Sediment incipient velocity （SIV） is a vital parameter for sediment research and river dynamics. This paper describes a novel method of estimating SIV based on the known flow velocity in the movable-bed model experi...Sediment incipient velocity （SIV） is a vital parameter for sediment research and river dynamics. This paper describes a novel method of estimating SIV based on the known flow velocity in the movable-bed model experiment. In this method, we use B-mode ultrasound imaging technique to get video images of moving particles and topography under water. By statistical analysis of video images, the relationship between the average number of imaging particles and flow velocity is obtained. The relationship between the change rate of average number and flow velocity is analyzed in sediment incipient process. These relationships are used to estimate the SIV. Lastly, the changed topography verifies the estimated velocity. The results show there is a sudden change in these relationships which can be used to estimate the SIV with high resolution by using a B-mode ultrasound device. The estimated SIV of plastic sands （particle size is about 0.25 mm） is 3.64 cm · s^-1 and the estimated SIV of natural sands （particle size is about 0.25 mm） is 5.47 cm · s^-1in the same condition.展开更多
This paper deals with the temperature correlation of gray scale of B-mode ultrasound image from heated tissue. In this study, many in-vitro fresh pig livers are heated in a temperature range from 28 ℃ to 45℃, from w...This paper deals with the temperature correlation of gray scale of B-mode ultrasound image from heated tissue. In this study, many in-vitro fresh pig livers are heated in a temperature range from 28 ℃ to 45℃, from which a series of B-mode ultrasonic images of livers were obtained. The gray-value is evaluated from the ultrasound images respectively. A correlation of the mean gray value of the selected regions (12×12 pixels) in B-mode ultrasonic images of liver and its temperature was pointed out. And the experiment results agreed the evaluation well. And it is possible to monitor the tissue temperature changing in hyperthermia using this correlation.展开更多
Objective To study the role of bladder trabeculation found by B-mode ultrasound in evaluating the degree of bladder outlet obstruction ( BOO ) and the bladder function in benign prostatic hyperplasia ( BPH) patients. ...Objective To study the role of bladder trabeculation found by B-mode ultrasound in evaluating the degree of bladder outlet obstruction ( BOO ) and the bladder function in benign prostatic hyperplasia ( BPH) patients. Methods Conducted prospective research to determine differences in clinical data and urodynamic展开更多
The image features, such as enhancements, shadows and lateral shadows, posterior to a tumor in a B-Mode diagnostic ultrasonography, are studied. The tumor is simplified as a homogeneous cylinder in an inhomogeneous me...The image features, such as enhancements, shadows and lateral shadows, posterior to a tumor in a B-Mode diagnostic ultrasonography, are studied. The tumor is simplified as a homogeneous cylinder in an inhomogeneous medium. Using the ray theory we derive the intensity in the region behind the cylinder. Taking the influences of sound attenuation and the time gain compensation of the B-Mode displyer into acount , we simulate with a computer five images corresponding to our experiments conducted with a tissue equivalent phantom. The theoratical images seem to be in good qualitative agreement with the experimental results.展开更多
目的综合考虑B型超声(B-mode ultrasound,B-US)和对比增强超声(contrast-enhanced ultrasound,CEUS)双模态信息有助于提升乳腺肿瘤诊断的准确性,从而利于提高患者生存率。然而,目前大多数模型只关注B-US的特征提取,忽视了CEUS特征的学...目的综合考虑B型超声(B-mode ultrasound,B-US)和对比增强超声(contrast-enhanced ultrasound,CEUS)双模态信息有助于提升乳腺肿瘤诊断的准确性,从而利于提高患者生存率。然而,目前大多数模型只关注B-US的特征提取,忽视了CEUS特征的学习和双模态信息的融合处理。为解决上述问题,提出了一个融合时空特征与时间约束的双模态乳腺肿瘤诊断模型(spatio-temporal feature and temporal-constrained model,STFTCM)。方法首先,基于双模态信息的数据特点,采用异构双分支网络学习B-US和CEUS包含的时空特征。然后,设计时间注意力损失函数引导CEUS分支关注造影剂流入病灶区的时间窗口,从该窗口期内提取CEUS特征。最后,借助特征融合模块实现双分支网络之间的横向连接,通过将B-US特征作为CEUS分支补充信息的方式,完成双模态特征融合。结果在收集到的数据集上进行对比实验,STFTCM预测的正确率、敏感性、宏平均F1和AUC(area under the curve)指标均表现优秀,其中预测正确率达88.2%,领先于其他先进模型。消融实验中,时间注意力约束将模型预测正确率提升5.8%,特征融合使得模型诊断正确率相较于单分支模型至少提升2.9%。结论本文提出的STFTCM能有效地提取并融合处理B-US和CEUS双模态信息,给出准确的诊断结果。同时,时间注意力约束和特征融合模块可以显著地提升模型性能。展开更多
基金Supported by the Fundamental Research Funds for the Central Universities(2014212020205)
文摘Sediment incipient velocity （SIV） is a vital parameter for sediment research and river dynamics. This paper describes a novel method of estimating SIV based on the known flow velocity in the movable-bed model experiment. In this method, we use B-mode ultrasound imaging technique to get video images of moving particles and topography under water. By statistical analysis of video images, the relationship between the average number of imaging particles and flow velocity is obtained. The relationship between the change rate of average number and flow velocity is analyzed in sediment incipient process. These relationships are used to estimate the SIV. Lastly, the changed topography verifies the estimated velocity. The results show there is a sudden change in these relationships which can be used to estimate the SIV with high resolution by using a B-mode ultrasound device. The estimated SIV of plastic sands （particle size is about 0.25 mm） is 3.64 cm · s^-1 and the estimated SIV of natural sands （particle size is about 0.25 mm） is 5.47 cm · s^-1in the same condition.
基金The research was supported by National Nature Science Foundation (30470450) Education Committee Foundation( KP0608200201 ) Elitist Foundation( KW5800200351 ) from Beijing City,China.
文摘This paper deals with the temperature correlation of gray scale of B-mode ultrasound image from heated tissue. In this study, many in-vitro fresh pig livers are heated in a temperature range from 28 ℃ to 45℃, from which a series of B-mode ultrasonic images of livers were obtained. The gray-value is evaluated from the ultrasound images respectively. A correlation of the mean gray value of the selected regions (12×12 pixels) in B-mode ultrasonic images of liver and its temperature was pointed out. And the experiment results agreed the evaluation well. And it is possible to monitor the tissue temperature changing in hyperthermia using this correlation.
文摘Objective To study the role of bladder trabeculation found by B-mode ultrasound in evaluating the degree of bladder outlet obstruction ( BOO ) and the bladder function in benign prostatic hyperplasia ( BPH) patients. Methods Conducted prospective research to determine differences in clinical data and urodynamic
基金Supported by National Natural Science Fundation of China
文摘The image features, such as enhancements, shadows and lateral shadows, posterior to a tumor in a B-Mode diagnostic ultrasonography, are studied. The tumor is simplified as a homogeneous cylinder in an inhomogeneous medium. Using the ray theory we derive the intensity in the region behind the cylinder. Taking the influences of sound attenuation and the time gain compensation of the B-Mode displyer into acount , we simulate with a computer five images corresponding to our experiments conducted with a tissue equivalent phantom. The theoratical images seem to be in good qualitative agreement with the experimental results.
文摘目的综合考虑B型超声(B-mode ultrasound,B-US)和对比增强超声(contrast-enhanced ultrasound,CEUS)双模态信息有助于提升乳腺肿瘤诊断的准确性,从而利于提高患者生存率。然而,目前大多数模型只关注B-US的特征提取,忽视了CEUS特征的学习和双模态信息的融合处理。为解决上述问题,提出了一个融合时空特征与时间约束的双模态乳腺肿瘤诊断模型(spatio-temporal feature and temporal-constrained model,STFTCM)。方法首先,基于双模态信息的数据特点,采用异构双分支网络学习B-US和CEUS包含的时空特征。然后,设计时间注意力损失函数引导CEUS分支关注造影剂流入病灶区的时间窗口,从该窗口期内提取CEUS特征。最后,借助特征融合模块实现双分支网络之间的横向连接,通过将B-US特征作为CEUS分支补充信息的方式,完成双模态特征融合。结果在收集到的数据集上进行对比实验,STFTCM预测的正确率、敏感性、宏平均F1和AUC(area under the curve)指标均表现优秀,其中预测正确率达88.2%,领先于其他先进模型。消融实验中,时间注意力约束将模型预测正确率提升5.8%,特征融合使得模型诊断正确率相较于单分支模型至少提升2.9%。结论本文提出的STFTCM能有效地提取并融合处理B-US和CEUS双模态信息,给出准确的诊断结果。同时,时间注意力约束和特征融合模块可以显著地提升模型性能。
