Ultrasound is a commonly used imaging modality for breast cancer diagnosis and prognosis but suffers from false positives, false negatives and interobserver variability. Deep learning (DL), a subset of artificial intelligence, has the potential to improve the efficiency and accuracy of breast ultrasound. This article provides a comprehensive overview of DL applications for breast cancer diagnosis and treatment in ultrasound, including methodological descriptions of various DL models, and clinical applications on noise reduction, lesion localization, risk assessment, diagnosis, response evaluation and outcome prediction. Furthermore, the review highlights the importance of interpretability and small sample size learning of DL-based systems in clinical practice; specific recommendations for further expanding the clinical impact of DL-based systems are also provided.

In recent years, ultrasound imaging has become an important means of medical diagnosis because of its safety and radiation-free advantages. With the continuous progress of deep learning, Artificial Intelligence (AI) models can process large amounts of ultrasound data quickly and accurately, providing decision support for clinicians in diagnosis. From the perspective of ultrasound image classification, detection and segmentation, this paper systemically introduces the latest progress of AI technology in ultrasound imaging, and summarizes the recent high-level related work. At the same time, we also discuss the development prospect and bottleneck of AI in ultrasound imaging processing, which provides the future research directions for researchers in related fields.

Artificial intelligence-based pelvic floor ultrasound helps the diagnosis, preoperative assessment, and postoperative monitoring of female pelvic floor dysfunction (FPFD). The application of artificial intelligence in pelvic floor ultrasound mainly includes automatic segmentation and measurement, the diagnosis of muscle injury, childbirth prediction and postoperational evaluation. It can not only overcome the problem of operator experience dependence but also improve work efficiency and simplify the workflow, which has popularized the application of pelvic floor ultrasound. However, most of the current research is still limited to the automatic segmentation of three-dimensional axial plane levator hiatus (LH). The automatic reconstruction, real-time tracking of 3D/4D images and the imaging navigation of pelvic floor surgery remain major challenges for researchers.

Artificial intelligent (AI) based on deep learning has been used in medical imaging analysis for years. Improvements have been made in the diagnosis of various diseases with the help of deep learning. Multimodal medical imaging combines two or more imaging modalities, providing comprehensive diagnostic information of the diseases. However, some modality problems always exist in clinical practice. Recently, AI-based deep learning technologies have realized the modality conversion. Investigations on modality conversion have gradually been reported in order to acquire multimodal information. MRI images could be generated from CT images while ultrasound elastography could be generated from B mode ultrasonography. Continuous researches and development of new technologies around deep learning are still under investigation and provide huge clinical potentials in the future. The purpose of this review is to summarize an overview of the current applications and prospects of deep learning-based modality conversion of medical imaging.

ChatGPT, an artificial intelligence (AI) software developed by OpenAI, is a powerful language model. ChatGPT is expected to perform a variety of tasks in the field of medical writing and publishing, including writing drafts, extracting article abstracts, and embellishing language. At the same time, ChatGPT has technical shortcomings and ethical challenges that have raised concerns. This review summarizes the issues faced by ChatGPT in the field of medical writing and publishing, and provides a reference for the development of standards and systems for the use of AI products such as ChatGPT.

Objective: An observational cross-sectional study to assess sonographic splenomegaly quantitatively in thalassemia patients grouped with respect to transfusion given whole blood vs packed red cells.
Methods: A study was conducted among 330 patients equally divided into two groups, undergoing an abdominal ultrasound examination with a transducer frequency ranging from 3-5 MHz during the period December 2021 to August 2022. An independent t-test was applied to compare the splenic volume in thalassemia patients given whole blood transfusions versus packed red cells transfusions, and Cohen's d was used to indicate the standardized difference between two ultrasound splenic volume means.
Results: The mean splenic volume of the patients who received whole blood cells was 320.62 ± 219.05 cm³, which is greater than the patients who received packed red cells, whose mean was 60.72 ± 58.72 cm³, The splenomegaly was quantitatively assessed in six age groups ranging from 1 to 3 years, 4 to 6 years, 7 to 9 years, 10 to 12 years, 13 to 15 years, and 16-18 years and mean splenic volume in each age group was compared to those receiving whole blood or packed red cells transfusion. there is a statistically significant difference between both transfusion receiving groups, having a larger Cohen’s d size effect of 1.62.
Conclusion: Ultrasound is a reliable imaging modality for assessing splenic volume and linear parameters of the spleen with greater splenomegaly in thalassemia patients with whole blood transfusions than those with packed red cells when quantitatively assessed according to relevant age groups. Thalassemia patients should be transfused packed red cells to delay splenomegaly, that should be assessed sonographically.

ChatGPT is attracting widespread attention from all walks of life with its excellent multi-round dialogue ability and strong user intent understanding ability, triggering a new wave of artificial intelligence. From the perspective of technical analysis, this article sorts out the various related technologies used in the GPT (Generative Pre-training Transformer) series models as well as large-scale multimodal models, which are more powerful and perform better in multiple downstream tasks. Meanwhile, we guide users to use LLM (Large Language Model) along with GPT more scientifically to maximize their potential. Finally, we analyze the application prospect of the GPT as well as the large-scale multimodal models in the medical field, and the problems are discussed from the perspectives of the risks and limitations of large-scale models applied into the medical field.

Since the 1990s, researchers have been seeking approaches for applying artificial intelligence (AI) to prenatal ultrasound. With the breakthrough of cloud computing technology and the development of deep learning technology, AI in prenatal ultrasound has already entered the clinical application stage in recent years. How does AI combine with clinical prenatal ultrasound? Is the clinical application of AI in prenatal ultrasound effective? What can we expect from AI in prenatal ultrasound? This review introduces the latest developments in this field and explores the challenges and opportunities brought by AI to prenatal ultrasound.

Cardiovascular disease (CVD) is one of the ten leading causes of death worldwide. Atherosclerotic disease, which can lead to myocardial infarction and stroke, is the main cause of CVD. The two main ultrasound image phenotypes used to monitor atherosclerotic load are carotid intima-media thickness (IMT) and plaque area (PA). Early segmentation and measurement methods were based on manual or threshold segmentation, snake models, etc. Usually, these methods are semi-automatic and have poor repeatability and accuracy. Segmentation of the carotid intima-media complex (IMC) and plaque in ultrasound based on artificial intelligence can achieve good accuracy. Compared with two-dimensional ultrasound, three-dimensional/four-dimensional ultrasound can provide spatial dynamic vascular information, which is helpful for doctors to evaluate. This study reviews the progress of artificial intelligence (AI) segmentation methods based on machine learning (ML) and deep learning (DL) used in the segmentation of the IMC and plaque as well as the 3D / 4D reconstruction of carotid ultrasound.

Pelvic floor dysfunction (PFD) is a series of diseases with anatomical and/or functional abnormalities of the pelvic organs, which is common in women and can considerably interfere with their quality of life. Imaging is increasingly being used and can contribute towards better understanding, management, and prediction of long-term outcomes in women who suffer from PFD. Of the available techniques such as X-ray, computed tomography, magnetic resonance imaging, and ultrasound, the latter is generally superior for female pelvic floor imaging, especially in the form of transperineal imaging. This technique is safe, cost-effective, simple, widely available, and can provide an overview of the female pelvic floor. This review will outline the basic methodology, introduce recent researches in the field, and provide an overview of likely future utility of this technique in the evaluation of PFD.

Neuromuscular disease includes a wide range of muscular disorders, but it lacks convenient and effective tools for clinical diagnosis and therapeutic monitoring. As a widely used imaging tool, ultrasound can clearly display muscle structure and create basic conditions for accurate image analysis. At present, many studies have tried to obtain information on muscle function and pathological changes by analyzing the features of muscle ultrasound images, and have shown reliable results. However, the minimal changes in muscle structure and image texture are easy to be neglected, and manual segmentation and data analysis are time-consuming tasks. Artificial intelligence (AI) can accurately identify image changes and improve the efficiency of image analysis, and the muscle ultrasonic image analysis model developed based on AI has shown advantages in a large number of research results. This review summarizes the relevant studies of muscle ultrasound imaging and AI in the field of it, including a variety of research based on traditional AI methods or deep learning methods, as well as discusses the clinical significance of ultrasound analysis assisted by AI and the future exploration directions in this field.

Objectives To measure the CSA of the HST musculature measured with ultrasonography in soccer players undergoing ACLR and compare limb differences with healthy controls.

Methods A case-control study was performed with patients after anterior cruciate ligament repair (ACLR) and healthy controls in which cross-sectional areas (CSA) obtained using a model TE7 ultrasound machine (MINDRAY ®, USA) in B mode (4.2 to 13 MHz) with a multifrequency linear array transducer (L12-4S). Three CSA images were taken of the semitendinosus muscle (ST) and the long head of the biceps femoris (BFlh), at a distance of 30% and 70% of the ischial tuberosity insertion. Mean differences between groups were analyzed using SPSS v.20 (IBM®, USA), and statistical analyses were performed using non-parametric techniques to determine differences between groups (Student's t-test) and Cohen's correlation coefficient to quantify effect size.

Results 14 ACLR operated 17 ± 5.4 months ago and 12 healthy controls (W = 6; M = 20M; 24.5 ± 3.92 years; BMI = 25.1 ± 2.32 kg/m²) were recruited. There were differences between groups in CSA-ST70 (Post-ACLR = 1.43 ± 1.029 cm² vs Control 2.65 ± 0.664 cm², T Student = -3.68, 95% CI [-Inf, -0.648], P < 0. 001, ES = -1.418), but not in CSA-ST30 (Post-ACLR = 8.42 ± 1.596 cm² vs Control 9.16 ± 0.945 cm², T Student = -1.535; 95% CI [-Inf, -0.0793], P = 0. 068, ES = -0.5607), CSA-BFlh30 (Post-ACLR = 8.79 ± 1.47 cm² vs Control 8.87 ± 2.312 cm², T Student = -0.123; 95% CI [-Inf, 1.1049], P = 0.452, ES = -0. 049) or CSA-BFlh70 (Post-ACLR = 6.91 ± 1.011 cm² vs Control 7.01 ± 1.453 cm², T Student = -0.214; 95% CI [-Inf, 0.6795], P = 0.416, ES = -0.0783).

Conclusion Ultrasound measurement of the CSA can be an image marker to identify muscle weakness or atrophy that predicts functional loss early.

Objective: To evaluate the real-time accuracy of cloud handheld ultrasound system using AI technology in screening carotid plaque.

Methods: 2627 ultrasound images of the carotid artery are collected using the cloud handheld system. Bounding boxes of carotid plaques are labeled by qualified sonographers, and the dataset is trained using a lightweight YOLOv3 model. An additional and separate 390 images are collected and tested using the evaluation metrics average recall (AR), average precision (AP), and frames per second (FPS) for quantifying classification performance and time consumption.

Results: We use a plaque grading definition with a thickness of 1.2-1.5 mm defined as small plaque, 1.5-3 mm as medium plaque, and more than 3 mm thick as large plaque. Our model achieves AP^IoU=0.50 with 96.5%, with AP^large is 79.9%, AP^medium is 90.7%, AP^small is 93.5%; AR^IoU=0.50 is 64.5%, where AR^large is 60.6%, AR^medium is 58.3%, AR^small is 70.8%, and FPS is 33.3.

Conclusion: We establish a framework for data set construction, model selection, training, and testing of carotid ultrasound images and verify the effectiveness of real-time AI technology in the automatic detection of carotid artery plaque.

Objective To evaluate the feasibility and capability of UT in detecting musculoskeletal system lesions in the limbs and to explore its image quality.

Materials and Methods The Institutional Review Board has approved this prospective single-center study. This study included participants with various musculoskeletal and neurologic disorders in the limbs who provided written consent from October 2022 to April 2023. In addition to other radiological examinations (X-rays, CT, or MRI) and conventional handheld ultrasound scans requested by clinicians based on the conditions, each participant also underwent UT scanning using our developed limb ultrasound imaging system during the same period. Four radiologists and ultrasound physicians with more than five years of experience in musculoskeletal diagnostics analyzed the two-dimensional and three-dimensional images of the examination area.

Results Overall, 50 participants were evaluated (mean age, 36 years ± 18 [SD]; 26 males). The conditions included musculoskeletal tumors (n = 10), postoperative follow-up of musculoskeletal tumors (n = 20), peripheral nerve disorders in the limbs (n = 10) and postoperative pain in orthopedic surgery (n = 10). In all UT images, the region of interest was completely displayed, and internal structures such as muscles and nerves were clearly visible. Compared to conventional ultrasound images, the reconstructed three-dimensional images intuitively displayed the relationship between the lesions and surrounding tissues. Furthermore, UT did not exhibit metal artifacts when observing soft tissues around metallic implants, providing more comprehensive soft tissue information and more intuitive stereoscopic images.

Conclusion Clinical results of the UT system have demonstrated its feasibility as an automated and standardized imaging technique for musculoskeletal imaging, providing a new imaging modality for the diagnosis of musculoskeletal diseases in the human body.

Breast cancer is the most common malignancy and the leading cause of death for women. Ultrasound is the main tool for breast cancer screening, but it can be influenced by the subjective factors of sonographers. With the continuous development of medical technology and artificial intelligence (AI), the application of breast ultrasound imaging technology is becoming increasingly widespread. Among them, the application of AI and automated breast volume scanning (ABVS) brings new opportunities and challenges for ultrasound diagnosis of breast diseases, while making breast ultrasound diagnosis more accurate and efficient. This article explores the application and prospects of AI and ABVS in ultrasound diagnosis of breast diseases.

Interventional ultrasound (IUS) is an important branch of modern minimally invasive medicine that has been widely applied in clinical practice due to its unique techniques and advantages. As a relatively emerging field, IUS has progressed towards standardization, precision, intelligence, and cutting-edge directions alone with more than 40 years of development, which is becoming increasingly important techniques in clinical medicine. This article will briefly review the development and advancement of IUS for diagnosis and treatment in China in the era of precision medicine from the aspects of artificial intelligence, virtual navigation, molecular imaging, and nanotechnology.

Objective: Computed tomography (CT) imaging of the chest is an effective diagnostic tool assisting physicians in making a diagnosis. This study aimed to propose a new convolutional neural network for classifying the lung nodules of the patient through chest CT scan data to determine whether the patient has related disease genes.
Methods: We proposed a DenseNet-based neural network structure that uses multi-scale convolutional kernels to obtain features of different receptive fields, which are fed into a DenseNet containing four improved DenseBlocks, followed by a classification module to obtain the model output, i.e., whether a lung nodule contains a cancer gene. We conducted classification experiments on a CT scan dataset containing 465 training samples and 117 test samples.
Results: The results showed that DenseNet was better than ResNet in terms of classification, whereas ResNet was better than VGG, which was consistent with the findings of previous studies. However, because these models were more complex, they suffered from overfitting problems. Among all of the models used in this paper, our proposed network achieved the best results in terms of accuracy, F1 score, and sensitivity without an over fitting. The accuracy was 72.0%, sensitivity was 78%, and F1 score was 68%.
Conclusion: The proposed DenseNet neural network can improve and assist medical imaging diagnostic physicians in the initial diagnosis of lung nodules.

Background and Purpose: Surgery of ultrasound-detected non-palpable breast lesions should be guided by ultrasound. Commonly radiologists localize the lesion under ultrasound preoperatively, which necessitates the availability of a localization device and may involve a substantial cost. We performed a study to prospectively assess the feasibility of ultrasound-guided localization without any special device.
Methods: Women with non-palpable benign breast masses were assigned to the “guide wire insertion” (GWI) or the “local coordinates and skin inking” (LOCSI) groups. In both groups, the tumor was marked as a shadow on the skin by the radiologist under ultrasound. In the GWI group, a guidewire was inserted, and in the LOCSI group, the local coordinates of the lesion relative to the skin and the nipple as well as its clockwise placement were reported.
Results: Overall, 29 cases were included in the study, 11 in the GWI and 18 in the LOCSI groups. In all cases, the specimen was correctly excised. The weights of the resected specimens were significantly higher with GWI; LOCSI prevented excessive tissue extraction. Clinicians reported LOCSI as “very easy” more frequently, and surgery took less time.
Conclusions: Overall, our study showed that LOCSI was feasible and can be a suitable method in areas with limited resources. We propose similar studies with a larger sample size, inclusion of malignant cases for margin assessment, and estimation of the cost-effectiveness of the technique.

Objective: General anesthesia (GA) can decrease cerebral flow velocities and predispose patients to cerebral hyperperfusion syndrome (CHS) and other perioperative adverse events after carotid endarterectomy (CEA). The aim of this study was to investigate whether decreased pre-operative flow velocity is associated with an increased risk of CHS and perioperative cerebral infarct, and to further identify risk factors if there is any.
Methods: We retrospectively evaluated 920 consecutive patients who received CEA from 2010 to 2020 at a major academic hospital in China. Middle cerebral artery (MCA) blood flow velocities were measured before and after induction of the GA by transcranial Doppler (TCD). Patients were classified into two groups: the NORMAL group if flow velocity decreased<30% and the LOW group if flow velocity decreased ≥30%. The ultrasonographic diagnostic criterion of CHS was defined as the 100% increase in flow velocity by TCD from the baseline to post-CEA. The occurrence of CHS, perioperative cerebral infarction was compared between the two groups.
Results: 399 (43.4%) were classified as LOW measurement, and 521 (56.6%) patients were classified as NORMAL measurement. In the LOW group, there were more patients with diabetes, fewer patients with ipsilateral ICA severe stenosis and the opening of anterior/posterior communicating artery. Although the occurrence of CHS per ultrasonography criteria was higher in the LOW group (21.3% vs 15.7%, P = 0.03), the occurrence of CHS per clinical criteria (3.2%, vs 2.1%, P = 0.28) or the perioperative cerebral infarct between the two groups (5.8% vs 5.0%, P = 0.60) is equivalent.
Conclusion: Patients with decreased flow velocities post-GA were more likely to meet the ultrasonography criteria for CHS, but they are not at risk of developing clinical CHS or perioperative cerebral infarct.

With the development of network technology and intelligent robot technology, Robot-assisted teleultrasound has played an important role in clinical fields. The application of real-time remote ultrasound technology has made the ultrasonic diagnosis break through the limitation of time and space distance, and solved the problem of shortage of medical resources to a certain extent. This article introduces the development and application basis of robot-assisted teleultrasound, summarizes the clinical application status, and discusses the advantages and limitations of its current application. In addition, we discuss the value in application scenario, interventional therapy and intracavitary ultrasound in the future.

Ultrasound intelligent diagnosis is an emerging technology that combines artificial intelligence (AI) and medical ultrasonography. It has gained significant attention in recent years due to its potential to improve the accuracy and efficiency of medical diagnosis. The core elements of ultrasound artificial intelligence are the construction of data and algorithm models. Therefore, developing autonomous and controllable models, algorithms, and data platforms is extremely important. In this paper, we provide a comprehensive review of the current state-of-the-art in ultrasound intelligent diagnosis including the aspects of the construction of ultrasonic database, deep learning techniques in ultrasound intelligent diagnosis, and the clinical application of ultrasound-AI products. With continued advancements in AI and ultrasound imaging technology, we believe ultrasound intelligent diagnosis will be a valuable tool in the hands of healthcare professionals, providing them with more accurate and efficient diagnoses and treatment plans in the coming years.

The integration of medical imaging and artificial intelligence (AI) has revolutionized interventional therapy of valvular heart diseases (VHD), owing to rapid development in multimodality imaging and healthcare big data. Medical imaging techniques, such as echocardiography, cardiovascular magnetic resonance (CMR) and computed tomography (CT), play an irreplaceable role in the whole process of pre-, intra- and post-procedural intervention of VHD. Different imaging techniques have unique advantages in different stages of interventional therapy. Therefore, single imaging technique can’t fully meet the requirements of complicated clinical scenarios. More importantly, a single intraoperative image provides only limited vision of the surgical field, which could be a potential source for unsatisfactory prognosis. Besides, the non-negligible inter- and intra-observer variability limits the precise quantification of heart valve structure and function in daily clinical practice. With the help of analysis clustered and regressed by big data and exponential growth in computing power, AI broken grounds in the interventional therapy of VHD, including preoperative planning, intraoperative navigation, and postoperative follow-up. This article reviews the state-of-the-art progress and directions in the application of AI for medical imaging in the interventional therapy of VHD.

ObjectiveThe objective of this study was to develop a sonographic technique using two-dimensional (2D) markers for detecting isolated fetal cleft palate (no cleft lip) and to evaluate the ability of 2D and three-dimensional (3D) sonography to image the normal and abnormal palate.

Methods Seventy-three fetuses with a high risk of cleft palate at 12-39 weeks of gestation were referred for specialist ultrasound. A detailed evaluation of the palate was performed through 2D ultrasound, which revealed the appearance of the palatine line in the sagittal plane; the palate and alveolar ridge in the coronal plane of the fetal face; the horizontal plate of the palatine bone in the axial maxillary plane; and the soft palate in the transverse plane of the cavum pharyngis. Subsequently, 3D ultrasound imaging of the palate was performed in all fetuses. Antenatal diagnoses were compared with postnatal findings or autopsy findings.

Results Visualization of 2D markers was accomplished in all fetuses, and 3D assessment was achieved in 97% of fetuses. Cleft palate was suspected in 16 cases (21.9%), among which 14 were suspected on the basis of both 2D and 3D evaluation, and two were suspected only on the basis of 3D evaluation. A normal palate was observed in 57 fetuses (78.1%). The mean gestational age was 27 weeks (range of 12 weeks to 39 weeks). All 16 fetuses with suspected cleft palate were confirmed by postnatal or autopsy findings, no false-positives were observed, and one case with a bifid uvula was missed among 57 fetuses with a presumed normal palate.

Conclusions The fetal palate can be evaluated with 2D markers and 3D sonography. The detection of isolated cleft palate is more sensitive when 2D markers are present in all four planes.

Objective: To compare the diagnostic sensitivity and consistency of ultrasound-guided attenuation parameter (UGAP) with B-mode ultrasound in nonalcoholic fatty liver disease (NAFLD) patients, and explored their correlation with clinical indicators.
Methods: Patients suspected of NAFLD from July to November 2021 were enrolled in this prospective study. After performing the B-mode ultrasound and UGAP examination, all patients were divided into four groups according to the grade of NAFLD obtained by two modalities, respectively. The diagnostic agreement of the two modalities were evaluated, and the diagnostic sensitivity was compared by the McNemar test. The correlation between clinical indicators and the attenuation coefficient (AC) of UGAP was analyzed by linear regression.
Results: The intraclass correlation coefficient of UGAP was 0.958 (95%CI: 0.943,0.970), while the kappa value of B-mode ultrasound grading was 0.799 (95%CI: 0.686, 0.912). The diagnostic sensitivity of UGAP was higher than that of B-mode ultrasound (99.0% vs. 32%, P < 0.001). BMI and TG can be distinguished in different grades of NAFLD diagnosed by B-mode ultrasound, while BMI, ALT, HDL, and Apo A can be distinguished in different grades of NAFLD diagnosed by UGAP. BMI (r = 0.502, P < 0.001), ALT (r = 0. 396, P < 0.001), TG (r = 0.418, P < 0.001), HDL (r = -0. 359, P < 0.001) and Apo A (r = -0.228, P = 0.020) were linearly correlated with the AC value of UGAP.
Conclusions: Compared with the B-mode ultrasound, UGAP had a higher sensitivity and consistency in diagnosing NAFLD, and correlated well with some laboratory indicators, which may be more valuable in screening and diagnosis of NAFLD.

Diabetic peripheral neuropathy (DPN) is one of the most common chronic complications of diabetes, which can lead to neuropathic pain, foot ulcers, and even disability, and greatly reduces survival. Therefore, early diagnosis and prevention of DPN is of great importance to reduce symptoms and disability rate. Ultrasound elastography is a noninvasive method to evaluate changes in nerve tissue composition by obtaining the elastic modulus of tissue and visually displaying the stiffness in the form of images. This paper summarizes the application progress of ultrasound elastography in the evaluation of peripheral neuropathy in recent years, in order to provide reference for the future clinical application of large samples.

Castleman disease (CD), first described in 1954, is a rare lymphoproliferative disorder. A 20-year-old female patient was treated for a "painless mass below the right ear for 1 year". The imagological examination of the patient showed a soft tissue mass about 2.8 cm in diameter in the right parotid gland. Moreover, the patient underwent a right superficial parotidectomy and facial nerve dissection. The mass in the parotid gland was removed, and CD was confirmed by further pathological examination. Imagological examination contributes to the diagnosis and preoperative evaluation of CD.

The American College of Radiology has implemented the Liver Imaging Reporting and Data System (LI-RADS) to help detect, interpret, and guide the management of suspected lesions on surveillance imaging for hepatocellular carcinoma (HCC) in patients with cirrhosis. The classification of indeterminate nodules with a grading algorithm can be used for multiple imaging modalities (US, CT, and MRI) and incorporates multiple imaging features to appropriately classify observations with different likelihood of being HCC. Contrast-enhanced ultrasound (CEUS) LI-RADS has been fully implemented since 2017. The aim of this pictorial article is to provide a comprehensive review of CEUS LI-RADS utilization, discuss its advantages, and highlight areas for potential improvement.

This article discusses the potential benefits of using virtual reality (VR) technology in the teaching of ultrasonography. VR technology can provide an immersive learning experience, enabling students to interact with simulated environments and practice various tasks. Ultrasonography has the characteristics of convenient, rapid, real-time feedback, and dynamic, and is indispensable in practical clinical disease diagnosis applications. Combining VR and ultrasound technology can provide a unique and effective teaching method for medical students and medical professionals. This article mainly discusses the current situation, advantages, and challenges of virtual reality technology in the teaching of ultrasonography to ensure their successful implementation in an educational environment.

The rapid advancement of 5G technology has opened new possibilities for remote ultrasound education, offering the potential to enhance training, real-time consultation, and quality control for primary ultrasound doctors. The 5G remote ultrasound education has the potential to revolutionize the way primary ultrasound doctors are trained and supported, ultimately leading to improved patient care and outcomes. By understanding the current status and development trends of this cutting-edge educational approach, the medical community can better prepare for and contribute to its ongoing evolution. Looking towards the future, the development trends in 5G remote ultrasound education are expected to revolve around continuous improvement and innovation in educational methods and technologies. This includes the exploration of artificial intelligence and machine learning applications, the expansion of telemedicine and telementoring programs, and the development of personalized learning plans tailored to individual learners' needs. This article aims to offer an overview of the current status and applications of 5G remote ultrasound education, including the development of theoretical courses and network construction within our institutes, and to discuss future trends in this field.

Arteriovenous malformation (AVM) is a kind of congenital endothelial malformation that results from errors in vascular morphogenesis. Here we present a case of maxillofacial AVM in a 19-year-old male for its imaging characteristics on contrast-enhanced ultrasound (CEUS). On conventional ultrasound, the lesion was a large subcutaneous mass with abnormal vasculature. On CEUS, it manifested as a “quick wash-in and slow wash-out” enhancement. Moreover, CEUS showed a much larger range of the lesion than gray-scale imaging, with left ophthalmic artery involved. CEUS enabled real-time observation of the microcirculatory perfusion of the lesion and played a very important role in the detection and evaluation of maxillofacial AVM.

Throughout pregnancy, maternal thyroid-related hormones are transported to the fetus via the placenta to allow normal fetal growth and development and are particularly important in the first and second trimesters of pregnancy. During maternal-fetal transport, in addition to thyroid-related hormones, thyroid-stimulating hormone receptor antibodies and antithyroid drugs can enter the fetus and interfere with development of the fetal thyroid gland and endocrine function, potentially leading to hyperthyroidism or hypothyroidism in the fetus or newborn. Several basic studies have been performed to demonstrate the important role of thyroid-related hormones in fetal and neonatal bone development. Ultrasound can assess neonatal skeletal maturity and bone development safely, rapidly, and effectively. This review aims to communicate the latest knowledge about maternal and fetal thyroid function in both normal and pathological pregnancies and summarize the latest advances in the potential effects of abnormal maternal thyroid function on bone development in the fetus and neonate. Finally, it discusses recent advances in research on ultrasound in the assessment of fetal and neonatal bone development.

Objective To investigate the effectiveness of SMI in evaluating hepatic IRI and detecting its therapy response.

Methods Thirty rats were randomly divided into sham (n = 12), IRI (n = 12), and andrographolide pretreatment (n = 6) groups. SMI, pathological, and biochemical examinations were conducted for the sham and IRI groups at 4 (n = 6) and 24 h (n = 6) after reperfusion, respectively. Two ultrasonologists measured the vascular index (VI). The interobserver agreement was evaluated using the intraclass correlation coefficient (ICC). The rat liver parameters, including Suzuki's score, alanine aminotransferase (ALT), and aspartate aminotransferase (AST), were obtained at different time steps in each group. For the andrographolide pretreatment group, data were obtained at 24 h after reperfusion to further verify the advantage of VI. Parameters were analyzed for correlations and compared between each group at 4 and 24 h.

Results The ICC between two ultrasonologists who measured the VI was 0.912 (95%CI: 0.889-0.940). Suzuki's score and VI were negatively correlated (r = -0.504, P = 0.012). Compared with the sham group, the IRI group showed a significant decrease in the VI at 4 and 24 h after reperfusion [(24.78 ± 3.23) versus (20.22 ± 2.55); (22.67 ± 1.36) versus (19.27 ± 2.23), P < 0.05)]. The VI in the andrographolide pretreatment group was higher than that in the IRI group [(21.90 ± 1.47) versus (19.27 ± 2.23), P <0.05].

Conclusions The VI on SMI can be used as a noninvasive and sensitive index to evaluate hepatic IRI and detect its therapeutic response.

Undescended testis or cryptorchidism is the most common reproductive system disease in children. About 3% of full-term and 30% of premature male infants are born with one or both testicles undescended. Most cryptorchidism is asymptomatic and is occasionally found due to inguinal pain caused by testicular torsion. Herein, we presented a case of pediatric patient with abdominal pain as initial symptom who was found the left inguinal cryptorchidism complicated by testicular torsion, and then spontaneous remission during the color Doppler ultrasonography examination.

With the increasing utilization of semi-thyroidectomy and rapid advancements in ultrasound-guided thermal ablation therapy for the management of papillary thyroid carcinoma (PTC) and PTC cervical lymph node metastasis, ultrasound-guided fine-needle aspiration biopsy (FNAB) has emerged as the predominant approach for the pre-treatment cytopathologic diagnosis of PTC. Numerous expert consensuses and practice guidelines have delineated the acquisition of sufficient, high-quality cellular specimens for cytological examination. However, new challenges keep emerging in the real-world practice of thyroid FNAB, primarily stemming from the perceptions and expertise of physicians or technicians who perform FNAB. The aim of this study was to delineate the key deficiencies in specimen collection during FNAB, elucidate principles of systematic thinking, and propose preventive measures for these issues, along with a range of innovative concepts and technical approaches. Effectively addressing these concerns will enhance FNAB implementation and facilitate advancements in novel therapeutic modalities, such as thermal ablation, to ameliorate prognosis.

Schistosoma japonicum causes intestinal schistosomiasis, which affects mainly the rectum and sigmoid colon but rarely the urinary tract. Here we report a rare case of S. japonicum infection of the bladder and ureter, and describe its clinical features and imaging findings. We also put this case in perspective with the literature on S. japonicum infections in humans.

This review article introduces the main concepts and architectures of deep learning networks for medical imaging tasks, such as classification, detection, segmentation, and generation. It then surveys how deep learning has been applied to ultrasound imaging for various purposes, such as image processing, diagnosis, and workflow enhancement. It covers different organs and body parts that can be imaged by ultrasound, such as liver, breast, thyroid, heart, kidney, prostate, nerve, muscle, and fetus. It also discusses how deep learning can help with view recognition, registration, and quantification, measurement, image registration for interventional guidance, and real-time assistance while scanning. Moreover, it explores how generative AI can be used in the future medical field by leveraging deep learning for ultrasound imaging, such as generating realistic and diverse images, virtual organs/patients with diseases, synthesizing missing or corrupted data and augmenting existing data for training and testing.

The gastrointestinal tract, particularly the colon, represents a rare site of metastatic carcinoma. It is difficult to diagnose because of its nonspecific clinical and imaging features. Herein, we reported the case of a 64-year-old Asian male who presented with a solitary mass on his right colon. We successively performed contrast-enhanced ultrasonography, computed tomography (CT) and colonoscopy. Colonoscopy-guided biopsy revealed inflammatory granulomatous and necrotic tissue. The mass was resected and identified as a metastatic carcinoma of renal cell carcinoma (RCC). The details of the case and review of related articles are presented.

Real-time intelligent segmentation of ultrasound video object is a demanding task in the field of medical image processing and serves as an essential and critical step in image-guided clinical procedures. However, obtaining reliable and accurate medical image annotations often necessitates expert guidance, making the acquisition of large-scale annotated datasets challenging and costly. This presents obstacles for traditional supervised learning methods. Consequently, semi-supervised learning (SSL) has emerged as a promising solution, capable of utilizing unlabeled data to enhance model performance and has been widely adopted in medical image segmentation tasks. However, striking a balance between segmentation accuracy and inference speed remains a challenge for real-time segmentation. This paper provides a comprehensive review of research progress in real-time intelligent semi-supervised ultrasound video object segmentation (SUVOS) and offers insights into future developments in this area.

Early detection of vascular disease is fundamental to the prevention and treatment of systemic vascular lesions. The timely identification of vascular damage can be achieved by comprehensively assessing the structural anomaly and/or functional degeneration of the vasculature. The assessment may to some extent indicate the long-term detrimental effects of cardiovascular disease (CVD) risk factors on vascular health. A key aspect in the evaluation of vascular function is the measurement of arterial stiffness. In 2012, the arterial velocity-pulse index (AVI) and arterial pressure-volume index (API) were introduced, which are noninvasively measured with a brachial cuff, and can reflect the status of arterial stiffness in both the aorta and the brachial artery. A large number of relevant studies have demonstrated the strong associations between AVI/API and various CVD risk factors, underlining the substantial relevance of the indices in CVD risk assessment. In this review, we provide a systematic review of the progresses made in brachial cuff-based measurements of arterial stiffness. In addition, we summarize the results of the recent studies focused on exploring the associations of AVI/API with relevant risk factors as well as their roles in CVD assessment.