To develop a deep learning model to automatically measure the curvature-related sagittal parameters on cervical spinal X-ray images. This retrospective study collected a total of 700 lateral cervical spine X-ray images from three hospitals, consisting of 500 training sets, 100 internal test sets, and 100 external test sets. 6 measured parameters and 34 landmarks were measured and labeled by two doctors and averaged as the gold standard. A Convolutional neural network (CNN) model was built by training on 500 images and testing on 200 images. Statistical analysis is used to evaluate labeling differences and model performance. The percentages of the difference in distance between landmarks within 4 mm were 96.90% (Dr. A vs. Dr. B), 98.47% (Dr. A vs. model), and 97.31% (Dr. B vs. model); within 3 mm were 94.88% (Dr. A vs. Dr. B), 96.43% (Dr. A vs. model), and 94.16% (Dr. B vs. model). The mean difference of the algorithmic model in labeling landmarks was 1.17 ± 1.14 mm. The mean absolute error (MAE) of the algorithmic model for the Borden method, Cervical curvature index (CCI), Vertebral centroid measurement cervical lordosis (CCL), C₀-C₇ Cobb, C₁-C₇ Cobb, C₂-C₇ Cobb in the test sets are 1.67 mm, 2.01%, 3.22°, 2.37°, 2.49°, 2.81°, respectively; symmetric mean absolute percentage error (SMAPE) was 20.06%, 21.68%, 20.02%, 6.68%, 5.28%, 20.46%, respectively. Also, the algorithmic model of the six cervical sagittal parameters is in good agreement with the gold standard (intraclass correlation efficiency was 0.983; p < 0.001). Our deep learning algorithmic model had high accuracy in recognizing the landmarks of the cervical spine and automatically measuring cervical spine-related parameters, which can help radiologists improve their diagnostic efficiency.

The aim of this study was to automatically detect teeth with dens invaginatus (DI) in panoramic radiographs using deep learning algorithms and to compare the success of the algorithms. For this purpose, 400 panoramic radiographs with DI were collected from the faculty database and separated into 60% training, 20% validation and 20% test images. The training and validation images were labeled by oral, dental and maxillofacial radiologists and augmented with various augmentation methods, and the improved models were asked for the images allocated for the test phase and the results were evaluated according to performance measures including accuracy, sensitivity, F1 score and mean detection time. According to the test results, YOLOv8 achieved a precision, sensitivity and F1 score of 0.904 and was the fastest detection model with an average detection time of 0.041. The Faster R-CNN model achieved 0.912 precision, 0.904 sensitivity and 0.907 F1 score, with an average detection time of 0.1 s. The YOLOv9 algorithm showed the most successful performance with 0.946 precision, 0.930 sensitivity, 0.937 F1 score value and the average detection speed per image was 0.158 s. According to the results obtained, all models achieved over 90% success. YOLOv8 was relatively more successful in detection speed and YOLOv9 in other performance criteria. Faster R-CNN ranked second in all criteria.

We present ReXVQA, the largest and most comprehensive benchmark for visual question answering (VQA) in chest radiology, comprising approximately 696,000 questions paired with 160,000 chest X-rays studies across training, validation, and test sets. Unlike prior efforts that rely heavily on template based queries, ReXVQA introduces a diverse and clinically authentic task suite reflecting five core radiological reasoning skills: presence assessment, location analysis, negation detection, differential diagnosis, and geometric reasoning. We evaluate eight state-of-the-art multimodal large language models, including MedGemma-4B-it, Qwen2.5-VL, Janus-Pro-7B, and Eagle2-9B. The best-performing model (MedGemma) achieves 83.24% overall accuracy. To bridge the gap between AI performance and clinical expertise, we conducted a comprehensive human reader study involving 3 radiology residents on 200 randomly sampled cases. Our evaluation demonstrates that MedGemma achieved superior performance (83.84% accuracy) compared to human readers (best radiology resident: 77.27%), representing a significant milestone where AI performance exceeds expert human evaluation on chest X-ray interpretation. The reader study reveals distinct performance patterns between AI models and human experts, with strong inter-reader agreement among radiologists while showing more variable agreement patterns between human readers and AI models. ReXVQA establishes a new standard for evaluating generalist radiological AI systems, offering public leaderboards, fine-grained evaluation splits, structured explanations, and category-level breakdowns. This benchmark lays the foundation for next-generation AI systems capable of mimicking expert-level clinical reasoning beyond narrow pathology classification. Our dataset will be open-sourced at https://huggingface.co/datasets/rajpurkarlab/ReXVQA

During radiation therapy, the natural movement of organs can lead to underdosing the cancer and overdosing the healthy tissue, compromising treatment efficacy. Real-time image-guided adaptive radiation therapy can track the tumour and account for the motion. Typically, fiducial markers are implanted as a surrogate for the tumour position due to the low radiographic contrast of soft tissues in kilovoltage (kV) images. A segmentation approach that does not require markers would eliminate the costs, delays, and risks associated with marker implantation. We trained patient-specific conditional Generative Adversarial Networks for prostate segmentation in kV images. The networks were trained using synthetic kV images generated from each patient's own imaging and planning data, which are available prior to the commencement of treatment. We validated the networks on two treatment fractions from 30 patients using multi-centre data from two clinical trials. Here, we present a large-scale proof-of-principle study of x-ray-based markerless prostate segmentation for globally available cancer therapy systems. Our results demonstrate the feasibility of a deep learning approach using kV images to track prostate motion across the entire treatment arc for 30 patients with prostate cancer. The mean absolute deviation is 1.4 and 1.6 mm in the anterior-posterior/lateral and superior-inferior directions, respectively. Markerless segmentation via deep learning may enable real-time image guidance on conventional cancer therapy systems without requiring implanted markers or additional hardware, thereby expanding access to real-time adaptive radiation therapy.

This study addresses the critical challenge of detecting out-of-distribution (OOD) chest X-rays, where subtle view differences between lateral and frontal radiographs can lead to diagnostic errors. We develop a GAN-based framework that learns the inherent feature distribution of frontal views from the MIMIC-CXR dataset through latent space optimization and Kolmogorov-Smirnov statistical testing. Our approach generates similarity scores to reliably identify OOD cases, achieving exceptional performance with 100% precision, and 97.5% accuracy in detecting lateral views. The method demonstrates consistent reliability across operating conditions, maintaining accuracy above 92.5% and precision exceeding 93% under varying detection thresholds. These results provide both theoretical insights and practical solutions for OOD detection in medical imaging, demonstrating how GANs can establish feature representations for identifying distributional shifts. By significantly improving model reliability when encountering view-based anomalies, our framework enhances the clinical applicability of deep learning systems, ultimately contributing to improved diagnostic safety and patient outcomes.

Deep learning models trained with spatial omics data uncover complex patterns and relationships among cells, genes, and proteins in a high-dimensional space. State-of-the-art in silico spatial multi-cell gene expression methods using histological images of tissue stained with hematoxylin and eosin (H&E) allow us to characterize cellular heterogeneity. We developed a vision transformer (ViT) framework to map histological signatures to spatial single-cell transcriptomic signatures, named SPiRiT. SPiRiT predicts single-cell spatial gene expression using the matched H&E image tiles of human breast cancer and whole mouse pup, evaluated by Xenium (10x Genomics) datasets. Importantly, SPiRiT incorporates rigorous strategies to ensure reproducibility and robustness of predictions and provides trustworthy interpretation through attention-based model explainability. SPiRiT model interpretation revealed the areas, and attention details it uses to predict gene expressions like marker genes in invasive cancer cells. In an apple-to-apple comparison with ST-Net, SPiRiT improved the predictive accuracy by 40%. These gene predictions and expression levels were highly consistent with the tumor region annotation. In summary, SPiRiT highlights the feasibility to infer spatial single-cell gene expression using tissue morphology in multiple-species.

Diagnostic imaging interpretation involves distilling multimodal clinical information into text form, a task well-suited to augmentation by generative artificial intelligence (AI). However, to our knowledge, impacts of AI-based draft radiological reporting remain unstudied in clinical settings. To prospectively evaluate the association of radiologist use of a workflow-integrated generative model capable of providing draft radiological reports for plain radiographs across a tertiary health care system with documentation efficiency, the clinical accuracy and textual quality of final radiologist reports, and the model's potential for detecting unexpected, clinically significant pneumothorax. This prospective cohort study was conducted from November 15, 2023, to April 24, 2024, at a tertiary care academic health system. The association between use of the generative model and radiologist documentation efficiency was evaluated for radiographs documented with model assistance compared with a baseline set of radiographs without model use, matched by study type (chest or nonchest). Peer review was performed on model-assisted interpretations. Flagging of pneumothorax requiring intervention was performed on radiographs prospectively. The primary outcomes were association of use of the generative model with radiologist documentation efficiency, assessed by difference in documentation time with and without model use using a linear mixed-effects model; for peer review of model-assisted reports, the difference in Likert-scale ratings using a cumulative-link mixed model; and for flagging pneumothorax requiring intervention, sensitivity and specificity. A total of 23 960 radiographs (11 980 each with and without model use) were used to analyze documentation efficiency. Interpretations with model assistance (mean [SE], 159.8 [27.0] seconds) were faster than the baseline set of those without (mean [SE], 189.2 [36.2] seconds) (P = .02), representing a 15.5% documentation efficiency increase. Peer review of 800 studies showed no difference in clinical accuracy (χ2 = 0.68; P = .41) or textual quality (χ2 = 3.62; P = .06) between model-assisted interpretations and nonmodel interpretations. Moreover, the model flagged studies containing a clinically significant, unexpected pneumothorax with a sensitivity of 72.7% and specificity of 99.9% among 97 651 studies screened. In this prospective cohort study of clinical use of a generative model for draft radiological reporting, model use was associated with improved radiologist documentation efficiency while maintaining clinical quality and demonstrated potential to detect studies containing a pneumothorax requiring immediate intervention. This study suggests the potential for radiologist and generative AI collaboration to improve clinical care delivery.

In modern healthcare, telemedicine, health records, and AI-driven diagnostics depend on medical image watermarking to secure chest X-rays for pneumonia diagnosis, ensuring data integrity, confidentiality, and authenticity. A 2024 study found over 70 % of healthcare institutions faced medical image data breaches. Yet, current methods falter in imperceptibility, robustness against attacks, and deployment efficiency. ViTU-Net integrates cutting-edge techniques to address these multifaceted challenges in medical image security and analysis. The model's core component, the Vision Transformer (ViT) encoder, efficiently captures global dependencies and spatial information, while the U-Net decoder enhances image reconstruction, with both components leveraging the Adaptive Hierarchical Spatial Attention (AHSA) module for improved spatial processing. Additionally, the patch-based LSB embedding mechanism ensures focused embedding of reversible fragile watermarks within each patch of the segmented non-diagnostic region (RONI), guided dynamically by adaptive masks derived from the attention mechanism, minimizing impact on diagnostic accuracy while maximizing precision and ensuring optimal utilization of spatial information. The hybrid meta-heuristic optimization algorithm, TuniBee Fusion, dynamically optimizes watermarking parameters, striking a balance between exploration and exploitation, thereby enhancing watermarking efficiency and robustness. The incorporation of advanced cryptographic techniques, including SHA-512 hashing and AES encryption, fortifies the model's security, ensuring the authenticity and confidentiality of watermarked medical images. A PSNR value of 60.7 dB, along with an NCC value of 0.9999 and an SSIM value of 1.00, underscores its effectiveness in preserving image quality, security, and diagnostic accuracy. Robustness analysis against a spectrum of attacks validates ViTU-Net's resilience in real-world scenarios.

Medical vision-language alignment through cross-modal contrastive learning shows promising performance in image-text matching tasks, such as retrieval and zero-shot classification. However, conventional cross-modal contrastive learning (CLIP-based) methods suffer from suboptimal visual representation capabilities, which also limits their effectiveness in vision-language alignment. In contrast, although the models pretrained via multimodal masked modeling struggle with direct cross-modal matching, they excel in visual representation. To address this contradiction, we propose ALTA (ALign Through Adapting), an efficient medical vision-language alignment method that utilizes only about 8% of the trainable parameters and less than 1/5 of the computational consumption required for masked record modeling. ALTA achieves superior performance in vision-language matching tasks like retrieval and zero-shot classification by adapting the pretrained vision model from masked record modeling. Additionally, we integrate temporal-multiview radiograph inputs to enhance the information consistency between radiographs and their corresponding descriptions in reports, further improving the vision-language alignment. Experimental evaluations show that ALTA outperforms the best-performing counterpart by over 4% absolute points in text-to-image accuracy and approximately 6% absolute points in image-to-text retrieval accuracy. The adaptation of vision-language models during efficient alignment also promotes better vision and language understanding. Code is publicly available at https://github.com/DopamineLcy/ALTA.

When taking chest X-rays, the patient is encouraged to take maximum inspiration and the radiological technologist takes the images at the appropriate time. If the image is not taken at maximum inspiration, retaking of the image is required. However, there is variation in the judgment of whether retaking is necessary between the operators. Therefore, we considered that it might be possible to reduce variation in judgment by developing a retaking assessment system that evaluates whether retaking is necessary using a convolutional neural network (CNN). To train the CNN, the input chest X-ray image and the corresponding correct label indicating whether retaking is necessary are required. However, chest X-ray images cannot distinguish whether inspiration is sufficient and does not need to be retaken, or insufficient and retaking is required. Therefore, we generated input images and labels from dynamic digital radiography (DDR) and conducted the training. Verification using 18 dynamic chest X-ray cases (5400 images) and 48 actual chest X-ray cases (96 images) showed that the VGG16-based architecture achieved an assessment accuracy of 82.3% even for actual chest X-ray images. Therefore, if the proposed method is used in hospitals, it could possibly reduce the variability in judgment between operators.