Search

This research aimed to improve diagnosis of non-alcoholic fatty liver disease (NAFLD) by deep learning with ultrasound Images and reduce the impact of the professional competence and personal bias of the diagnostician. Three convolutional neural network models were used to classify and identify the ultrasound images to obtain the best network. Then, the features in the ultrasound images were extracted and a new convolutional neural network was created based on the best network. Finally, the accuracy of several networks was compared and the best network was evaluated using AUC. Models of VGG16, ResNet50, and Inception-v3 were individually applied to classify and identify 710 ultrasound images containing NAFLD, demonstrating accuracies of 66.2%, 58.5%, and 59.2%, respectively. To further improve the classification accuracy, two features are presented: the ultrasound echo attenuation coefficient (θ), derived from fitting brightness values within sliding region of interest (ROIs), and the ratio of Doppler effect (ROD), identified through analyzing spots exhibiting the Doppler effect. Then, a multi-input deep learning network framework based on the VGG16 model is established, where the VGG16 model processes ultrasound image, while the fully connected layers handle θ and ROD. Ultimately, these components are combined to jointly generate predictions, demonstrating robust diagnostic capabilities for moderate to severe fatty liver (AUC = 0.95). Moreover, the average accuracy is increased from 64.8% to 77.5%, attributed to the introduction of two advanced features with domain knowledge. This research holds significant potential in aiding doctors for more precise and efficient diagnosis of ultrasound images related to NAFLD. [ABSTRACT FROM AUTHOR]

Aim: To demonstrate a case of Actinomyces infection in which ultrasound was used as a diagnostic aid tool. Materials and Methods: Description of a case of cervicofacial actinomycosis in a patient, whose treatment and long‐term evaluation were performed using ultrasound, a crucial tool to ensure accuracy and real‐time images, without exposing the patient to radiation, contributing to the success of treatment. Results: In the case of actinomycosis, the ultrasound findings consisted of hyperechoic foci surrounded by hypoechoic areas. Conclusions: Ultrasonography plays a broad role in dentistry, being able to assist in diagnosis, guided surgical interventions and post‐treatment monitoring. Furthermore, ultrasound is an accessible and portable examination that can be used at the patient's bedside, especially in critical cases, proving to be a promising tool in clinical management in a hospital environment. [ABSTRACT FROM AUTHOR]

Deep vein thrombosis (DVT) refers to the formation of abnormal blood clots within the inner vascular veins, typically in the legs, posing significant health risks. Traditional treatment involves suctioning the clot, monitored by X-ray angiography, which exposes patients and medical staff to radiation. This study aims to enhance DVT diagnosis and treatment by developing a 3D reconstruction method using B-mode ultrasound, linear 3D interpolation, and a multi-denoising filter approach for improved image segmentation. The research methodology includes ultrasound data acquisition with a B-mode scanner and optical tracking system, followed by 3D volume reconstruction through bin-filling and hole-filling processes. Deep learning techniques are employed to segment the blood clot in ultrasound images, and the thrombus volume is estimated. Experiments were conducted in two scenarios: 3D reconstruction using a 2D ultrasound dataset from a DVT patient and thrombus area determination using artificial datasets with fat-injected balloon phantoms. Results demonstrate the proposed method achieved an accuracy of 0.824, a specificity of 0.583, and a sensitivity of 0.955. Thrombus volume estimation yielded a mean absolute percentage error (MAPE) of 27.5%. The findings indicate that the novel method is eligible to be an alternative to reconstruct thrombus volume and accurately identifies thrombus areas in ultrasound images, offering a safer alternative to traditional X-ray-based methods. [ABSTRACT FROM AUTHOR]

Background Purpose Methods Results Conclusions In clinical anesthesia, precise segmentation of muscle layers from abdominal ultrasound images is crucial for identifying nerve block locations accurately. Despite deep learning advancements, challenges persist in segmenting muscle layers with accurate topology due to pseudo and weak edges caused by acoustic artifacts in ultrasound imagery.To assist anesthesiologists in locating nerve block areas, we have developed a novel deep learning algorithm that can accurately segment muscle layers in abdominal ultrasound images with interference.We propose a comprehensive approach emphasizing the preservation of the segmentation's low‐rank property to ensure correct topology. Our methodology integrates a Semantic Feature Extraction (SFE) module for redundant encoding, a Low‐rank Reconstruction (LR) module to compress this encoding, and an Edge Reconstruction (ER) module to refine segmentation boundaries. Our evaluation involved rigorous testing on clinical datasets, comparing our algorithm against seven established deep learning‐based segmentation methods using metrics such as Mean Intersection‐over‐Union (MIoU) and Hausdorff distance (HD). Statistical rigor was ensured through effect size quantification with Cliff's Delta, Multivariate Analysis of Variance (MANOVA) for multivariate analysis, and application of the Holm‐Bonferroni method for multiple comparisons correction.We demonstrate that our method outperforms other industry‐recognized deep learning approaches on both MIoU and HD metrics, achieving the best outcomes with 88.21%/4.98 (pmax=0.1893$p_{max}=0.1893$) on the standard test set and 85.48%/6.98 (pmax=0.0448$p_{max}=0.0448$) on the challenging test set. The best&worst results for the other models on the standard test set were (87.20%/5.72)&(83.69%/8.12), and on the challenging test set were (81.25%/10.00)&(71.74%/16.82). Ablation studies further validate the distinct contributions of the proposed modules, which synergistically achieve a balance between maintaining topological integrity and edge precision.Our findings validate the effective segmentation of muscle layers with accurate topology in complex ultrasound images, leveraging low‐rank constraints. The proposed method not only advances the field of medical imaging segmentation but also offers practical benefits for clinical anesthesia by improving the reliability of nerve block localization. [ABSTRACT FROM AUTHOR]

Ultrasound imaging is widely used in medical diagnostics due to its non-invasive and real-time capabilities. However, existing methods often overlook the benefits of fractional-order filters for denoising and dehazing. Thus, this work introduces an efficient multi-scale wavelet method for dehazing and denoising ultrasound images using a fractional-order filter, which integrates a guided filter, directional filter, fractional-order filter, and haze removal to the different resolution images generated by a multi-scale wavelet. In the directional filter stage, an eigen-analysis of each pixel is conducted to extract structural features, which are then classified into edges for targeted filtering. The guided filter subsequently reduces speckle noise in homogeneous anatomical regions. The fractional-order filter allows the algorithm to effectively denoise while improving edge definition, irrespective of the edge size. Haze removal can effectively eliminate the haze caused by attenuation. Our method achieved significant improvements, with PSNR reaching 31.25 and SSIM 0.905 on our ultrasound dataset, outperforming other methods. Additionally, on external datasets like McMaster and Kodak24, it achieved the highest PSNR (29.68, 28.62) and SSIM (0.858, 0.803). Clinical evaluations by four radiologists confirmed its superiority in liver and carotid artery images. Overall, our approach outperforms existing speckle reduction and structural preservation techniques, making it highly suitable for clinical ultrasound imaging. [ABSTRACT FROM AUTHOR]

The evaluation of carotid ultrasound image plaque requires a large number of experienced clinicians, and the ultrasound image has the characteristics of blurred boundary and strong noise interference, making the evaluation of plaques time-consuming and laborious. Therefore, a fully automated carotid plaque segmentation method is urgently needed to solve the problem of manpower scarcity. This study proposes a deep neural network model based on Swin-Transformer (Shifted-Windows Transformer) block for the automatic segmentation of carotid plaques. Based on the U-Net(U-Convolutional Network) architecture, the encoding part uses three convolutional blocks for image down-sampling to obtain feature images of different resolution sizes, and then adds six pairs of two consecutive Swin-Transformer blocks for more refined feature extraction. The decoding part up-samples the refined features output by the Swin-Transformer module step by step, and jump-joints them with the feature maps of each resolution level in the encoding part, respectively. The comparison experiments based on the data set of Tong Ren Hospital show that the Dice index of the proposed deep neural network model reaches 0.814 2, which is higher than that of other comparison networks. The results demonstrate that the proposed model can effectively extract the features of carotid ultrasound image plaques and achieve automated and high-precision plaque segmentation. [ABSTRACT FROM AUTHOR]

Ultrasound image examination has become the preferred choice for diagnosing metabolic dysfunction-associated steatotic liver disease (MASLD) due to its non-invasive nature. Computer-aided diagnosis (CAD) technology can assist doctors in avoiding deviations in the detection and classification of MASLD. We propose a hybrid model that integrates the pre-trained VGG16 network with an attention mechanism and a stacking ensemble learning model, which is capable of multi-scale feature aggregation based on the self-attention mechanism and multi-classification model fusion (Logistic regression, random forest, support vector machine) based on stacking ensemble learning. The proposed hybrid method achieves four classifications of normal, mild, moderate, and severe fatty liver based on ultrasound images. Our proposed hybrid model reaches an accuracy of 91.34% and exhibits superior robustness against interference, which is better than traditional neural network algorithms. Experimental results show that, compared with the pre-trained VGG16 model, adding the self-attention mechanism improves the accuracy by 3.02%. Using the stacking ensemble learning model as a classifier further increases the accuracy to 91.34%, exceeding any single classifier such as LR (89.86%) and SVM (90.34%) and RF (90.73%). The proposed hybrid method can effectively improve the efficiency and accuracy of MASLD ultrasound image detection. [ABSTRACT FROM AUTHOR]

It is crucial to preoperatively diagnose lateral cervical lymph node (LN) metastases (LNMs) in papillary thyroid carcinoma (PTC) patients. This study aims to develop deep-learning models for the automatic segmentation and classification of LNM on original ultrasound images. This study included 1000 lateral cervical LN ultrasound images (consisting of 512 benign and 558 metastatic LNs) collected from 728 patients at the Chongqing General Hospital between March 2022 and July 2023. Three instance segmentation models (MaskRCNN, SOLO and Mask2Former) were constructed to segment and classify ultrasound images of lateral cervical LNs by recognizing each object individually and in a pixel-by-pixel manner. The segmentation and classification results of the three models were compared with an experienced sonographer in the test set. Upon completion of a 200-epoch learning cycle, the loss among the three unique models became negligible. To evaluate the performance of the deep-learning models, the intersection over union threshold was set at 0.75. The mean average precision scores for MaskRCNN, SOLO and Mask2Former were 88.8%, 86.7% and 89.5%, respectively. The segmentation accuracies of the MaskRCNN, SOLO, Mask2Former models and sonographer were 85.6%, 88.0%, 89.5% and 82.3%, respectively. The classification AUCs of the MaskRCNN, SOLO, Mask2Former models and sonographer were 0.886, 0.869, 0.90.2 and 0.852 in the test set, respectively. The deep learning models could automatically segment and classify lateral cervical LNs with an AUC of 0.92. This approach may serve as a promising tool to assist sonographers in diagnosing lateral cervical LNMs among patients with PTC. • Accurate segmentation and classification of lateral lymph nodes are paramount for papillary thyroid carcinoma patients. • The segmentation performance of the deep-learning model was superior to the sonographer. • The deep-learning model showed the better diagnosis performance than the sonographer. [ABSTRACT FROM AUTHOR]

Purpose: To develop a deep neural network system for the automatic segmentation and risk stratification prediction of gastrointestinal stromal tumors (GISTs). Methods: A total of 980 ultrasound (US) images from 245 GIST patients were retrospectively collected. These images were randomly divided (6:2:2) into a training set, a validation set, and an internal test set. Additionally, 188 US images from 47 prospective GIST patients were collected to evaluate the segmentation and diagnostic performance of the model. Five deep learning‐based segmentation networks, namely, UNet, FCN, DeepLabV3+, Swin Transformer, and SegNeXt, were employed, along with the ResNet 18 classification network, to select the most suitable network combination. The performance of the segmentation models was evaluated using metrics such as the intersection over union (IoU), Dice similarity coefficient (DSC), recall, and precision. The classification performance was assessed based on accuracy and the area under the receiver operating characteristic curve (AUROC). Results: Among the compared models, SegNeXt‐ResNet18 exhibited the best segmentation and classification performance. On the internal test set, the proposed model achieved IoU, DSC, precision, and recall values of 82.1, 90.2, 91.7, and 88.8%, respectively. The accuracy and AUC for GIST risk prediction were 87.4 and 92.0%, respectively. On the external test set, the segmentation models exhibited IoU, DSC, precision, and recall values of 81.0, 89.5, 92.8, and 86.4%, respectively. The accuracy and AUC for GIST risk prediction were 86.7 and 92.5%, respectively. Conclusion: This two‐stage SegNeXt‐ResNet18 model achieves automatic segmentation and risk stratification prediction for GISTs and demonstrates excellent segmentation and classification performance. [ABSTRACT FROM AUTHOR]

Ultrasound images can be utilized as a non-invasive method in the reproductive program. Striped catfish in various gonadal development were examined with a portable Mindray Model Z60Vet using the Convex probe at a frequency of 5 MHz. The results showed that the male gonads of immature striped catfishes were not distinguished from other visceral organs by using an ultrasonic technique. Ultrasound images in ovaries in stages II and III had significant differences. In ovary ‘s group of stage III, the oocyte’s dimensions on the ultrasound images were twice bigger than the group of stage II was. Before the premilitary injection, eggs displayed small, smooth, black dots, and 12h after the definitive dose, egg diameters raised twice and had brighter dots on the ultrasound images.

Ultrasound imaging is vital for diagnosing carotid artery vascular lesions, highlighting the importance of accurately segmenting lumens in ultrasound images to prevent, diagnose and treat vascular diseases. However, noise artifacts, blood residue and discontinuous lumens significantly affect segmentation accuracy. To achieve accurate lumen segmentation in low-quality images, we propose a novel segmentation algorithm which is guided by an adaptively generated shape prior. To tackle the above challenges, we introduce a shape-prior-based segmentation method for carotid artery lumen walls. The shape prior in this study is adaptively generated based on the evolutionary trend of vessel growth. Shape priors guide and constrain the active contour, resulting in precise segmentation. The efficacy of the proposed model was confirmed using 247 carotid artery ultrasound images, with experimental results showing an average Dice coefficient of 92.38%, demonstrating superior segmentation performance compared to existing mathematical models. Our method can quickly and effectively perform accurate lumen segmentation on low-quality carotid artery ultrasound images, which is of great significance for the diagnosis of cardiovascular and cerebrovascular diseases. [ABSTRACT FROM AUTHOR]

Objective: In clinical practice, thyroid nodules are typically visually evaluated by expert physicians using 2D ultrasound images. Based on their assessment, a fine needle aspiration (FNA) may be recommended. However, visually classifying thyroid nodules from ultrasound images may lead to unnecessary fine needle aspirations for patients. The aim of this study is to develop an automatic thyroid ultrasound image classification system to prevent unnecessary FNAs. Methods: An automatic computer-aided artificial intelligence system is proposed for classifying thyroid nodules using a fine-tuned deep learning model based on the DenseNet architecture, which incorporates an attention module. The dataset comprises 591 thyroid nodule images categorized based on the Bethesda score. Thyroid nodules are classified as either requiring FNA or not. The challenges encountered in this task include managing variability in image quality, addressing the presence of artifacts in ultrasound image datasets, tackling class imbalance, and ensuring model interpretability. We employed techniques such as data augmentation, class weighting, and gradient-weighted class activation maps (Grad-CAM) to enhance model performance and provide insights into decision making. Results: Our approach achieved excellent results with an average accuracy of 0.94, F1-score of 0.93, and sensitivity of 0.96. The use of Grad-CAM gives insights on the decision making and then reinforce the reliability of the binary classification for the end-user perspective. Conclusions: We propose a deep learning architecture that effectively classifies thyroid nodules as requiring FNA or not from ultrasound images. Despite challenges related to image variability, class imbalance, and interpretability, our method demonstrated a high classification accuracy with minimal false negatives, showing its potential to reduce unnecessary FNAs in clinical settings. [ABSTRACT FROM AUTHOR]

Muscle dysfunction and muscle atrophy are common complications resulting from Chronic Obstructive Pulmonary Disease (COPD). The evaluation of the peripheral muscles can be carried out through the assessment of their structural components from ultrasound images or their functional components through isometric and isotonic strength tests. This evaluation, performed mainly on the quadriceps muscle, is not only of great interest for diagnosis, prognosis and monitoring of COPD, but also for the evaluation of the benefits of therapeutic interventions. In this work, bioimpedance spectroscopy technology is proposed as a low-cost and easy-to-use alternative for the evaluation of peripheral muscles, becoming a feasible alternative to ultrasound images and strength tests for their application in routine clinical practice. For this purpose, a laboratory prototype of a bioimpedance device has been adapted to perform segmental measurements in the quadriceps region. The validation results obtained in a pseudo-randomized study in patients with COPD in a controlled clinical environment which involved 33 volunteers confirm the correlation and correspondence of the bioimpedance parameters with respect to the structural and functional parameters of the quadriceps muscle, making it possible to propose a set of prediction equations. The main contribution of this manuscript is the discovery of a linear relationship between quadriceps muscle properties and the bioimpedance Cole model parameters, reaching a correlation of 0.69 and an average error of less than 0.2 cm regarding the thickness of the quadriceps estimations from ultrasound images, and a correlation of 0.77 and an average error of 3.9 kg regarding the isometric strength of the quadriceps muscle. [ABSTRACT FROM AUTHOR]

Cirrhosis is a liver disease resulting from abnormal continuation of fibrosis, and ultrasound imaging is widely used for cirrhosis diagnosis because of its non‐invasiveness. However, due to unclear appearances of cirrhosis on ultrasound images, diagnoses are difficult and individual results possibly differ depending on the physician's experience. Recently, computer‐aided diagnostic systems using image processing and machine learning have been developed to help physicians detect cirrhosis as a 'Second opinion'. Some related studies have focused on a scenario where physicians set ROIs (Region of Interests) manually because selecting reliable ROIs for training a classifier and classification of patients is indispensable. But, the accuracy of such systems depends inherently on the quality of ROIs, and thus the workloads of physicians increase. In this paper, we propose a reliability evaluation method (REM) for each ROI based on its posterior probability and relationship to peripheral ROIs. The assumption of our proposal is that reliable regions of cirrhosis and normal can be observed in certain regions predominantly. We evaluated the effectiveness of the REM and its optimization for practical use. Experimental results showed that our proposed method curated reliable ROIs and improved classification performance in terms of AUC (Area Under the Curve). © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Speckle noise in ultrasound images (UIs) significantly reduces the accuracy of disease diagnosis. The aim of this study was to quantitatively evaluate its feasibility in salivary gland ultrasound imaging by modeling the adaptive non-local means (NLM) algorithm. UIs were obtained using an open-source device provided by SonoSkills and FUJIFILM Healthcare Europe. The adaptive NLM algorithm automates optimization by modeling the isotropic search window, eliminating the need for manual configuration in conventional NLM methods. The coefficient of variation (COV), contrast-to-noise ratio (CNR), and edge rise distance (ERD) were used as quantitative evaluation parameters. UIs of the salivary glands revealed evident visualization of the internal echo shape of the malignant tumor and calcification line using the adaptive NLM algorithm. Improved COV and CNR results (approximately 4.62 and 2.15 times, respectively) compared with noisy images were achieved. Additionally, when the adaptive NLM algorithm was applied to the UIs of patients with salivary gland sialolithiasis, the noisy images and ERD values were calculated almost similarly. In conclusion, this study demonstrated the applicability of the adaptive NLM algorithm in optimizing search window parameters for salivary gland UIs. [ABSTRACT FROM AUTHOR]

With the increasing dominance of artificial intelligence (AI) techniques, the important prospects for their application have extended to various medical fields, including domains such as in vitro diagnosis, intelligent rehabilitation, medical imaging, and prognosis. Breast cancer is a common malignancy that critically affects women’s physical and mental health. Early breast cancer screening—through mammography, ultrasound, or magnetic resonance imaging (MRI)—can substantially improve the prognosis for breast cancer patients. AI applications have shown excellent performance in various image recognition tasks, and their use in breast cancer screening has been explored in numerous studies. This paper introduces relevant AI techniques and their applications in the field of medical imaging of the breast (mammography and ultrasound), specifically in terms of identifying, segmenting, and classifying lesions; assessing breast cancer risk; and improving image quality. Focusing on medical imaging for breast cancer, this paper also reviews related challenges and prospects for AI.

Objective: To develop a ultrasound images based dual-channel deep learning model to achieve accurate early diagnosis of thyroid nodules less than 1 cm. Methods: A dual-channel deep learning model called thyroid nodule transformer network (TNT-Net) was proposed. The model has two input channels for transverse and longitudinal ultrasound images of thyroid nodules, respectively. A total of 9649 nodules from 8455 patients across five hospitals were retrospectively collected. The data were divided into a training set (8453 nodules, 7369 patients), an internal test set (565 nodules, 512 patients), and an external test set (631 nodules, 574 patients). Results: TNT-Net achieved an area under the curve (AUC) of 0.953 (95 % confidence interval (CI): 0.934, 0.969) on the internal test set and 0.941 (95 % CI: 0.921, 0.957) on the external test set, significantly outperforming traditional deep convolutional neural network models and single-channel swin transformer model, whose AUCs ranged from 0.800 (95 % CI: 0.759, 0.837) to 0.856 (95 % CI: 0.819, 0.881). Furthermore, feature heatmap visualization showed that TNT-Net could extract richer and more energetic malignant nodule patterns. Conclusion: The proposed TNT-Net model significantly improved the recognition capability for thyroid nodules with size less than 1 cm. This model has the potential to reduce overdiagnosis and overtreatment of such nodules, providing essential support for precise management of thyroid nodules while complementing fine-needle aspiration biopsy.

This paper presents a multi-scale hybrid method for speckle reduction in ultrasound (US) images. Speckle is removed by guided filtering, directional filtering, and fractional order filtering on the coarse to fine resolution images of a wavelet pyramid. For the directional filter, the eigen-analysis of each pixel is firstly carried out to obtain its structural features, and then it is classified into edges for filtering. Speckle noise, corresponding to the homogeneous anatomical regions, is then alleviated by the guided filter. Thereby, the algorithm reduces speckle noise while enhancing edge sharpness regardless of the size of the edges. In the synthetic images, the proposed method showed statistically significant improvements in peak signal-to-noise ratio(PSNR), structural similarity(SSIM), feature similarity index(FSIM) index and Mean Squared Error(MSE) compared with other speckle reduction methods, e.g., the squeeze boxes (SBF) filter, optimal Bayesian NLM (OBNLM) filter, speckle reducing anisotropic diffusion filter (SRAD), nonlocal low-rank framework (NLLRF) and multi-scale attention-guided neural network (MSANN). Similarly, our method outperformed the other methods in terms of mainly metrics. All the clinical images that were denoised using the six speckle reduction methods were reviewed by four radiologists for evaluation based on each radiologist's diagnostic preferences. All the radiologists showed a significant preference for the liver images and arotid artery images obtained using our methods in terms of effectively suppresses speckle noise while preserving the structural details. For the kidney and thyroid images, our method showed similar improvement over other methods. The experimental results show that this method has better performance than other state-of-the-art medical ultrasonic image speckle removal methods. [ABSTRACT FROM AUTHOR]

Most women, in general, have an ovarian cyst, which causes a variety of disorders. Cervical cysts occur when multiple cysts appear in or on top of the uterus. This is especially true for women who have a good reason for having a baby. Related to menstrual problems and cyst problems in women during pregnancy. Ultrasound imaging techniques are used to detect ovarian cysts. Doctors have many difficulties identifying these types of tumors that are not clearly visible from ultrasound images and what type of ovarian cyst they are To make these problems more useful to the doctors, the system of automatic detection of various cyst type has been implemented. The cyst detection and classification method are implemented using the features extracted from the ultrasound image. Automated detection methods and various ovarian cyst classification are implemented using a 2D Convolutional Neural network, and the proposed prediction model has yielded 99.37% accurate results. [ABSTRACT FROM AUTHOR]

Ultrasound imaging is widely used in clinical diagnoses because of its nonionizing radiation, low cost, and noninvasive operation. However, making a diagnosis based on ultrasound images is a labor-intensive process. An accurate lesion classification system can thus be used to assist doctors in making diagnoses. The performance of classification algorithms typically improves when they are trained on large, labeled datasets. However, collecting labeled data is an expensive and time-consuming task. Therefore, performing lesion classification via ultrasound images is still challenging due to the small number of available training samples. To address this issue, a data augmentation method for ultrasound images based on a conditional generative adversarial network was proposed in this study to perform lesion classification. A real image was input into the generative adversarial network to constrain the mapping between the images. Then, the data augmentation process based on the conditional generative adversarial network generated the corresponding segmentation masks by category. Considering that the data augmentation method based on affine transformation can generate only fake ultrasound images or segmentation masks separately, this study proposed to use image-to-image translation to generate fake ultrasound images from the corresponding segmentation masks. The ResNet-50 was used to classify benign and malignant lesions to validate the effectiveness of the proposed approach. The results showed that, by comparing to the traditional data augmentation method based on affine transformation in terms of four evaluation metrics, the average performances of the proposed method increased by approximately 13.05% and 12.85% for the classification of lesions in the segmented masks and ultrasound images of lymph nodes and breasts, respectively. The results suggested that the proposed method could realize the purpose of data augmentation and greatly improve the classification performance. [ABSTRACT FROM AUTHOR]

Non‐cancerous growths called uterine fibroids develop in the uterus. They can vary in size, location, and number, and can produce symptoms including excessive menstrual flow, pelvic discomfort, and reproductive problems. Early detection of uterine fibroids is important because it allows for timely intervention and appropriate management strategies. Extracting meaningful features from ultrasound (US) images requires robust and effective techniques. However, different feature extraction methods may yield varying results, and the choice of technique can influence the accuracy of fibroid detection. This paper presents an efficient approach for early detection of uterine fibroids in US images. Our proposed technique combines efficient feature extraction methods and a hybrid deep learning approach. Firstly, we employ the well‐known canny edge detection algorithm to accurately identify the edges of the fibroid region in the US images. This step helps in segmenting the target edge portion for further analysis. Additionally, we introduce an improved bird swarm optimization (IBSO) algorithm to extract a comprehensive set of features, utilizing both known features and newly obtained features. This approach enhances the accuracy of fibroid detection in uterine images. The proposed method uses a convolutional recurrent neural network (CRNN) to accurately detect and treat uterine fibroids, ensuring timely diagnosis and treatment. Our proposed IBSO‐CRNN model produces accuracy rates of 99.897% and 97.568% on the augmented and original datasets, respectively. [ABSTRACT FROM AUTHOR]

Medical ultrasonic imaging is affected by an inherent phenomenon called speckle noise, which prevents the identification of details in images. While several state-of-the-art methods have been already proposed for speckle noise reduction, they often suffer from blurring, artifacts, and losing the useful details and features of image which limits the accuracy of medical diagnosis. To address such challenges, in this paper, taking the advantage of convolutional neural network (CNN), a hybrid algorithm composed of anisotropic spatial filter and denoising CNN (DnCNN) is proposed for speckle noise reduction. To further eliminate the blurring effect and increase the contrast of image edges, we incorporate Wiener filter and fast local Laplacian filter as post-processing. The experimental results on medical images show that the proposed method, in addition to an effective noise suppression, can preserve the edges and structural details of the image. The proposed algorithm outperforms state-of-the-art noise removal filters, including Frost, Lee, Median, the speckle reducing anisotropic diffusion (SRAD) filter, Wiener filter, DnCNN, and fusion filters including SRAD + DnCNN, and SRAD + DnCNN + Wiener, in terms of PSNR, SSI, and SSIM metrics. [ABSTRACT FROM AUTHOR]

With the increasing dominance of artificial intelligence (AI) techniques, the important prospects for their application have extended to various medical fields, including domains such as in vitro diagnosis, intelligent rehabilitation, medical imaging, and prognosis. Breast cancer is a common malignancy that critically affects women's physical and mental health. Early breast cancer screening—through mammography, ultrasound, or magnetic resonance imaging (MRI)—can substantially improve the prognosis for breast cancer patients. AI applications have shown excellent performance in various image recognition tasks, and their use in breast cancer screening has been explored in numerous studies. This paper introduces relevant AI techniques and their applications in the field of medical imaging of the breast (mammography and ultrasound), specifically in terms of identifying, segmenting, and classifying lesions; assessing breast cancer risk; and improving image quality. Focusing on medical imaging for breast cancer, this paper also reviews related challenges and prospects for AI. [ABSTRACT FROM AUTHOR]

Thyroid nodule (TYN) is a life-threatening disease that is commonly observed among adults globally. The applications of deep learning in computer-aided diagnosis systems (CADs) for diagnosing thyroid nodules have attracted attention among clinical professionals due to their significantly potential role in reducing the occurrence of missed diagnoses. However, most techniques for segmenting thyroid nodules rely on U-Net structures or deep convolutional neural networks, which have limitations in obtaining different context information due to the diversities in the shapes and sizes, ambiguous boundaries, and heterostructure of thyroid nodules. To resolve these challenges, we present an encoder-decoder-based architecture (referred to as CIL-Net) for boosting TYN segmentation. There are three contributions in the proposed CIL-Net. First, the encoder is established using dense connectivity for efficient feature extraction and the triplet attention block (TAB) for highlighting essential feature maps. Second, we design a feature improvement block (FIB) using dilated convolutions and attention mechanisms to capture the global context information and also build up robust feature maps between the encoder-decoder branches. Third, we introduce the residual context block (RCB), which leverages residual units (ResUnits) to accumulate the context information from the multiple blocks of decoders in the decoder branch. We assess the segmentation quality of our proposed method using six different evaluation metrics on two standard datasets (DDTI and TN3K) of TYN and demonstrate competitive performance against advanced state-of-the-art methods. We consider that the proposed method advances the performance of TYN region localization and segmentation, which heavily rely on an accurate assessment of different context information. This advancement is primarily attributed to the comprehensive incorporation of dense connectivity, TAB, FIB, and RCB, which effectively capture both extensive and intricate contextual details. We anticipate that this approach reliability and visual explainability make it a valuable tool that holds the potential to significantly enhance clinical practices by offering reliable predictions to facilitate cognitive and healthcare decision-making. [ABSTRACT FROM AUTHOR]

Assessing fetal growth and development requires accurate identification of the fetal area contour and measurement of the Crown-Rump Length (CRL). In this paper, we presented a unique method for autonomously segmenting the fetal region in ultrasound images and calculating the CRL based on the U-Net architecture. Because of its capacity to capture both global and local information, the U-Net model is a popular choice for image segmentation tasks. Our method employs the U-Net model to extract the fetal region contour and measure the CRL, resulting in a dependable and efficient prenatal evaluation solution. [ABSTRACT FROM AUTHOR]

Simple Summary: In this study, we focus on improving radiation therapy (RT), particularly advances in managing the intra-refractive movement of pancreatic tumors and surrounding tissues during treatment. It is hard to track internal anatomy changes, such as those induced by respiration, in conventional RT, which can lead to the inadequate treatment of pancreatic cancer as well as cause potential harm to surrounding normal tissues. To address this issue, we focused on the use of ultrasound imaging, specifically the use of novelty flexible array transducers, for the real-time monitoring of these movements. However, challenges arise due to the nature of flexible array transducers that change with the shape of the body. Our study developed a new method, the Aligned Peak Response (APR) method, and combined it with an auxiliary structure with embedded markers. The method aims to improve the accuracy of the beamforming and motion tracking of ultrasound images during RT. We tested the effectiveness of the method using simulation and in vitro data aimed at improving RT accuracy and reducing patient risk. To develop ultrasound-guided radiotherapy, we proposed an assistant structure with embedded markers along with a novel alternative method, the Aligned Peak Response (APR) method, to alter the conventional delay-and-sum (DAS) beamformer for reconstructing ultrasound images obtained from a flexible array. We simulated imaging targets in Field-II using point target phantoms with point targets at different locations. In the experimental phantom ultrasound images, image RF data were acquired with a flexible transducer with in-house assistant structures embedded with needle targets for testing the accuracy of the APR method. The lateral full width at half maximum (FWHM) values of the objective point target (OPT) in ground truth ultrasound images, APR-delayed ultrasound images with a flat shape, and images acquired with curved transducer radii of 500 mm and 700 mm were 3.96 mm, 4.95 mm, 4.96 mm, and 4.95 mm. The corresponding axial FWHM values were 1.52 mm, 4.08 mm, 5.84 mm, and 5.92 mm, respectively. These results demonstrate that the proposed assistant structure and the APR method have the potential to construct accurate delay curves without external shape sensing, thereby enabling a flexible ultrasound array for tracking pancreatic tumor targets in real time for radiotherapy. [ABSTRACT FROM AUTHOR]

The facet joint injection is the most common procedure used to release lower back pain. In this paper, we proposed a deep learning method for detecting and segmenting facet joints in ultrasound images based on convolutional neural networks (CNNs) and enhanced data annotation. In the enhanced data annotation, a facet joint was considered as the first target and the ventral complex as the second target to improve the capability of CNNs in recognizing the facet joint. A total of 300 cases of patients undergoing pain treatment were included. The ultrasound images were captured and labeled by two professional anesthesiologists, and then augmented to train a deep learning model based on the Mask Region-based CNN (Mask R-CNN). The performance of the deep learning model was evaluated using the average precision (AP) on the testing sets. The data augmentation and data annotation methods were found to improve the AP. The AP50 for facet joint detection and segmentation was 90.4% and 85.0%, respectively, demonstrating the satisfying performance of the deep learning model. We presented a deep learning method for facet joint detection and segmentation in ultrasound images based on enhanced data annotation and the Mask R-CNN. The feasibility and potential of deep learning techniques in facet joint ultrasound image analysis have been demonstrated. [ABSTRACT FROM AUTHOR]

The level‐set based active contours have been found popular for medical image segmentation tasks, because of their inherent support for the topological changes—splitting and merging. Meanwhile, contour's leakage through the weak edges and premature convergence due to intensity inhomogeneity diminish its accuracy. Adjusting energy weights according to image features, local to the contour can be helpful. However, weight adjusted as deterministic function of the features is not adequate, limiting the segmentation accuracy. To address the problem, a new relevant edge probability based adaptively weighted level‐set evolution (REP‐WLSE) method for medical image segmentation is investigated. The weights used in this proposal are adaptive to an image relative value, obtained statistically from the feature‐explorations during the contour's evolution. The value is basically an estimate of contour's probability of finding relevant boundary edges on the image plane. Spatial intensity‐range filtering provides the feature space. An adaptive time‐step management scheme is also implemented, which controls the speed variation of the contour evolution. Time‐step is adjusted as a function of contour's alignment with the edges. The merits of the suggested methodology are—(i) reduced leakage through the weak edges, (ii) ability to handle the inhomogeneity, and (iii) increased chances of convergence around the foreground/region of interest (ROI). Experimental results using brain MR, abdominal ultrasound, and breast ultrasound images are presented. State‐of‐the‐art methods are compared using different metrics. The suggested methodology achieved better results. [ABSTRACT FROM AUTHOR]

Background: The quality of segmentation of thyroid nodules in ultrasound images is a crucial factor in preventing the cancerization of thyroid nodules. However, the existing standards for the ultrasound imaging of cancerous nodules have limitations, and changes of the echo pattern of thyroid nodules pose challenges in accurately segmenting nodules, which can affect the diagnostic results of medical professionals. Purpose: The aim of this study is to address the challenges related to segmentation accuracy due to noise, low contrast, morphological scale variations, and blurred edges of thyroid nodules in ultrasound images and improve the accuracy of ultrasound‐based thyroid nodule segmentation, thereby aiding the clinical diagnosis of thyroid nodules. Method: In this study, the dataset of thyroid ultrasound images was obtained from Hunan Provincial People's Hospital, consisting of a total of 3572 samples used for the training, validation, and testing of this model at a ratio of 8:1:1. A novel SK‐Unet++ network was used to enhance the segmentation accuracy of thyroid nodules. SK‐Unet++ is a novel deep learning architecture that adds the adaptive receptive fields based on the selective kernel (SK) attention mechanisms into the Unet++ network. The convolution blocks of the original UNet++ encoder part were replaced with finer SK convolution blocks in SK‐Unet++. First, multiple skip connections were incorporated so that SK‐Unet++ can make information from previous layers of the neural network to bypass certain layers and directly propagate to subsequent layers. The feature maps of the corresponding locations were fused on the channel, resulting in enhanced segmentation accuracy. Second, we added the adaptive receptive fields. The adaptive receptive fields were used to capture multiscale spatial features better by dynamically adjusting its receptive field. The assessment metrics contained dice similarity coefficient (Dsc), accuracy (Acc), precision (Pre), recall (Re), and Hausdorff distance, and all comparison experiments used the paired t‐tests to assess whether statistically significant performance differences existed (p < 0.05). And to address the multi‐comparison problem, we performed the false discovery rate (FDR) correction after the test. Results: The segmentation model had an Acc of 80.6%, Dsc of 84.7%, Pre of 77.5%, Re of 71.7%, and an average Hausdorff distance of 15.80 mm. Ablation experimental results demonstrated that each module in the network could contribute to the improved performance (p < 0.05) and determined the best combination of parameters. A comparison with other state‐of‐the‐art methods showed that SK‐Unet++ significantly outperformed them in terms of segmentation performance (p < 0.05), with a more accurate segmentation contour. Additionally, the adaptive weight changes of the SK module were monitored during the training process, and the resulting change curves demonstrated their convergence. Conclusion: Our proposed method demonstrates favorable performance in the segmentation of ultrasound images of thyroid nodules. Results confirmed that SK‐Unet++ is a feasible and effective method for the automatic segmentation of thyroid nodules in ultrasound images. The high accuracy achieved by our method can facilitate efficient screening of patients with thyroid nodules, ultimately reducing the workload of clinicians and radiologists. [ABSTRACT FROM AUTHOR]

Accurately segmenting thyroid nodules in ultrasound images is crucial for computer-aided diagnosis. Despite the success of Convolutional Neural Networks (CNNs) and Transformers in natural images processing, they struggle with precise boundaries and small-object segmentation in ultrasound images. To address this, a novel BFG&MSF-Net model is proposed in this paper, utilizing four newly designed modules: (1) a Boundary Feature Guidance Module (BFGM) for improving the edge details capturing; (2) a Multi-Scale Perception Fusion Module (MSPFM) for enhancing the information capture by combining a novel Positional Blended Attention (PBA) with the Pyramid Squeeze Attention (PSA); (3) a Depthwise Separable Atrous Spatial Pyramid Pooling Module (DSASPPM), used in the bottleneck to improve the contextual information capturing; and (4) a Refinement Module (RM) optimizing the low-level features for better organ and boundary identification. Evaluated on the TN3K and DDTI open-access datasets, BFG&MSF-Net demonstrates effective reduction of boundary segmentation errors and superior segmentation performance compared to commonly used segmentation models and state-of-the-art models, which makes it a promising solution for accurate thyroid nodule segmentation in ultrasound images.

Ultrasound images acquired through various measuring devices may have different styles, and each style may be specialized for diagnosing specific diseases. Accordingly, ultrasound image-to-image translation (US I2I) has become an essential research field. However, direct application of conventional I2I techniques to US I2I is difficult because it causes content deformation and has the problem of not being able to accurately translate fine textures. To solve the aforementioned problems, this paper proposes a novel feature decomposition scheme specialized for US I2I. The proposed feature decomposition explicitly separates texture and content information in latent space. Then, fine textures of the US image are effectively translated through translation of only the texture features. Moreover, I2I is carried out in a way that minimizes changes to the original content through reuse of content features. In addition to the feature decomposition scheme, we present a contrastive loss designed for content preservation. Specifically, the contrastive loss can maximize the content preservation effect because it preferentially performs query selection, which allows regions containing organ structures to be selected as queries (i.e., anchors). The proposed US image-specific learning scheme leads to qualitatively superior results, and the excellence of each method has been experimentally verified through various quantitative metrics.

Recently, deep learning-based methods have emerged as the preferred approach for ultrasound data analysis. However, these methods often require large-scale annotated datasets for training deep models, which are not readily available in practical scenarios. Additionally, the presence of speckle noise and other imaging artifacts can introduce numerous hard examples for ultrasound data classification. In this paper, drawing inspiration from self-supervised learning techniques, we present a pre-training method based on mask modeling specifically designed for ultrasound data. Our study investigates three different mask modeling strategies: random masking, vertical masking, and horizontal masking. By employing these strategies, our pre-training approach aims to predict the masked portion of the ultrasound images. Notably, our method does not rely on externally labeled data, allowing us to extract representative features without the need for human annotation. Consequently, we can leverage unlabeled datasets for pre-training. Furthermore, to address the challenges posed by hard samples in ultrasound data, we propose a novel hard sample mining strategy. To evaluate the effectiveness of our proposed method, we conduct experiments on two datasets. The experimental results demonstrate that our approach outperforms other state-of-the-art methods in ultrasound image classification. This indicates the superiority of our pre-training method and its ability to extract discriminative features from ultrasound data, even in the presence of hard examples.

Carotid plaque classification from ultrasound images is crucial for predicting ischemic stroke risk. While deep learning has shown effectiveness, it heavily relies on substantial labeled datasets. Achieving high performance with limited labeled images is essential for clinical use. Self-supervised learning (SSL) offers a potential solution; however, the existing works mainly focus on constructing the SSL tasks, neglecting the use of multiple tasks for pretraining. To overcome these limitations, this study proposed a self-supervised fusion network (Fusion-SSL) for carotid plaque ultrasound image classification with limited labeled data. Fusion-SSL consists of two SSL tasks: classifying image block order (Ordering) and predicting image rotation angle (Rotating). A dual-branch residual neural network was developed to fuse feature presentations learned by the two tasks, which can extract richer visual boundary shape and contour information than a single task. In this experiment, 1270 carotid plaque ultrasound images were collected from 844 patients at Zhongnan Hospital (Wuhan, China). The results showed that Fusion-SSL outperforms single SSL methods across different percentages of labeled training data, ranging from 10 to 100%. Moreover, with only 40% labeled training data, Fusion-SSL achieved comparable results to a single SSL method (predicting image rotation angle) with 100% labeled data. These results indicate that Fusion-SSL could be beneficial for the classification of carotid plaques and the early warning of a stroke in clinical practice.

Retinal tears (RTs) are usually detected by B-scan ultrasound images, particularly for individuals with complex eye conditions. However, traditional manual techniques for reading ultrasound images have the potential to overlook or inaccurately diagnose conditions. Thus, the development of rapid and accurate approaches for the diagnosis of an RT is highly important and urgent. The present study introduces a novel hybrid deep-learning model called DCT-Net to enable the automatic and precise diagnosis of RTs. The implemented model utilizes a vision transformer as the backbone and feature extractor. Additionally, in order to accommodate the edge characteristics of the lesion areas, a novel module called the residual deformable convolution has been incorporated. Furthermore, normalization is employed to mitigate the issue of overfitting and, a Softmax layer has been included to achieve the final classification following the acquisition of the global and local representations. The study was conducted by using both our proprietary dataset and a publicly available dataset. In addition, interpretability of the trained model was assessed by generating attention maps using the attention rollout approach. On the private dataset, the model demonstrated a high level of performance, with an accuracy of 97.78%, precision of 97.34%, recall rate of 97.13%, and an F1 score of 0.9682. On the other hand, the model developed by using the public funds image dataset demonstrated an accuracy of 83.82%, a sensitivity of 82.69% and a specificity of 82.40%. The findings, therefore present a novel framework for the diagnosis of RTs that is characterized by a high degree of efficiency, accuracy and interpretability. Accordingly, the technology exhibits considerable promise and has the potential to serve as a reliable tool for ophthalmologists.

Objective A non-invasive assessment of the risk of benign and malignant follicular thyroid cancer is invaluable in the choice of treatment options. The extraction and fusion of multidimensional features from ultrasound images of follicular thyroid cancer is decisive in improving the accuracy of identifying benign and malignant thyroid cancer. This paper presents a non-invasive preoperative benign and malignant risk assessment system for follicular thyroid cancer, based on the proposed deep feature extraction and fusion of ultrasound images of follicular thyroid cancer. Methods First, this study uses a convolution neural network (CNN) to obtain a global feature map of the image, and the fusion of global features cropped to local features to identify tumor images. Secondly, this tumour image is also extracted by googleNet and ResNet respectively to extract features and recognize the image. Finally, we employ an averaging algorithm to obtain the final recognition results.Results The experimental results show that the method proposed in this study achieved 89.95% accuracy, 88.46% sensitivity, 91.30% specificity and an AUC value of 96.69% in the local dataset obtained from Peking University Shenzhen Hospital, all of which are far superior to other models.Conclusion In this study, a non-invasive risk prediction system is proposed for ultrasound images of thyroid follicular tumours. We solve the problem of unbalanced sample distribution by means of an image enhancement algorithm. In order to obtain enough features to differentiate ultrasound images, a three-branched feature extraction network was designed in this study, and a balance of sensitivity and specificity is ensured by an averaging algorithm.

Ultrasound imaging is widely used in medical diagnostics due to its non-invasive and real-time capabilities. However, existing methods often overlook the benefits of fractional-order filters for denoising and dehazing. Thus, this work introduces an efficient multi-scale wavelet method for dehazing and denoising ultrasound images using a fractional-order filter, which integrates a guided filter, directional filter, fractional-order filter, and haze removal to the different resolution images generated by a multi-scale wavelet. In the directional filter stage, an eigen-analysis of each pixel is conducted to extract structural features, which are then classified into edges for targeted filtering. The guided filter subsequently reduces speckle noise in homogeneous anatomical regions. The fractional-order filter allows the algorithm to effectively denoise while improving edge definition, irrespective of the edge size. Haze removal can effectively eliminate the haze caused by attenuation. Our method achieved significant improvements, with PSNR reaching 31.25 and SSIM 0.905 on our ultrasound dataset, outperforming other methods. Additionally, on external datasets like McMaster and Kodak24, it achieved the highest PSNR (29.68, 28.62) and SSIM (0.858, 0.803). Clinical evaluations by four radiologists confirmed its superiority in liver and carotid artery images. Overall, our approach outperforms existing speckle reduction and structural preservation techniques, making it highly suitable for clinical ultrasound imaging.

Purpose: The aim of this study was to investigate the value of a deep learning model (DLM) based on breast tumor ultrasound image segmentation in predicting pathological response to neoadjuvant chemotherapy (NAC) in breast cancer. Methods: The dataset contains a total of 1393 ultrasound images of 913 patients from Renmin Hospital of Wuhan University, of which 956 ultrasound images of 856 patients were used as the training set, and 437 ultrasound images of 57 patients underwent NAC were used as the test set. A U-Net-based end-to-end DLM was developed for automatically tumor segmentation and area calculation. The predictive abilities of the DLM, manual segmentation model (MSM), and two traditional ultrasound measurement methods (longest axis model [LAM] and dual-axis model [DAM]) for pathological complete response (pCR) were compared using changes in tumor size ratios to develop receiver operating characteristic curves. Results: The average intersection over union value of the DLM was 0.856. The early-stage ultrasound-predicted area under curve (AUC) values of pCR were not significantly different from those of the intermediate and late stages (p< 0.05). The AUCs for MSM, DLM, LAM and DAM were 0.840, 0.756, 0.778 and 0.796, respectively. There was no significant difference in AUC values of the predictive ability of the four models. Conclusion: Ultrasonography was predictive of pCR in the early stages of NAC. DLM have a similar predictive value to conventional ultrasound for pCR, with an add benefit in effectively improving workflow. [ABSTRACT FROM AUTHOR]

Ultrasound (US) examination has unique advantages in diagnosing carpal tunnel syndrome (CTS), although identification of the median nerve (MN) and diagnosis of CTS depend heavily on the expertise of examiners. In the aim of alleviating this problem, we developed a one-stop automated CTS diagnosis system (OSA-CTSD) and evaluated its effectiveness as a computer-aided diagnostic tool. We combined real-time MN delineation, accurate biometric measurements and explainable CTS diagnosis into a unified framework, called OSA-CTSD. We then collected a total of 32,301 static images from US videos of 90 normal wrists and 40 CTS wrists for evaluation using a simplified scanning protocol. The proposed model exhibited better segmentation and measurement performance than competing methods, with a Hausdorff distance (95th percentile) score of 7.21 px, average symmetric surface distance score of 2.64 px, Dice score of 85.78% and intersection over union score of 76.00%. In the reader study, it exhibited performance comparable to the average performance of experienced radiologists in classifying CTS and outperformed inexperienced radiologists in terms of classification metrics (e.g., accuracy score 3.59% higher and F1 score 5.85% higher). Diagnostic performance of the OSA-CTSD was promising, with the advantages of real-time delineation, automation and clinical interpretability. The application of such a tool not only reduces reliance on the expertise of examiners but also can help to promote future standardization of the CTS diagnostic process, benefiting both patients and radiologists. [ABSTRACT FROM AUTHOR]

Ultrasound speckle is an interference pattern that reduces the quality of medical ultrasound images and decreases the capability of interpreting the medical information included in these images. Therefore, speckle suppression is vital whenever ultrasound imaging is used. In this research, an efficient adaptive ultrasound image filter is proposed to reduce ultrasound speckle while maintaining the edge cues in the image. In particular, the pixels that correspond to tissue boundaries are identified by computing the edge map of ultrasound image, and the noisy pixels are determined by calculating the image radiation. The proposed filtering method is designed to adaptively apply different window sizes and bandwidths based on the edge map and image radiation. The proposed method is employed to process a set of ultrasound images and the filtering performance is assessed using a set of broadly accepted filtering metrics. These metrics aim to evaluate the filtering performance, including the capability of the filter to preserve the edges and maintain the image quality. The performance of the proposed filtering method is compared with eight existing image filtering methods. The experimental results indicated that the proposed filtering method enables effective edge preservation and retains image quality. These results suggest the potential of the proposed filtering method to achieve effective speckle suppression in ultrasound images. [ABSTRACT FROM AUTHOR]

A clinical ultrasound imaging plays a significant role in the proper diagnosis of patients because, it is a cost-effective and non-invasive technique in comparison with other methods. The speckle noise contamination caused by ultrasound images during the acquisition process degrades its visual quality, which makes the diagnosis task difficult for physicians. Hence, to improve their visual quality, despeckling filters are commonly used for processing of such images. However, several disadvantages of existing despeckling filters discourage the use of existing despeckling filters to reduce the effect of speckle noise. In this paper, two dimensional cuckoo search optimization algorithm based despeckling filter is proposed for avoiding limitations of various existing despeckling filters. Proposed despeckling filter is developed by combining fast non-local means filter and 2D finite impulse response (FIR) filter with cuckoo search optimization algorithm. In the proposed despeckling filter, the coefficients of 2D FIR filter are optimized by using the cuckoo search optimization algorithm. The quantitative results comparison between the proposed despeckling filter and other existing despeckling filters are analyzed by evaluating PSNR, MSE, MAE, and SSIM values for different real ultrasound images. Results reveal that the visual quality obtained by the proposed despeckling filter is better than other existing despeckling filters. The numerical results also reveal that the proposed despeckling filter is highly effective for despeckling the clinical ultrasound images. [ABSTRACT FROM AUTHOR]

Objective: A non-invasive assessment of the risk of benign and malignant follicular thyroid cancer is invaluable in the choice of treatment options. The extraction and fusion of multidimensional features from ultrasound images of follicular thyroid cancer is decisive in improving the accuracy of identifying benign and malignant thyroid cancer. This paper presents a non-invasive preoperative benign and malignant risk assessment system for follicular thyroid cancer, based on the proposed deep feature extraction and fusion of ultrasound images of follicular thyroid cancer. Methods: First, this study uses a convolution neural network (CNN) to obtain a global feature map of the image, and the fusion of global features cropped to local features to identify tumor images. Secondly, this tumour image is also extracted by googleNet and ResNet respectively to extract features and recognize the image. Finally, we employ an averaging algorithm to obtain the final recognition results. Results: The experimental results show that the method proposed in this study achieved 89.95% accuracy, 88.46% sensitivity, 91.30% specificity and an AUC value of 96.69% in the local dataset obtained from Peking University Shenzhen Hospital, all of which are far superior to other models. Conclusion: In this study, a non-invasive risk prediction system is proposed for ultrasound images of thyroid follicular tumours. We solve the problem of unbalanced sample distribution by means of an image enhancement algorithm. In order to obtain enough features to differentiate ultrasound images, a three-branched feature extraction network was designed in this study, and a balance of sensitivity and specificity is ensured by an averaging algorithm. [ABSTRACT FROM AUTHOR]

Background The aim of this study is to design a deep learning (DL) model to preoperatively predict the occurrence of central lymph node metastasis (CLNM) in patients with papillary thyroid microcarcinoma (PTMC). Methods This research collected preoperative ultrasound (US) images and clinical factors of 611 PTMC patients. The clinical factors were analyzed using multivariate regression. Then, a DL model based on US images and clinical factors was developed to preoperatively predict CLNM. The model’s efficacy was evaluated using the receiver operating characteristic (ROC) curve, along with accuracy, sensitivity, specificity, and the F1 score. Results The multivariate analysis indicated an independent correlation factors including age ≥55 (OR = 0.309, p < 0.001), tumor diameter (OR = 2.551, p = 0.010), macrocalcifications (OR = 1.832, p = 0.002), and capsular invasion (OR = 1.977, p = 0.005). The suggested DL model utilized US images achieved an average area under the curve (AUC) of 0.65, slightly outperforming the model that employed traditional clinical factors (AUC = 0.64). Nevertheless, the model that incorporated both of them did not enhance prediction accuracy (AUC = 0.63). Conclusions The suggested approach offers a reference for the treatment and supervision of PTMC. Among three models used in this study, the deep model relied generally more on image modalities than the data modality of clinic records when making the predictions.

Ultrasound imaging is vital for diagnosing carotid artery vascular lesions, highlighting the importance of accurately segmenting lumens in ultrasound images to prevent, diagnose and treat vascular diseases. However, noise artifacts, blood residue and discontinuous lumens significantly affect segmentation accuracy. To achieve accurate lumen segmentation in low-quality images, we propose a novel segmentation algorithm which is guided by an adaptively generated shape prior. To tackle the above challenges, we introduce a shape-prior-based segmentation method for carotid artery lumen walls. The shape prior in this study is adaptively generated based on the evolutionary trend of vessel growth. Shape priors guide and constrain the active contour, resulting in precise segmentation. The efficacy of the proposed model was confirmed using 247 carotid artery ultrasound images, with experimental results showing an average Dice coefficient of 92.38%, demonstrating superior segmentation performance compared to existing mathematical models. Our method can quickly and effectively perform accurate lumen segmentation on low-quality carotid artery ultrasound images, which is of great significance for the diagnosis of cardiovascular and cerebrovascular diseases.

Speckle noise in ultrasound images (UIs) significantly reduces the accuracy of disease diagnosis. The aim of this study was to quantitatively evaluate its feasibility in salivary gland ultrasound imaging by modeling the adaptive non-local means (NLM) algorithm. UIs were obtained using an open-source device provided by SonoSkills and FUJIFILM Healthcare Europe. The adaptive NLM algorithm automates optimization by modeling the isotropic search window, eliminating the need for manual configuration in conventional NLM methods. The coefficient of variation (COV), contrast-to-noise ratio (CNR), and edge rise distance (ERD) were used as quantitative evaluation parameters. UIs of the salivary glands revealed evident visualization of the internal echo shape of the malignant tumor and calcification line using the adaptive NLM algorithm. Improved COV and CNR results (approximately 4.62 and 2.15 times, respectively) compared with noisy images were achieved. Additionally, when the adaptive NLM algorithm was applied to the UIs of patients with salivary gland sialolithiasis, the noisy images and ERD values were calculated almost similarly. In conclusion, this study demonstrated the applicability of the adaptive NLM algorithm in optimizing search window parameters for salivary gland UIs.