning diverse anatomical regions and evaluated across 12 datasets and 20 downstream tasks, including classification, retrieval, segmentation, localization, visual grounding, and report generation. XR-0 achieves state-of-the-art performance on most multi-anatomy tasks and remains competitive on chest-specific benchmarks. Our results demonstrate that anatomical diversity and supervision are critical for building robust, general-purpose medical vision models, paving the way for scalable and adaptable AI systems in radiology.