Ultrasonic Imaging最新文献

{"title":"Calcification Detection in Intravascular Ultrasound (IVUS) Images Using Transfer Learning Based MultiSVM model.","authors":"Priyanka Arora,&nbsp;Parminder Singh,&nbsp;Akshay Girdhar,&nbsp;Rajesh Vijayvergiya","doi":"10.1177/01617346231164574","DOIUrl":"https://doi.org/10.1177/01617346231164574","url":null,"abstract":"<p><p>Cardiovascular disease serves as the leading cause of death worldwide. Calcification detection is considered an important factor in cardiovascular diseases. Currently, medical practitioners visually inspect the presence of calcification using intravascular ultrasound (IVUS) images. The study aims to detect the extent of calcification as belonging to class I, II as mild calcification, and class III, IV as dense calcification from IVUS images acquired at 40 MHz. To detect calcification, the features were extracted using improved AlexNet architecture and then were fed into machine learning classifiers. The experiments were carried out using 14 real IVUS pullbacks of 10 patients. Experimental results show that the combination of traditional machine learning with deep learning approaches significantly improves accuracy. The results show that support vector machines outperform all other classifiers. The proposed model is compared with two other pre-trained models GoogLeNet (98.8%), SqueezeNet (99.2%), and exhibits considerable improvement in classification accuracy (99.8%). In the future other models such as Vision Transformers could be explored with additional feature selection methods such as ReliefF, PSO, ACO, etc. to improve the overall accuracy of diagnosis.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 3","pages":"136-150"},"PeriodicalIF":2.3,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9684965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Infrared Thermography, Intratendon Vascular Resistance, and Echotexture in Athletes with Patellar Tendinopathy: A Cross-Sectional Study.","authors":"Francisco J Molina-Payá,&nbsp;José Ríos-Díaz,&nbsp;Francisco Carrasco-Martínez,&nbsp;Jacinto J Martínez-Payá","doi":"10.1177/01617346231153581","DOIUrl":"https://doi.org/10.1177/01617346231153581","url":null,"abstract":"<p><p>Ultrasonographic signs of tendinopathies are an increase in thickness, loss of alignment in collagen fibers and the presence of neovascularization. Nevertheless, analysis of intratendinous vascular resistance (IVR) can be more useful for understanding the physiological state of the tissue. To show thermal, echotextural, and Doppler signal differences in athletes with patellar tendinopathy and controls. Twenty-six athletes with patellar tendinopathy (PT) participants (30.1 years; <i>SD</i> = 9.0 years) and 27 asymptomatic athletes (23.3 years; <i>SD</i> = 5.38 years) were evaluated with thermographic and Doppler ultrasonography (DS). Area of Doppler signals (DS), echotextural parameters (echointensity and echovariation) and IVR were determined by image analysis. The statistical analysis was performed by Bayesian methods and the results were showed by Bayes Factor (BF10: probability of alternative hypothesis over null hypothesis), and Credibility intervals (CrI) of the effect. The absolute differences of temperature (TD) were clearly greater (BF10 = 19) in the tendinopathy group (patients) than in controls. Regarding temperature differences between the affected and healthy limb, strong evidence was found (BF₁₀ = 14) for a higher temperature (effect = 0.53°C; 95% CrI = 0.15°C-0.95°C) and very strong for reduced IVR compared (BF₁₀ = 71) (effect = -0.67; 95% CrI = -1.10 to 0.25). The differences in area of DS (BF₁₀ = 266) and EV (BF₁₀ = 266) were higher in tendinopathy group. TD showed a moderate positive correlation with VISA-P scores (tau-B = .29; 95% CrI = .04-.51) and strong correlation with IVR (<i>r</i> = -.553; 95%CrI = -.75 to .18). Athletes with patellar tendinopathy showed a more pronounced thermal difference, a larger area of Doppler signal, a lower IVR and a moderately higher echovariaton than controls. The correlation between temperature changes and IVR might be related with the coexistence of degenerative and inflammatory process in PT.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 2","pages":"47-61"},"PeriodicalIF":2.3,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9684441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Boundary-oriented Network for Automatic Breast Tumor Segmentation in Ultrasound Images.","authors":"Mengmeng Zhang,&nbsp;Aibin Huang,&nbsp;Debiao Yang,&nbsp;Rui Xu","doi":"10.1177/01617346231162925","DOIUrl":"https://doi.org/10.1177/01617346231162925","url":null,"abstract":"<p><p>Breast cancer is considered as the most prevalent cancer. Using ultrasound images is a momentous clinical diagnosis method to locate breast tumors. However, accurate segmentation of breast tumors remains an open problem due to ultrasound artifacts, low contrast, and complicated tumor shapes in ultrasound images. To address this issue, we proposed a boundary-oriented network (BO-Net) for boosting breast tumor segmentation in ultrasound images. The BO-Net boosts tumor segmentation performance from two perspectives. Firstly, a boundary-oriented module (BOM) was designed to capture the weak boundaries of breast tumors by learning additional breast tumor boundary maps. Second, we focus on enhanced feature extraction, which takes advantage of the Atrous Spatial Pyramid Pooling (ASPP) module and Squeeze-and-Excitation (SE) block to obtain multi-scale and efficient feature information. We evaluate our network on two public datasets: Dataset B and BUSI. For the Dataset B, our network achieves 0.8685 in Dice, 0.7846 in Jaccard, 0.8604 in Precision, 0.9078 in Recall, and 0.9928 in Specificity. For the BUSI dataset, our network achieves 0.7954 in Dice, 0.7033 in Jaccard, 0.8275 in Precision, 0.8251 in Recall, and 0.9814 in Specificity. Experimental results show that BO-Net outperforms the state-of-the-art segmentation methods for breast tumor segmentation in ultrasound images. It demonstrates that focusing on boundary and feature enhancement creates more efficient and robust breast tumor segmentation.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 2","pages":"62-73"},"PeriodicalIF":2.3,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9740593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prediction of Renal Function 1 Year After Transplantation Using Machine Learning Methods Based on Ultrasound Radiomics Combined With Clinical and Imaging Features.","authors":"Lili Zhu,&nbsp;Renjun Huang,&nbsp;Zhiyong Zhou,&nbsp;Qingmin Fan,&nbsp;Junchen Yan,&nbsp;Xiaojing Wan,&nbsp;Xiaojun Zhao,&nbsp;Yao He,&nbsp;Fenglin Dong","doi":"10.1177/01617346231162910","DOIUrl":"https://doi.org/10.1177/01617346231162910","url":null,"abstract":"<p><p>Kidney transplantation is the most effective treatment for advanced chronic kidney disease (CKD). If the prognosis of transplantation can be predicted early after transplantation, it might improve the long-term survival of patients with transplanted kidneys. Currently, studies on the assessment and prediction of renal function by radiomics are limited. Therefore, the present study aimed to explore the value of ultrasound (US)-based imaging and radiomics features, combined with clinical features to develop and validate the models for predicting transplanted kidney function after 1 year (TKF-1Y) using different machine learning algorithms. A total of 189 patients were included and classified into the abnormal TKF-1Y group, and the normal TKF-1Y group based on their estimated glomerular filtration rate (eGFR) levels 1 year after transplantation. The radiomics features were derived from the US images of each case. Three machine learning methods were employed to establish different models for predicting TKF-1Y using selected clinical and US imaging as well as radiomics features from the training set. Two US imaging, four clinical, and six radiomics features were selected. Then, the clinical (including clinical and US image features), radiomics, and combined models were developed. The area under the curves (AUCs) of the models was 0.62 to 0.82 within the test set. Combined models showed statistically higher AUCs than the radiomics models (all <i>p</i>-values <.05). The prediction performance of different models was not significantly affected by the different machine learning algorithms (all <i>p</i>-values >.05). In conclusion, US imaging features combined with clinical features could predict TKF-1Y and yield an incremental value over radiomics features. A model integrating all available features may further improve the predictive efficacy. Different machine learning algorithms may not have a significant impact on the predictive performance of the model.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 2","pages":"85-96"},"PeriodicalIF":2.3,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9740585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Breast Tumor Classification using Short-ResNet with Pixel-based Tumor Probability Map in Ultrasound Images.","authors":"You-Wei Wang,&nbsp;Tsung-Ter Kuo,&nbsp;Yi-Hong Chou,&nbsp;Yu Su,&nbsp;Shing-Hwa Huang,&nbsp;Chii-Jen Chen","doi":"10.1177/01617346231162906","DOIUrl":"https://doi.org/10.1177/01617346231162906","url":null,"abstract":"<p><p>Breast cancer is the most common form of cancer and is still the second leading cause of death for women in the world. Early detection and treatment of breast cancer can reduce mortality rates. Breast ultrasound is always used to detect and diagnose breast cancer. The accurate breast segmentation and diagnosis as benign or malignant is still a challenging task in the ultrasound image. In this paper, we proposed a classification model as short-ResNet with DC-UNet to solve the segmentation and diagnosis challenge to find the tumor and classify benign or malignant with breast ultrasonic images. The proposed model has a dice coefficient of 83% for segmentation and achieves an accuracy of 90% for classification with breast tumors. In the experiment, we have compared with segmentation task and classification result in different datasets to prove that the proposed model is more general and demonstrates better results. The deep learning model using short-ResNet to classify tumor whether benign or malignant, that combine DC-UNet of segmentation task to assist in improving the classification results.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 2","pages":"74-84"},"PeriodicalIF":2.3,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9740594","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Application of Contrast-Enhanced Ultrasound Galactography in Patients With Pathologic Nipple Discharge.","authors":"Yongmei Wang,&nbsp;Yongzhu Pu,&nbsp;Mei Yin,&nbsp;Yawen Wang,&nbsp;Song Zhao,&nbsp;Jianli Wang,&nbsp;Rong Ma","doi":"10.1177/01617346221141470","DOIUrl":"https://doi.org/10.1177/01617346221141470","url":null,"abstract":"<p><p>Twenty patients with pathologic nipple discharge underwent conventional galactography and contrast-enhanced ultrasound (CEUS) galactography. Images were reviewed for detection of suspicious lesions. Lesion localization information from CEUS galactography was recorded. We included 25 lesions from the 20 included patients. The pathological results revealed 13 intraductal papillomas. The detective rates of intraductal papilloma by conventional galactography and CEUS galactography were 92.31% and 100%, respectively. All the preoperative localizations of lesions from CEUS galactography were in accordance with the surgical detections. CEUS galactography is a highly effective tool for the detection of intraductal breast lesions, and it could provide accurate lesion localization information for an optimal surgical design.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 1","pages":"17-21"},"PeriodicalIF":2.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10634736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nonlinear Harmonic Distortion of Complementary Golay Codes.","authors":"Fraser Hamilton,&nbsp;Peter Hoskins,&nbsp;George Corner,&nbsp;Zhihong Huang","doi":"10.1177/01617346221147820","DOIUrl":"https://doi.org/10.1177/01617346221147820","url":null,"abstract":"<p><p>Recent advances in electronics miniaturization have led to the development of low-power, low-cost, point-of-care ultrasound scanners. Low-cost systems employing simple bi-level pulse generation devices need only utilize binary phase modulated coded excitations to significantly improve sensitivity; however the performance of complementary codes in the presence of nonlinear harmonic distortion has not been thoroughly investigated. Through simulation, it was found that nonlinear propagation media with little attenuative properties can significantly deteriorate the Peak Sidelobe Level (PSL) performance of complementary Golay coded pulse compression, resulting in PSL levels of -62 dB using nonlinear acoustics theory contrasted with -198 dB in the linear case. Simulations of 96 complementary pairs revealed that some pairs are more robust to sidelobe degradation from nonlinear harmonic distortion than others, up to a maximum PSL difference of 17 dB between the best and worst performing codes. It is recommended that users consider the effects of nonlinear harmonic distortion when implementing binary phase modulated complementary Golay coded excitations.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 1","pages":"22-29"},"PeriodicalIF":2.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9893299/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10644752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"B-line Elastography Measurement of Lung Parenchymal Elasticity.","authors":"Ren Koda,&nbsp;Hayato Taniguchi,&nbsp;Kei Konno,&nbsp;Yoshiki Yamakoshi","doi":"10.1177/01617346221149141","DOIUrl":"https://doi.org/10.1177/01617346221149141","url":null,"abstract":"<p><p>This paper proposes a method to determine the elasticity of the lung parenchyma from the B-line Doppler signal observed using continuous shear wave elastography, which uses a small vibrator placed on the tissue surface to propagate continuous shear waves with a vibration frequency of approximately 100 Hz. Since the B-line is generated by multiple reflections in fluid-storing alveoli near the lung surface, the ultrasonic multiple-reflection signal from the B-line is affected by the Doppler shift due to shear waves propagating in the lung parenchyma. When multiple B-lines are observed, the propagation velocity can be estimated by measuring the difference in propagation time between the B-lines. Therefore, continuous shear wave elastography can be used to determine the elasticity of the lung parenchyma by measuring the phase difference of shear wave between the B-lines. In this study, three elastic sponges (soft, medium, and hard) with embedded glass beads were used to simulate fluid-storing alveoli. Shear wave velocity measured using the proposed method was compared with that calculated using Young's modulus obtained from compression measurement. Using the proposed method, the measured shear wave velocities (mean ± S.D.) were 3.78 ± 0.23, 4.24 ± 0.12, and 5.06 ± 0.05 m/s for soft, medium, and hard sponges, respectively, which deviated by a maximum of 5.37% from the values calculated using the measured Young's moduli. The shear wave velocities of the sponge phantom were in a velocity range similar to the mean shear wave velocities of healthy and diseased lungs reported by magnetic resonance elastography (3.25 and 4.54 m/s, respectively). B-line elastography may enable emergency diagnoses of acute lung disease using portable ultrasonic echo devices.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 1","pages":"30-41"},"PeriodicalIF":2.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9188483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving Image Quality by Deconvolution Recovery Filter in Ultrasound Imaging.","authors":"Jingwen Pan,&nbsp;Hu Peng,&nbsp;Zhihui Han,&nbsp;Dan Hu,&nbsp;Yadan Wang,&nbsp;Yuanguo Wang","doi":"10.1177/01617346221141634","DOIUrl":"https://doi.org/10.1177/01617346221141634","url":null,"abstract":"<p><p>Due to the advantages of non-radiation and real-time performance, ultrasound imaging is essential in medical imaging. Image quality is affected by the performance of the transducer in an ultrasound imaging system. For example, the bandwidth controls the pulse length, resulting in different axial resolutions. Therefore, a transducer with a large bandwidth helps to improve imaging quality. However, large bandwidths lead to increased system cost and sometimes a loss of sensitivity and lateral resolution in attenuating media. In this paper, a deconvolution recovery method combined with a frequency-domain filtering technique (DRF) is proposed to improve the imaging quality, especially for the axial resolution. In this method, the received low-bandwidth echo signals are converted into high-bandwidth signals, which is similar to the echo signals produced by a high-bandwidth transducer, and the imaging quality is improved. Simulation and experiment results show that, compared with Delay-and-sum (DAS) method, the DRF method improved axial resolution from 0.60 to 0.41 mm in simulation and from 0.62 to 0.47 mm in the tissue-mimicking phantom experiment. The contrast ratio performance is improved to some extent compared with the DAS in experimental and in-vivo images. Besides, the proposed method has the potential to further improve image quality by combining it with adaptive weightings, such as the minimum variance method.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"45 1","pages":"3-16"},"PeriodicalIF":2.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9200705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Liver Fibrosis Assessment Using Radiomics of Ultrasound Homodyned-K imaging Based on the Artificial Neural Network Estimator.","authors":"Zhuhuang Zhou,&nbsp;Zijing Zhang,&nbsp;Anna Gao,&nbsp;Dar-In Tai,&nbsp;Shuicai Wu,&nbsp;Po-Hsiang Tsui","doi":"10.1177/01617346221120070","DOIUrl":"https://doi.org/10.1177/01617346221120070","url":null,"abstract":"<p><p>The homodyned-K distribution is an important ultrasound backscatter envelope statistics model of physical meaning, and the parametric imaging of the model parameters has been explored for quantitative ultrasound tissue characterization. In this paper, we proposed a new method for liver fibrosis characterization by using radiomics of ultrasound backscatter homodyned-K imaging based on an improved artificial neural network (iANN) estimator. The iANN estimator was used to estimate the ultrasound homodyned-K distribution parameters <i>k</i> and <i>α</i> from the backscattered radiofrequency (RF) signals of clinical liver fibrosis (<i>n</i> = 237), collected with a 3-MHz convex array transducer. The RF data were divided into two groups: Group I corresponded to liver fibrosis with no hepatic steatosis (<i>n</i> = 94), and Group II corresponded to liver fibrosis with mild to severe hepatic steatosis (<i>n</i> = 143). The estimated homodyned-K parameter values were then used to construct <i>k</i> and <i>α</i> parametric images using the sliding window technique. Radiomics features of <i>k</i> and <i>α</i> parametric images were extracted, and feature selection was conducted. Logistic regression classification models based on the selected radiomics features were built for staging liver fibrosis. Experimental results showed that the proposed method is overall superior to the radiomics method of uncompressed envelope images when assessing liver fibrosis. Regardless of hepatic steatosis, the proposed method achieved the best performance in staging liver fibrosis ≥<i>F1</i>, ≥<i>F4</i>, and the area under the receiver operating characteristic curve was 0.88, 0.85 (Group I), and 0.85, 0.86 (Group II), respectively. Radiomics has improved the ability of ultrasound backscatter statistical parametric imaging to assess liver fibrosis, and is expected to become a new quantitative ultrasound method for liver fibrosis characterization.</p>","PeriodicalId":49401,"journal":{"name":"Ultrasonic Imaging","volume":"44 5-6","pages":"229-241"},"PeriodicalIF":2.3,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"33438466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}