Fibroadenoma is a common benign breast disease that affects women of all ages. Early diagnosis can greatly improve the treatment outcomes and reduce the associated pain. Computer-aided diagnosis (CAD) has great potential to improve diagnosis accuracy and efficiency. However, its application in sonography is limited. A network that utilizes expansive receptive fields and local information learning was proposed for the accurate segmentation of breast fibroadenomas in sonography. The architecture comprises the Hierarchical Attentive Fusion module, which conducts local information learning through channel-wise and pixel-wise perspectives, and the Residual Large-Kernel module, which utilizes multiscale large kernel convolution for global information learning. Additionally, multiscale feature fusion in both modules was included to enhance the stability of our network. Finally, an energy function and a data augmentation method were incorporated to fine-tune low-level features of medical images and improve data enhancement. The performance of our model is evaluated using both our local clinical dataset and a public dataset. Mean pixel accuracy (MPA) of 93.93% and 86.06% and mean intersection over union (MIOU) of 88.16% and 73.19% were achieved on the clinical and public datasets, respectively. They are significantly improved over state-of-the-art methods such as SegFormer (89.75% and 78.45% in MPA and 83.26% and 71.85% in MIOU, respectively). The proposed feature extraction strategy, combining local pixel-wise learning with an expansive receptive field for global information perception, demonstrates excellent feature learning capabilities. Due to this powerful and unique local-global feature extraction capability, our deep network achieves superior segmentation of breast fibroadenoma in sonography, which may be valuable in early diagnosis.

UNASSIGNED: Rotator cuff injuries and tears are common causes of shoulder pain and dysfunction, necessitating accurate diagnostic methods to guide clinical decision-making. This study evaluates the diagnostic utility of three-dimensional (3D) shoulder sonography in identifying rotator cuff injury and tear patterns.
UNASSIGNED: A comprehensive search across seven electronic databases, which included Cochrane Library, Embase, PubMed, Cochrane Library, China Biology Medicine (CBM) database, CNKI, Wanfang, and VIP database. These databases were utilized to retrieve articles that assess the diagnostic value of 3D shoulder sonography for identifying rotator cuff injuries and tear patterns. The effectiveness of 3D shoulder sonography was assessed in terms of sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR). For each parameter, the 95% confidence intervals were calculated. Additionally, summary receiver operating characteristic curves (SROCs) were constructed, allowing for a comprehensive evaluation of diagnostic accuracy, which is reflected in the area under the SROC curve (AUC).
UNASSIGNED: Screening of 8,508 identified nine literatures eligible for inclusion in the meta-analysis, encompassing a total of 366 patients. The analysis of detecting any rotator cuff tear revealed a sensitivity of 0.97 and specificity of 0.87, yielding a DOR of 90.03 and an AUC of 0.98. Furthermore, 3D shoulder sonography demonstrated satisfactory accuracy in detecting both full and partial-thickness rotator cuff tears (Sensitivity: 0.92 vs. 0.83, specificity: 0.94 vs. 097, and AUC: 0.96 vs. 0.95).
UNASSIGNED: This study indicates that three-dimensional sonography has satisfied accuracy for detecting rotator cuff tears.

OBJECTIVE: This study aims to investigate the use of high-frequency sonography as a tool for detecting inflammatory and destructive changes in the hand and foot joints of patients with early and long-term RA.
METHODS: This study employs a prospective cohort design involving 162 patients diagnosed with Rheumatoid arthritis (RA) who meet the American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria. Patients were divided into two groups based on disease duration: Group 1 (n = 74) included patients with a disease duration of up to 2 years, or early РА (ERA;), Group 2 (n = 88) consisted of patients with a disease duration exceeding 2 years, or long-term persistent РА (LtRA). All patients underwent a clinical assessment of their joints, as well as radiography and arthrosonography, at the beginning of the study and again at 6 and 12 months later.
RESULTS: In the general group of patients, ultrasound examination revealed signs of synovitis in the joints of the hands more frequently (66%) compared with clinical examination (56% by a number of swollen joints [NSJ] and 55% by a number of painful joints [NPJ], P < .01). After 6 months of treatment, 12% of the patients achieved full US remission and 24% achieved partial US remission.
CONCLUSIONS: Within the scope of comprehensive RA diagnostics, arthrosonography of the joints of the hands and feet, utilizing a combination of greyscale and power Doppler, may surpass radiography in detecting early RA. This method allows for a more accurate assessment of disease activity and progression rates.

UNASSIGNED: Monitoring sensitivity of sonography in focused ultrasound ablation surgery (FUAS) is limited (no hyperechoes in ∼50% of successful coagulation in uterine fibroids). A more accurate and sensitive approach is required.
UNASSIGNED: The echo amplitudes of the focused ultrasound (FUS) transducer in a testing mode (short pulse duration and low power) were found to correlate with the ex vivo coagulation. To further evaluate its coagulation prediction capabilities, in vivo experiments were carried out. The liver, kidney, and leg muscles of three adult goats were treated using clinical FUAS settings, and the echo amplitude of the FUS transducer and grayscale in sonography before and after FUAS were collected. On day 7, animals were sacrificed humanely, and the treated tissues were dissected to expose the lesion. Echo amplitude changes and lesion areas were analyzed statistically, as were the coagulation prediction metrics.
UNASSIGNED: The echo amplitude changes of the FUS transducer correlate well with the lesion areas in the liver (R = 0.682). Its prediction in accuracy (94.4% vs. 50%), sensitivity (92.9% vs. 35.7%), and negative prediction (80% vs. 30.8%) is better than sonography, but similar in specificity (80% vs. 100%) and positive prediction (100% vs. 100%). In addition, the correlation between tissue depth and the lesion area is not good (|R| < 0.2). Prediction performances in kidney and leg muscles are similar.
UNASSIGNED: The FUS echo amplitudes are sensitive to the tissue properties and their changes after FUAS. They are sensitive and reliable in evaluating and predicting FUAS outcomes.

BACKGROUND: Breast fibroadenoma poses a significant health concern, particularly for young women. Computer-aided diagnosis has emerged as an effective and efficient method for the early and accurate detection of various solid tumors. Automatic segmentation of the breast fibroadenoma is important and potentially reduces unnecessary biopsies, but challenging due to the low image quality and presence of various artifacts in sonography.
METHODS: Human learning involves modularizing complete information and then integrating it through dense contextual connections in an intuitive and efficient way. Here, a human learning paradigm was introduced to guide the neural network by using two consecutive phases: the feature fragmentation stage and the information aggregation stage. To optimize this paradigm, three fragmentation attention mechanisms and information aggregation mechanisms were adapted according to the characteristics of sonography. The evaluation was conducted using a local dataset comprising 600 breast ultrasound images from 30 patients at Suining Central Hospital in China. Additionally, a public dataset consisting of 246 breast ultrasound images from Dataset_BUSI and DatasetB was used to further validate the robustness of the proposed network. Segmentation performance and inference speed were assessed by Dice similarity coefficient (DSC), Hausdorff distance (HD), and training time and then compared with those of the baseline model (TransUNet) and other state-of-the-art methods.
RESULTS: Most models guided by the human learning paradigm demonstrated improved segmentation on the local dataset with the best one (incorporating C3ECA and LogSparse Attention modules) outperforming the baseline model by 0.76% in DSC and 3.14 mm in HD and reducing the training time by 31.25%. Its robustness and efficiency on the public dataset are also confirmed, surpassing TransUNet by 0.42% in DSC and 5.13 mm in HD.
CONCLUSIONS: Our proposed human learning paradigm has demonstrated the superiority and efficiency of ultrasound breast fibroadenoma segmentation across both public and local datasets. This intuitive and efficient learning paradigm as the core of neural networks holds immense potential in medical image processing.

Segmenting breast tumors in ultrasonography is challenging due to the low image quality and presence of artifacts. Radiologists\' studying and diagnosis skills are integrated with artificial intelligence to establish a clinical learning-based deep learning network in order to robustly extract and delineate features of breast fibroadenoma. The spatial local feature contrast (SLFC) module captures overall tumor contours, while the channel recursive gated attention (CRGA) module enhances edge perception through high-dimensional information interaction. Additionally, full-scale feature fusion and enhanced deep supervision are applied to improve model stability and performance. To achieve smoother boundaries, we introduce a new loss function (cosh-smooth) that penalizes and finely tunes tumor edges. Our dataset comprises 1016 clinical ultrasound images of breast fibroadenoma with labeled masks, alongside a publicly available dataset of 246 ones. Segmentation performance is evaluated using the Dice similarity coefficient (DSC) and mean intersection over union (MIOU). Extensive experiments demonstrate that our proposed MS-CFNet outperforms state-of-the-art methods. Compared to TransUNet as a baseline model, MS-CFNet improves by 1.47% in DSC and 2.56% in MIOU. The promising result of MS-CFNet is attributed to the integration of radiologists\' clinical diagnosis procedure and the bionic mindset, enhancing the network\'s ability to recognize and segment breast fibroadenomas effectively.

UNASSIGNED: Induction of labor (IOL) is a common obstetric approach to start or encourage uterine contractions to achieve a vaginal birth. It is recommended when continuing the pregnancy may be more dangerous for the mother or baby. Different ultrasonographic measures, such as cervical length, have been investigated as possible predictors of the outcomes of IOL. This meta-analysis aimed to assess the accuracy of ultrasound measurements in anticipating successful IOL.
UNASSIGNED: The study conducted a thorough search on three databases (PubMed, Scopus, and Web of Science) until 04 March 2023, to find clinical studies published in English that reported different sonographic cervical measures and their ability to predict IOL outcomes. The chosen studies were stratified based on the type of indicator reported, and a meta-analysis was conducted to determine the best indicator for both successful and failed induction. The risk of bias and concerns about the applicability of the included studies was evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) method.
UNASSIGNED: This study analyzed 57 studies with 9,338 patients. Cervical length is moderately effective in predicting successful IOL, with pooled sensitivity (SN) and specificity (SP) of 0.67 and 0.70, respectively. However, cervical length had a pooled SN and SP of 0.70 and 0.61 for predicting failed IOL. The posterior cervical angle was found to have a higher pooled SN and SP of 0.79 and 0.73 for predicting successful IOL. Fetal head-perineum distance demonstrated moderate accuracy with a pooled SN, SP, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and area under the curve of 0.58, 0.66, 1.95, 0.36, 5.33, and 0.9992, respectively, for predicting successful IOL.
UNASSIGNED: Fetal head-perineum distance was the most effective predictor for successful IOL compared to cervical length, which only had a moderate predictive ability. Shortening of cervical length was not a useful indicator for successful IOL. On the other hand, the posterior cervical angle was the most reliable factor for predicting failed induction. The study\'s findings can aid in developing more effective management strategies for IOL.

UNASSIGNED: This study examined the implementation of a Doppler sonography imaging protocol to assess intraneural blood flow, within the median nerve, in healthy individuals.
UNASSIGNED: A total of 176 participants were examined, and this involved 717 retrospective observations of the images collected. The implemented imaging protocol was assessed, and the data that were collected were cleaned and checked for fidelity and validity.
UNASSIGNED: A large percentage of missing evidence (11%-35%) across proximal, mid, and distal carpal tunnel locations. Only a quarter of cases with evidence of intraneural blood flow had the strongest evidence of a power Doppler video clip, of which only three-quarters were valid. The study identified potential areas for improving the imaging protocol to reduce missing data and improve data quality.
UNASSIGNED: This study demonstrates the significance of a standardized imaging protocol to guide the sonographic acquisition of Doppler images and provides important insights into potential issues with data quality. The recommendations have the potential to help future studies assess intraneural blood flow in healthy populations in a more rigorous and reliable way. Incorporating the study\'s recommendations into a standardized protocol, there is potential to enhance the diagnostic accuracy of carpal tunnel syndrome and improve diagnosis and treatment.

Timely diagnosis and management of elevated intracranial pressure (ICP) in patients with traumatic brain injury (TBI) can significantly reduce mortality rates. Ultrasound examination of the optic nerve sheath diameter (ONSD) is considered a potential, noninvasive, and effective method for assessing ICP. We conducted a systematic review and meta-analysis of ONSD ultrasound detection and invasive ICP monitoring methods to compare and evaluate the diagnostic accuracy of ONSD ultrasound detection methods for intracranial hypertension (IH) in patients with TBI.
We searched the Web of Science, PubMed, and Embase databases to assess the diagnostic accuracy of ONSD sonography for predicting increased ICP. The 2 authors independently extracted the collected data. Simultaneously, the QUADAS-2 tool was used to evaluate the bias risk of each study and conducted random-effects meta-analyses for the accuracy and specificity of diagnosis, and calculated pooled estimates.
Ten studies with 512 patients were included. The diagnostic accuracy of ONSD sonography for IH was revealed as a pooled sensitivity of 0.85 (95% confidence interval [CI], 0.79-0.89) and specificity of 0.88 (95% CI, 0.80-0.93), compared with the invasive ICP monitoring standard for patients with TBI.
ONSD sonography may be a useful method for predicting increased ICP in adult patients with TBI. Further clinical studies are required to confirm the diagnostic value of ONSD sonography.

OBJECTIVE: To evaluate the occurrence of retained products of conception (RPOC) after termination of pregnancy in the first trimester and to assess the vascular signals with transvaginal ultrasonography (TVUS) examination in the detection of retained products.
METHODS: A retrospective cohort study was performed using TVUS examination in patients following termination of pregnancy. In cases of RPOC, 3 scales of vascular signal were identified: type 1, no or small amount, spot flow signals; type 2, medium amount, strip-like flow signals; type 3, rich amount, circumferential-like flow signals. The correlation between vascular signals and placenta accreta spectrum (PAS) staging was proposed by sonography and histopathology findings.
RESULTS: The 3 vascular patterns were differently distributed within non-RPOC as well as RPOC patients with and without PAS: type 1 vascular signal detection rates of non-RPOC and RPOC were 97.8% (262/268) and 28.1% (18/64), respectively. Of 64 cases of RPOC, 48.4% (31/64) of the patients had type 2 vascular signals. Vascular signals were enhanced in RPOC with PAS patients whose diagnosis was confirmed by histopathology.
CONCLUSIONS: The vascularity (amount of flow), vascular pattern (spot, strip- or circumferential-like flow), and the flow penetrating myometrium were significant findings for distinguishing concomitant RPOC with and without PAS. Additionally, RPOC may contribute to PAS progression, or PAS and RPOC in coordination strengthen the observed vascular signals.