Ultrasound examinations, while valuable, are time-consuming and often limited in availability. Consequently, many hospitals implement reservation systems; however, these systems typically lack prioritization for examination purposes. Hence, our hospital uses a waitlist system that prioritizes examination requests based on their clinical value when slots become available due to cancellations. This system, however, requires a manual review of examination purposes, which are recorded in free-form text. We hypothesized that artificial intelligence language models could preliminarily estimate the priority of requests before manual reviews. This study aimed to investigate potential challenges associated with using language models for estimating the priority of medical examination requests and to evaluate the performance of language models in processing Japanese medical texts. We retrospectively collected ultrasound examination requests from the waitlist system at Keio University Hospital, spanning January 2020 to March 2023. Each request comprised an examination purpose documented by the requesting physician and a 6-tier priority level assigned by a radiologist during the clinical workflow. We fine-tuned JMedRoBERTa, Luke, OpenCalm, and LLaMA2 under two conditions: (1) tuning only the final layer and (2) tuning all layers using either standard backpropagation or low-rank adaptation. We had 2335 and 204 requests in the training and test datasets post cleaning. When only the final layers were tuned, JMedRoBERTa outperformed the other models (Kendall coefficient=0.225). With full fine-tuning, JMedRoBERTa continued to perform best (Kendall coefficient=0.254), though with reduced margins compared with the other models. The radiologist's retrospective re-evaluation yielded a Kendall coefficient of 0.221. Language models can estimate the priority of examination requests with accuracy comparable with that of human radiologists. The fine-tuning results indicate that general-purpose language models can be adapted to domain-specific texts (ie, Japanese medical texts) with sufficient fine-tuning. Further research is required to address priority rank ambiguity, expand the dataset across multiple institutions, and explore more recent language models with potentially higher performance or better suitability for this task.

To propose a deep learning (DL) system for the preoperative diagnosis of follicular-like thyroid neoplasms (FNs) using routine ultrasound images. Preoperative diagnosis of malignancy in nodules suspicious for an FN remains challenging. Ultrasound, fine-needle aspiration cytology, and intraoperative frozen section pathology cannot unambiguously distinguish between benign and malignant FNs, leading to unnecessary biopsies and operations in benign nodules. This multicenter, retrospective study included 3634 patients who underwent ultrasound and received a definite diagnosis of FN from 11 centers, comprising thyroid follicular adenoma (n=1748), follicular carcinoma (n=299), and follicular variant of papillary thyroid carcinoma (n=1587). Four DL models including Inception-v3, ResNet50, Inception-ResNet-v2, and DenseNet161 were constructed on a training set (n=2587, 6178 images) and were verified on an internal validation set (n=648, 1633 images) and an external validation set (n=399, 847 images). The diagnostic efficacy of the DL models was evaluated against the ACR TI-RADS regarding the area under the curve (AUC), sensitivity, specificity, and unnecessary biopsy rate. When externally validated, the four DL models yielded robust and comparable performance, with AUCs of 82.2%-85.2%, sensitivities of 69.6%-76.0%, and specificities of 84.1%-89.2%, which outperformed the ACR TI-RADS. Compared to ACR TI-RADS, the DL models showed a higher biopsy rate of malignancy (71.6% -79.9% vs 37.7%, P<0.001) and a significantly lower unnecessary FNAB rate (8.5% -12.8% vs 40.7%, P<0.001). This study provides a noninvasive DL tool for accurate preoperative diagnosis of FNs, showing better performance than ACR TI-RADS and reducing unnecessary invasive interventions.

To validate the diagnostic performance of a B-mode ultrasound-based deep learning (DL) model in distinguishing benign and malignant cervical lymphadenopathy (CLP). A total of 210 CLPs with conclusive pathological results were retrospectively included and separated as training (n = 169) or test cohort (n = 41) randomly at a ratio of 4:1. A DL model integrating convolutional neural network, deformable convolution network and attention mechanism was developed. Three diagnostic models were developed: (a) Model I, CLPs with at least one suspicious B-mode ultrasound feature (ratio of longitudinal to short diameter < 2, irregular margin, hyper-echogenicity, hilus absence, cystic necrosis and calcification) were deemed malignant; (b) Model II: total risk score of B-mode ultrasound features obtained by multivariate logistic regression and (c) Model III: CLPs with positive DL output are deemed malignant. The diagnostic utility of these models was assessed by the area under the receiver operating curve (AUC) and corresponding sensitivity and specificity. Multivariate analysis indicated that DL positive result was the most important factor associated with malignant CLPs [odds ratio (OR) = 39.05, p < 0.001], only followed by hilus absence (OR = 6.01, p = 0.001) in the training cohort. In the test cohort, the AUC of the DL model (0.871) was significantly higher than that in model I (AUC = 0.681, p = 0.04) and model II (AUC = 0.679, p = 0.03), respectively. In addition, model III obtained 93.3% specificity, which was significantly higher than that in model I (40.0%, p = 0.002) and model II (60.0%, p = 0.03), respectively. Although the sensitivity of model I was the highest, it did not show a significant difference compared to that of model III (96.2% vs.80.8%, p = 0.083). B-mode ultrasound-based DL is a potentially robust tool for the differential diagnosis of benign and malignant CLPs.

Accurate and efficient simulation of wave equations is crucial in computational wave imaging applications, such as ultrasound computed tomography (USCT), which reconstructs tissue material properties from observed scattered waves. Traditional numerical solvers for wave equations are computationally intensive and often unstable, limiting their practical applications for quasi-real-time image reconstruction. Neural operators offer an innovative approach by accelerating PDE solving using neural networks; however, their effectiveness in realistic imaging is limited because existing datasets oversimplify real-world complexity. In this paper, we present OpenBreastUS, a large-scale wave equation dataset designed to bridge the gap between theoretical equations and practical imaging applications. OpenBreastUS includes 8,000 anatomically realistic human breast phantoms and over 16 million frequency-domain wave simulations using real USCT configurations. It enables a comprehensive benchmarking of popular neural operators for both forward simulation and inverse imaging tasks, allowing analysis of their performance, scalability, and generalization capabilities. By offering a realistic and extensive dataset, OpenBreastUS not only serves as a platform for developing innovative neural PDE solvers but also facilitates their deployment in real-world medical imaging problems. For the first time, we demonstrate efficient in vivo imaging of the human breast using neural operator solvers.

To construct an intelligent analysis model for classifying fetal orientation during intrapartum ultrasound videos based on multi-feature fusion. The proposed model consists of the Input, Backbone Network and Classification Head modules. The Input module carries out data augmentation to improve the sample quality and generalization ability of the model. The Backbone Network was responsible for feature extraction based on Yolov8 combined with CBAM, ECA, PSA attention mechanism and AIFI feature interaction module. The Classification Head consists of a convolutional layer and a softmax function to output the final probability value of each class. The images of the key structures (the eyes, face, head, thalamus, and spine) were annotated with frames by physicians for model training to improve the classification accuracy of the anterior occipital, posterior occipital, and transverse occipital orientations. The experimental results showed that the proposed model had excellent performance in the tire orientation classification task with the classification accuracy reaching 0.984, an area under the PR curve (average accuracy) of 0.993, and area under the ROC curve of 0.984, and a kappa consistency test score of 0.974. The prediction results by the deep learning model were highly consistent with the actual classification results. The multi-feature fusion model proposed in this study can efficiently and accurately classify fetal orientation in intrapartum ultrasound videos.

The assessment of diaphragm function is crucial for effective clinical management and the prevention of complications associated with diaphragmatic dysfunction. However, current measurement methodologies rely on manual techniques that are susceptible to human error: How does the performance of an automatic diaphragm measurement system based on a segmentation neural network focusing on diaphragm thickness and excursion compare with existing methodologies? The proposed system integrates segmentation and parameter measurement, leveraging a newly established ultrasound diaphragm dataset. This dataset comprises B-mode ultrasound images and videos for diaphragm thickness assessment, as well as M-mode images and videos for movement measurement. We introduce a novel deep learning-based segmentation network, the Multi-ratio Dilated U-Net (MDRU-Net), to enable accurate diaphragm measurements. The system additionally incorporates a comprehensive implementation plan for automated measurement. Automatic measurement results are compared against manual assessments conducted by clinicians, revealing an average error of 8.12% in diaphragm thickening fraction measurements and a mere 4.3% average relative error in diaphragm excursion measurements. The results indicate overall minor discrepancies and enhanced potential for clinical detection of diaphragmatic conditions. Additionally, we design a user-friendly automatic measurement system for assessing diaphragm parameters and an accompanying method for measuring ultrasound-derived diaphragm parameters. In this paper, we constructed a diaphragm ultrasound dataset of thickness and excursion. Based on the U-Net architecture, we developed an automatic diaphragm segmentation algorithm and designed an automatic parameter measurement scheme. A comparative error analysis was conducted against manual measurements. Overall, the proposed diaphragm ultrasound segmentation algorithm demonstrated high segmentation performance and efficiency. The automatic measurement scheme based on this algorithm exhibited high accuracy, eliminating subjective influence and enhancing the automation of diaphragm ultrasound parameter assessment, thereby providing new possibilities for diaphragm evaluation.

Previous studies have confirmed that fully automated three-dimensional echocardiography (3DE) right ventricular (RV) quantification software can accurately assess adult RV function. However, data on its accuracy in children are scarce. This study aimed to test the accuracy of the software in children using cardiac magnetic resonance (MR) as the gold standard. This study prospectively enrolled 82 children who underwent both echocardiography and cardiac MR within 24 h. The RV end-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF) were obtained using the novel 3DE-RV quantification software and compared with cardiac MR values across different groups. The novel 3DE-RV quantification software was feasible in all 82 children (100%). Fully automated analysis was achieved in 35% patients with an analysis time of 8 ± 2 s and 100% reproducibility. Manual editing was necessary in the remaining 65% patients. The 3DE-derived RV volumes and EF correlated well with cardiac MR measurements (RVEDV, r=0.93; RVESV, r=0.90; RVEF, r=0.82; all P <0.001). Although the automated approach slightly underestimated RV volumes and overestimated RVEF compared with cardiac MR in the entire cohort, the bias was smaller in children with RVEF ≥ 45%, normal RV size, and good 3DE image quality. Fully automated 3DE-RV quantification software provided accurate and completely reproducible results in 35% children without any adjustment. The RV volumes and EF measured using the automated 3DE method correlated well with those from cardiac MR, especially in children with RVEF ≥ 45%, normal RV size, and good 3DE image quality. Therefore, the novel automated 3DE method may achieve rapid and accurate assessment of RV function in children with normal heart anatomy.

Ultrasound technology enables safe, non-invasive imaging of dynamic tissue behavior, making it a valuable tool in medicine, biomechanics, and sports science. However, accurately tracking tissue motion in B-mode ultrasound remains challenging due to speckle noise, low edge contrast, and out-of-plane movement. These challenges complicate the task of tracking anatomical landmarks over time, which is essential for quantifying tissue dynamics in many clinical and research applications. This manuscript introduces DUSTrack (Deep learning and optical flow-based toolkit for UltraSound Tracking), a semi-automated framework for tracking arbitrary points in B-mode ultrasound videos. We combine deep learning with optical flow to deliver high-quality and robust tracking across diverse anatomical structures and motion patterns. The toolkit includes a graphical user interface that streamlines the generation of high-quality training data and supports iterative model refinement. It also implements a novel optical-flow-based filtering technique that reduces high-frequency frame-to-frame noise while preserving rapid tissue motion. DUSTrack demonstrates superior accuracy compared to contemporary zero-shot point trackers and performs on par with specialized methods, establishing its potential as a general and foundational tool for clinical and biomechanical research. We demonstrate DUSTrack's versatility through three use cases: cardiac wall motion tracking in echocardiograms, muscle deformation analysis during reaching tasks, and fascicle tracking during ankle plantarflexion. As an open-source solution, DUSTrack offers a powerful, flexible framework for point tracking to quantify tissue motion from ultrasound videos. DUSTrack is available at https://github.com/praneethnamburi/DUSTrack.

Artificial intelligence (AI) is already having a significant impact on healthcare. For example, AI-guided imaging can improve the diagnosis/treatment of vascular diseases, which affect over 200 million people globally. Recently, Chiu and colleagues (2024) developed an AI algorithm that supports nurses with no ultrasound training in diagnosing abdominal aortic aneurysms (AAA) with similar accuracy as ultrasound-trained physicians. This technology can therefore improve AAA screening; however, achieving clinical impact with new AI technologies requires careful consideration of commercialization strategies, including funding, compliance with safety and regulatory frameworks, health technology assessment, regulatory approval, reimbursement, and clinical guideline integration.

It is challenging to preoperatively diagnose follicular thyroid carcinoma (FTC) on ultrasound images. This study aimed to develop an end-to-end diagnostic model that can classify thyroid tumors into benign tumors, FTC and other malignant tumors based on deep learning. This retrospective multi-center study included 10,771 consecutive adult patients who underwent conventional ultrasound and postoperative pathology between January 2018 and September 2021. We proposed a novel data augmentation method and a mixed loss function to solve an imbalanced dataset and applied them to a pre-trained convolutional neural network and transformer model that could effectively extract image features. The proposed model can directly identify FTC from other malignant subtypes and benign tumors based on ultrasound images. The testing dataset included 1078 patients (mean age, 47.3 years ± 11.8 (SD); 811 female patients; FTCs, 39 of 1078 (3.6%); Other malignancies, 385 of 1078 (35.7%)). The proposed classification model outperformed state-of-the-art models on differentiation of FTC from other malignant sub-types and benign ones, achieved an excellent diagnosis performance with balanced-accuracy 0.87, AUC 0.96 (95% CI: 0.96, 0.96), mean sensitivity 0.87 and mean specificity 0.92. Meanwhile, it was superior to radiologists included in this study for thyroid tumor diagnosis (balanced-accuracy: Junior 0.60, p < 0.001; Mid-level 0.59, p < 0.001; Senior 0.66, p < 0.001). The developed classification model addressed the class-imbalanced problem and achieved higher performance in differentiating FTC from other malignant subtypes and benign tumors compared with existing methods. Question Deep learning has the potential to improve preoperatively diagnostic accuracy for follicular thyroid carcinoma (FTC). Findings The proposed model achieved high accuracy, sensitivity and specificity in diagnosing follicular thyroid carcinoma, outperforming other models. Clinical relevance The proposed model is a promising computer-aided diagnostic tool for the clinical diagnosis of FTC, which potentially could help reduce missed diagnosis and misdiagnosis for FTC.