Surgeons treating skeletally immature patients use skeletal age to determine appropriate surgical strategies. Traditional bone age estimation methods utilizing hand radiographs are time-consuming. To develop highly accurate/reliable deep learning (DL) models for determination of accurate skeletal age from hand radiographs. Cohort Study. The authors utilized 3 publicly available hand radiograph data sets for model development/validation from (1) the Radiological Society of North America (RSNA), (2) the Radiological Hand Pose Estimation (RHPE) data set, and (3) the Digital Hand Atlas (DHA). All 3 data sets report corresponding sex and skeletal age. The RHPE and DHA also contain chronological age. After image preprocessing, a ConvNeXt model was trained first on the RSNA data set using sex/skeletal age as inputs using 5-fold cross-validation, with subsequent training on the RHPE with addition of chronological age. Final model validation was performed on the DHA and an institutional data set of 200 images. The first model, trained on the RSNA, achieved a mean absolute error (MAE) of 3.68 months on the RSNA test set and 5.66 months on the DHA. This outperformed the 4.2 months achieved on the RSNA test set by the best model from previous work (12.4% improvement) and 3.9 months by the open-source software Deeplasia (5.6% improvement). After incorporation of chronological age from the RHPE in model 2, this error improved to an MAE of 4.65 months on the DHA, again surpassing the best previously published models (19.8% improvement). Leveraging newer DL technologies trained on >20,000 hand radiographs across 3 distinct, diverse data sets, this study developed a robust model for predicting bone age. Utilizing features extracted from an RSNA model, combined with chronological age inputs, this model outperforms previous state-of-the-art models when applied to validation data sets. These results indicate that the models provide a highly accurate/reliable platform for clinical use to improve confidence about appropriate surgical selection (eg, physeal-sparing procedures) and time savings for orthopaedic surgeons/radiologists evaluating skeletal age. Development of an accurate DL model for determination of bone age from the hand reduces the time required for age estimation. Additionally, streamlined skeletal age estimation can aid practitioners in determining optimal treatment strategies and may be useful in research settings to decrease workload and improve reporting.

Ultrasound imaging is critical for clinical diagnostics, providing insights into various diseases and organs. However, artificial intelligence (AI) in this field faces challenges, such as the need for large labeled datasets and limited task-specific model applicability, particularly due to ultrasound's low signal-to-noise ratio (SNR). To overcome these, we introduce the Ultrasound Representation Foundation Model (URFM), designed to learn robust, generalizable representations from unlabeled ultrasound images, enabling label-efficient adaptation to diverse diagnostic tasks. URFM is pre-trained on over 1M images from 15 major anatomical organs using representation-based masked image modeling (MIM), an advanced self-supervised learning. Unlike traditional pixel-based MIM, URFM integrates high-level representations from BiomedCLIP, a specialized medical vision-language model, to address the low SNR issue. Extensive evaluation shows that URFM outperforms state-of-the-art methods, offering enhanced generalization, label efficiency, and training-time efficiency. URFM's scalability and flexibility signal a significant advancement in diagnostic accuracy and clinical workflow optimization in ultrasound imaging.

Retrospective study. The study aimed to develop and authenticated a deep learning model based on X-ray images to accurately diagnose fresh thoracolumbar vertebral compression fractures. In clinical practice, diagnosing fresh vertebral compression fractures often requires MRI. However, due to the scarcity of MRI resources and the high time and economic costs involved, some patients may not receive timely diagnosis and treatment. Using a deep learning model combined with X-rays for diagnostic assistance could potentially serve as an alternative to MRI. In this study, the main collection included X-ray images suspected of thoracolumbar vertebral compression fractures from the municipal shared database between December 2012 and February 2024. Deep learning models were constructed using frameworks of EfficientNet, MobileNet, and MnasNet, respectively. We conducted a preliminary evaluation of the deep learning model using the validation set. The diagnostic performance of the models was evaluated using metrics such as AUC value, accuracy, sensitivity, specificity, F1 score, precision, and ROC curve. Finally, the deep learning models were compared with evaluations from two spine surgeons of different experience levels on the control set. This study included a total of 3025 lateral X-ray images from 2224 patients. The data set was divided into a training set of 2388 cases, a validation set of 482 cases, and a control set of 155 cases. In the validation set, the three groups of DL models had accuracies of 83.0%, 82.4%, and 82.2%, respectively. The AUC values were 0.861, 0.852, and 0.865, respectively. In the control set, the accuracies of the three groups of DL models were 78.1%, 78.1%, and 80.7%, respectively, all higher than spinal surgeons and significantly higher than junior spine surgeon. This study developed deep learning models for detecting fresh vertebral compression fractures, demonstrating high accuracy.

Non-contrast CT (NCCT) is widely used in clinical practice and holds potential for large-scale atherosclerosis screening, yet its application in detecting and grading aortic atherosclerosis remains limited. To address this, we propose Aortic-AAE, an automated segmentation system based on a cascaded attention mechanism within the nnU-Net framework. The cascaded attention module enhances feature learning across complex anatomical structures, outperforming existing attention modules. Integrated preprocessing and post-processing ensure anatomical consistency and robustness across multi-center data. Trained on 435 labeled NCCT scans from three centers and validated on 388 independent cases, Aortic-AAE achieved 81.12% accuracy in aortic stenosis classification and 92.37% in Agatston scoring of calcified plaques, surpassing five state-of-the-art models. This study demonstrates the feasibility of using deep learning for accurate detection and grading of aortic atherosclerosis from NCCT, supporting improved diagnostic decisions and enhanced clinical workflows.

The present study examined an integrated multiple neuroimaging modality (T1 structural, Diffusion Tensor Imaging (DTI), and resting-state FMRI (rsFMRI)) to predict aphasia severity using Western Aphasia Battery-Revised Aphasia Quotient (WAB-R AQ) in 76 individuals with post-stroke aphasia. We employed Support Vector Regression (SVR) and Random Forest (RF) models with supervised feature selection and a stacked feature prediction approach. The SVR model outperformed RF, achieving an average root mean square error (RMSE) of 16.38±5.57, Pearson's correlation coefficient (r) of 0.70±0.13, and mean absolute error (MAE) of 12.67±3.27, compared to RF's RMSE of 18.41±4.34, r of 0.66±0.15, and MAE of 14.64±3.04. Resting-state neural activity and structural integrity emerged as crucial predictors of aphasia severity, appearing in the top 20% of predictor combinations for both SVR and RF. Finally, the feature selection method revealed that functional connectivity in both hemispheres and between homologous language areas is critical for predicting language outcomes in patients with aphasia. The statistically significant difference in performance between the model using only single modality and the optimal multi-modal SVR/RF model (which included both resting-state connectivity and structural information) underscores that aphasia severity is influenced by factors beyond lesion location and volume. These findings suggest that integrating multiple neuroimaging modalities enhances the prediction of language outcomes in aphasia beyond lesion characteristics alone, offering insights that could inform personalized rehabilitation strategies.

Cortical morphological abnormalities in schizophrenia (SCZ), major depressive disorder (MDD), and bipolar disorder (BD) have been identified in past research. However, their potential as objective biomarkers to differentiate these disorders remains uncertain. Machine learning models may offer a novel diagnostic tool. Structural MRI (sMRI) of 220 SCZ, 220 MDD, 220 BD, and 220 healthy controls were obtained using a 3T scanner. Volume, thickness, surface area, and mean curvature of 68 cerebral cortices were extracted using FreeSurfer. 272 features underwent 3 feature selection techniques to isolate important variables for model construction. These features were incorporated into 3 classifiers for classification. After model evaluation and hyperparameter tuning, the best-performing model was identified, along with the most significant brain measures. The univariate feature selection-Naive Bayes model achieved the best performance, with an accuracy of 0.66, macro-average AUC of 0.86, and sensitivities and specificities ranging from 0.58-0.86 to 0.81-0.93, respectively. Key features included thickness of right isthmus-cingulate cortex, area of left inferior temporal gyrus, thickness of right superior temporal gyrus, mean curvature of right pars orbitalis, thickness of left transverse temporal cortex, volume of left caudal anterior-cingulate cortex, area of right banks superior temporal sulcus, and thickness of right temporal pole. The machine learning model based on sMRI data shows promise for aiding in the differential diagnosis of SCZ, MDD, and BD. Cortical features from the cingulate and temporal lobes may highlight distinct biological mechanisms underlying each disorder.

Burn injuries represent a significant clinical challenge due to the complexity of accurately assessing burn depth, which directly influences the course of treatment and patient outcomes. Traditional diagnostic methods primarily rely on visual inspection by experienced burn surgeons. Studies report diagnostic accuracies of around 76% for experts, dropping to nearly 50% for less experienced clinicians. Such inaccuracies can result in suboptimal clinical decisions-delaying vital surgical interventions in severe cases or initiating unnecessary treatments for superficial burns. This diagnostic variability not only compromises patient care but also strains health care resources and increases the likelihood of adverse outcomes. Hence, a more consistent and precise approach to burn classification is urgently needed. The objective is to determine whether a multimodal integrated artificial intelligence (AI) system for accurate classification of burn depth can preserve diagnostic accuracy and provide an important resource when used as part of the electronic medical record (EMR). This study used a novel multimodal AI system, integrating digital photographs and ultrasound tissue Doppler imaging (TDI) data to accurately assess burn depth. These imaging modalities were accessed and processed through an EMR system, enabling real-time data retrieval and AI-assisted evaluation. TDI was instrumental in evaluating the biomechanical properties of subcutaneous tissues, using color-coded images to identify burn-induced changes in tissue stiffness and elasticity. The collected imaging data were uploaded to the EMR system (DrChrono), where they were processed by a vision-language model built on GPT-4 architecture. This model received expert-formulated prompts describing how to interpret both digital and TDI images, guiding the AI in making explainable classifications. This study evaluated whether a multimodal AI classifier, designed to identify first-, second-, and third-degree burns, could be effectively applied to imaging data stored within an EMR system. The classifier achieved an overall accuracy of 84.38%, significantly surpassing human performance benchmarks typically cited in the literature. This highlights the potential of the AI model to serve as a robust clinical decision support tool, especially in settings lacking highly specialized expertise. In addition to accuracy, the classifier demonstrated strong performance across multiple evaluation metrics. The classifier's ability to distinguish between burn severities was further validated by the area under the receiver operating characteristic: 0.97 for first-degree, 0.96 for second-degree, and a perfect 1.00 for third-degree burns, each with narrow 95% CIs. The storage of multimodal imaging data within the EMR, along with the ability for post hoc analysis by AI algorithms, offers significant advancements in burn care, enabling real-time burn depth prediction on currently available data. Using digital photos for superficial burns, easily diagnosed through physical examinations, reduces reliance on TDI, while TDI helps distinguish deep second- and third-degree burns, enhancing diagnostic efficiency.

[¹⁸F]-meta-fluorobenzylguanidine ([¹⁸F]MFBG) PET/CT is a promising imaging modality for neural crest-derived tumors, particularly neuroblastoma. Accurate interpretation necessitates an understanding of normal biodistribution and variations in physiological uptake. This study aimed to systematically characterize the physiological distribution and variability of [¹⁸F]MFBG uptake in pediatric patients to enhance clinical interpretation and differentiate normal from pathological uptake. We retrospectively analyzed [¹⁸F]MFBG PET/CT scans from 169 pediatric neuroblastoma patients, including 20 in confirmed remission, for detailed biodistribution analysis. Organ uptake was quantified using both manual segmentation and deep learning(DL)-based automatic segmentation methods. Patterns of physiological uptake variants were categorized and illustrated using representative cases. [¹⁸F]MFBG demonstrated consistent physiological uptake in the salivary glands (SUVmax 9.8 ± 3.3), myocardium (7.1 ± 1.7), and adrenal glands (4.6 ± 0.9), with low activity in bone (0.6 ± 0.2) and muscle (0.8 ± 0.2). DL-based analysis confirmed uniform, mild uptake across vertebral and peripheral skeletal structures (SUVmean 0.47 ± 0.08). Three physiological liver uptake patterns were identified: uniform (43%), left-lobe predominant (31%), and marginal (26%). Asymmetric uptake in the pancreatic head, transient brown adipose tissue activity, gallbladder excretion, and symmetric epiphyseal uptake were also recorded. These variants were not associated with structural abnormalities or clinical recurrence and showed distinct patterns from pathological lesions. This study establishes a reference for normal [¹⁸F]MFBG biodistribution and physiological variants in children. Understanding these patterns is essential for accurate image interpretation and the avoidance of diagnostic pitfalls in pediatric neuroblastoma patients.

Accurate, noninvasive diagnosis of compensated advanced chronic liver disease (cACLD) is essential for effective clinical management but remains challenging. This study aimed to develop a deep learning-based radiomics model using international multicenter data and to evaluate its performance by comparing it to the two-dimensional shear wave elastography (2D-SWE) cut-off method covering multiple countries or regions, etiologies, and ultrasound device manufacturers. This retrospective study included 1937 adult patients with chronic liver disease due to hepatitis B, hepatitis C, or metabolic dysfunction-associated steatotic liver disease. All patients underwent 2D-SWE imaging and liver biopsy at 17 centers across China, Japan, and Europe using devices from three manufacturers (SuperSonic Imagine, General Electric, and Mindray). The proposed generalized deep learning radiomics of elastography model integrated both elastographic images and liver stiffness measurements and was trained and tested on stratified internal and external datasets. A total of 1937 patients with 9472 2D-SWE images were included in the statistical analysis. Compared to 2D-SWE, the model achieved a higher area under the receiver operating characteristic curve (AUC) (0.89 vs 0.83, P = 0.025). It also achieved a highly consistent diagnosis across all subanalyses (P values: 0.21-0.91), whereas 2D-SWE exhibited different AUCs in the country or region (P < 0.001) and etiology (P = 0.005) subanalyses but not in the manufacturer subanalysis (P = 0.24). The model demonstrated more accurate and robust performance in noninvasive cACLD diagnosis than 2D-SWE across different countries or regions, etiologies, and manufacturers.

This study evaluates the efficiency, accuracy, and cost-effectiveness of radiology reporting using audio multimodal large language models (LLMs) compared to conventional reporting with speech recognition software. We hypothesized that providing minimal audio input would enable a multimodal LLM to generate complete radiological reports. 480 reports from 80 retrospective multimodal imaging studies were reported by two board-certified radiologists using three workflows: conventional workflow (C-WF) with speech recognition software to generate findings and impressions separately and LLM-based workflow (LLM-WF) using the state-of-the-art LLMs GPT-4o and Claude Sonnet 3.5. Outcome measures included reporting time, corrections and personnel cost per report. Two radiologists assessed formal structure and report quality. Statistical analysis used ANOVA and Tukey's post hoc tests (p < 0.05). LLM-WF significantly reduced reporting time (GPT-4o/Sonnet 3.5: 38.9 s ± 22.7 s vs. C-WF: 88.0 s ± 60.9 s, p < 0.01), required fewer corrections (GPT-4o: 1.0 ± 1.1, Sonnet 3.5: 0.9 ± 1.0 vs. C-WF: 2.4 ± 2.5, p < 0.01), and lowered costs (GPT-4o: $2.3 ± $1.4, Sonnet 3.5: $2.4 ± $1.4 vs. C-WF: $3.0 ± $2.1, p < 0.01). Reports generated with Sonnet 3.5 were rated highest in quality, while GPT-4o and conventional reports showed no difference. Multimodal LLMs can generate high-quality radiology reports based solely on minimal audio input, with greater speed, fewer corrections, and reduced costs compared to conventional speech-based workflows. However, future implementation may involve licensing costs, and generalizability to broader clinical contexts warrants further evaluation. Question Comparing time, accuracy, cost, and report quality of reporting using audio input functionality of GPT-4o and Claude Sonnet 3.5 to conventional reporting with speech recognition. Findings Large language models enable radiological reporting via minimal audio input, reducing turnaround time and costs without quality loss compared to conventional reporting with speech recognition. Clinical relevance Large language model-based reporting from minimal audio input has the potential to improve efficiency and report quality, supporting more streamlined workflows in clinical radiology.