Quantitative analysis of ultrasonic B-mode images

Various sources of variability, such as speckle noise, depth dependence and inhomogeneous intervening tissue, are involved in B-mode images, even when using the same ultrasonic equipment with fixed settings. The behavior of these sources of variability was investigated by texture analysis of images obtained from simulations and from a tissue-mimicking phantom, a normal adult liver and a pediatric renal (Wilms') tumor. First-order statistics (MEAN and SNR) and second-order statistics from the co-occurrence matrix (ENT and COR) were calculated. In a phantom, the SNR and ENT show a clear depth dependence. In biological tissue, the variability is mainly caused by the speckle noise and inhomogeneous intervening tissue. In addition, almost the entire range of the COR feature is present in images of liver and tumor. Furthermore, all the features calculated in windows of 1 cm2 exhibit an overlap among the different media. With the second-order features, it is possible to discriminate 85% reliable (average) between the normal, adult, liver and the pediatric renal tumor above a window size of 9 cm2. The SNR can not discriminate between these tissues. The maximum resolution of 9 cm2 reveals a serious limitation of parametric imaging. Finally, the features reproduce well in the case of follow-up of an abdominal tumor during chemotherapy.     

Minimum requirement of artificial noise level for using noise-assisted correlation algorithm to suppress artifacts in ultrasonic Nakagami images

The Nakagami image is a complementary imaging mode for pulse-echo ultrasound B-scan to characterize tissues. White noise in anechoic areas induces artifacts in the Nakagami image. Recently, we proposed a noise-assisted correlation algorithm (NCA) for suppressing the Nakagami artifact. In the NCA, artificial white noise is intentionally added twice to backscattered signals to produce two noisy data, which are used to establish a correlation profile for rejecting noise. This study explored the effects of artificial noise level on the NCA to suppress the artifact of the Nakagami image. Simulations were conducted to produce B-mode images of anechoic regions under signal-to-noise ratios (SNRs) of 20, 10 and 5 dB. Various artificial noise levels ranging from 0.1- to 1-fold of the intrinsic noise amplitude were used in the NCA for constructing the Nakagami images. Phantom experiments were conducted to validate the performance of using the optimal artificial noise level suggested by the simulation results to suppress the Nakagami artifacts by the NCA. The simulation results indicated that the artifacts of the Nakagami image in the anechoic regions can be gradually suppressed by increasing the artificial noise level used in the NCA to improve the image contrast-to-noise ratio (CNR). The CNR of the Nakagami image reached 20 dB when the artificial noise level was 0.7-fold of the intrinsic noise amplitude. This criterion was demonstrated by the phantom results to provide the NCA with an excellent ability to obtain artifact-free Nakagami images.     

Segmentation of the luminal border in intravascular ultrasound B-mode images using a probabilistic approach

Intravascular ultrasound (IVUS) is a catheter-based medical imaging technique that produces cross-sectional images of blood vessels and is particularly useful for studying atherosclerosis. In this paper, we present a computational method for the delineation of the luminal border in IVUS B-mode images. The method is based in the minimization of a probabilistic cost function (that deforms a parametric curve) which defines a probability field that is regularized with respect to the given likelihoods of the pixels belonging to blood and non-blood. These likelihoods are obtained by a Support Vector Machine classifier trained using samples of the lumen and non-lumen regions provided by the user in the first frame of the sequence to be segmented. In addition, an optimization strategy is introduced in which the direction of the steepest descent and Broyden–Fletcher–Goldfarb–Shanno optimization methods are linearly combined to improve convergence. Our proposed method (MRK) is capable of segmenting IVUS B-mode images from different systems and transducer frequencies without the need of any parameter tuning, and it is robust with respect to changes of the B-mode reconstruction parameters which are subjectively adjusted by the interventionist. We validated the proposed method on six 20 MHz and six 40 MHz IVUS stationary sequences corresponding to regions with different degrees of stenosis, and evaluated its performance by comparing the segmentation results with manual segmentation by two observers. Furthermore, we compared our method with the segmentation results on the same sequences as provided by the authors of three other segmentation methods available in the literature. The performance of all methods was quantified using Dice and Jaccard similarity indexes, Hausdorff distance, linear regression and Bland–Altman analysis. The results indicate the advantages of our method for the segmentation of the lumen contour.     

超声弹性成像对乳腺小实性病灶的鉴别诊断价值

目的 探讨超声弹性成像在乳腺小实性病灶(最大径≤10mm)鉴别诊断中的价值.方法 回顾性分析经手术病理证实的274个乳腺小实性病灶的常规超声及超声弹性成像图像.常规超声采用BI-RADS-US标准判断病灶的良恶性.超声弹性成像采用改良5分评分法判断病灶的良恶性.以手术病理结果为诊断金标准,构建受试者工作特征曲线,比较两种方法对乳腺肿物的诊断率.结果 超声弹性成像及常规超声BI-RADS分级诊断乳腺良恶性病灶的曲线下面积分别为0.891和0.742,差异有统计学意义(Z=4.053,P＜0.001).结论 超声弹性成像有助于乳腺小实性病灶的鉴别诊断.     

B-mode ultrasonic examination of the testes

This study was carried out to determine the potential value of B-mode ultrasonography in the identification and diagnosis of intratesticular disorders, information heretofore obtainable only through biopsy. Fifty-nine cases have been examined to date. Fifty-three were referred from an infertility clinic and include oligospermic and azoospermic subjects as well as individuals with normal semen analyses. There have been two hydroceles, two neoplasms, one abscess and one Klinefelter's syndrome. Our initial results indicate no difference between individuals with normal or subnormal semen quality with respect to the presence and pattern of internal testicular echoes. A characteristic echo pattern is obtained from the region of the rete testis but otherwise the testis remains echo-free. The hydrocele, testicular abscess, and tumor had appearances distinctly different from the usual pattern. The technique affords an accurate measurement of testicular thickness and may have the potential for revealing early testicular neoplasms.     

超声背向散射积分在乳腺肿块鉴别诊断中的应用

目的 应用超声背向散射积分(IBS)技术对乳腺肿块进行检测,探讨IBS值在乳腺肿瘤鉴别诊断中的价值。方法 对152例以乳腺肿块收入院拟进行手术的患者行IBS联机分析测定和常规彩超检测,术后进行病理追踪,按病理结果分组分析,确立诊断乳腺肿瘤的IBS的临界值 。然后设立验证组,验证该标准诊断乳腺疾病的敏感度、特异度、阳性预测值(+PV)、阴性预测值(－PV)、准确率。结果 各组疾病的IBS值的诊断点各不相同(P均＜0.05)。PPI与SDI结果的一致性较好,PPI(SDI)的敏感度高,AII特异度高。PPI或AII与常规彩超结合,上述指标均有明显的提高。结论 应用IBS值有助于乳腺肿块良恶性的鉴别,IBS为无创判定乳腺疾病提供了一项数字半量化的新指标;与常规彩超联合应用可取长补短,可明显提高乳腺超声检查的准确率。     

Computer-aided diagnosis for the classification of breast masses in automated whole breast ultrasound images

New automated whole breast ultrasound (ABUS) machines have recently been developed and the ultrasound (US) volume dataset of the whole breast can be acquired in a standard manner. The purpose of this study was to develop a novel computer-aided diagnosis system for classification of breast masses in ABUS images. One hundred forty-seven cases (76 benign and 71 malignant breast masses) were obtained by a commercially available ABUS system. Because the distance of neighboring slices in ABUS images is fixed and small, these continuous slices were used for reconstruction as three-dimensional (3-D) US images. The 3-D tumor contour was segmented using the level-set segmentation method. Then, the 3-D features, including the texture, shape and ellipsoid fitting were extracted based on the segmented 3-D tumor contour to classify benign and malignant tumors based on the logistic regression model. The Student’s t test, Mann-Whitney U test and receiver operating characteristic (ROC) curve analysis were used for statistical analysis. From the Az values of ROC curves, the shape features (0.9138) are better than the texture features (0.8603) and the ellipsoid fitting features (0.8496) for classification. The difference was significant between shape and ellipsoid fitting features (p = 0.0382). However, combination of ellipsoid fitting features and shape features can achieve a best performance with accuracy of 85.0% (125/147), sensitivity of 84.5% (60/71), specificity of 85.5% (65/76) and the area under the ROC curve Az of 0.9466. The results showed that ABUS images could be used for computer-aided feature extraction and classification of breast tumors. (E-mail: rfchang@csie.ntu.edu.tw)     

超声弹性成像鉴别诊断乳腺小实性病灶的价值

目的：探讨超声弹性成像在乳腺小病灶（最大直径≤10mm）鉴别诊断中的价值。方法：对经手术病理证实的直径≤10mm的142个乳腺病灶（Ⅰ组）和直径〉10mm的421个乳腺病灶（Ⅱ组）进行超声弹性成像检查，采用弹性成像5分评分法进行评价，弹性成像评分4分及以上为恶性，3分及以下为良性。以手术病理结果为诊断金标准，计算超声弹性成像诊断乳腺良恶性病变的敏感性、特异性和准确性。结果：Ⅰ组超声弹性成像诊断的敏感性、特异性和准确性分别为90．9％、99．2％、98．6％，Ⅱ组超声弹性成像诊断的敏感性、特异性和准确性分别为82．5％、97．6％、93．1％。结论：超声弹性成像有助于乳腺小病灶良恶性的鉴别诊断。     

3D spatial compounding of ultrasound images using image-based nonrigid registration

Medical ultrasound images are often distorted enough to significantly limit resolution during compounding (i.e., summation of images from multiple views). A new, volumetric image registration technique has been used successfully to enable high spatial resolution in three-dimensional (3D) spatial compounding of ultrasound images. Volumetric ultrasound data were acquired by scanning a linear matrix array probe in the elevational direction in a focal lesion phantom and in a breast in vivo. To obtain partly uncorrelated views, the volume of interest was scanned at five different transducer tilt angles separated by 4° to 6°. Pairs of separate views were registered by an automatic procedure based on a mutual information metric, using global full affine and thin-plate spline warping transformations. Registration accuracy was analyzed automatically in the phantom data, and manually in vivo, yielding average registration errors of 0.31 mm and 0.65 mm, respectively. In the vicinity of the warping control points, registrations obtained with warping transformations were significantly more accurate than full affine registrations. Compounded images displayed the expected reduction in speckle noise and increase in contrast-to-noise ratio (CNR), as well as better delineation of connective tissues and reduced shadowing. Compounding also revealed some apparent low contrast lobulations that were not visible in the single-sweep images. Given expected algorithmic and hardware enhancements, nonrigid, image-based registration shows great promise for reducing tissue motion and refraction artifacts in 3D spatial compounding.     

Performance evaluation of ultrasonic Nakagami image in tissue characterization

Conventional ultrasonic B-mode images qualitatively describe tissue structures but are unsuitable for quantitative analyses of scatterer properties. We have recently developed an ultrasonic parametric imaging technique based on the Nakagami statistical distribution that is able to quantify scatterer concentrations. The aim of the present study is to further explore both the behavior of a Nakagami image in characterizing different scatterer structures at different signal-to-noise ratios (SNRs) and the feasibility of Nakagami imaging using a general commercial ultrasound scanner for tissue examinations. Simulations, experiments on a tissue-mimicking phantom and in vitro measurements on a muscle tissue before and after microwave treatment were carried out. The SNR and contrast-to-noise ratio (CNR) were estimated to quantify image performance. The results demonstrate that a Nakagami image can differentiate different scatterer concentrations for single, hypoechoic and hyperechoic targets. Also, a Nakagami image, when combined with an ultrasound scanner, can complement the B-scan to characterize tissue and to identify the region of interest with a larger CNR. However, the noise effect can degrade the performance of a Nakagami image. When the signal SNR decreased to 15 dB in simulations and to 8 dB in experiments, the CNR of the hyperechoic Nakagami image decreased by 4% and 27%, respectively, and that of the hypoechoic one decreased by 42% and 80%, respectively. These results indicate that a Nakagami image behaves well in identifying regions with high scatterer concentrations but does not perform well when both the scatterer concentration and SNR are low.     

超声弹性成像在乳腺良恶性肿块鉴别诊断中的应用

超声弹性成像技术是近年来发展较快的一种超声诊断新技术,为乳腺肿块良、恶性的鉴别诊断提供了新的途径.本文就超声弹性成像的原理、检查技术、图像的分析方法、评分标准、临床应用、局限性及展望做一综述.     

Study of ultrasonic contrast agents using a dual-frequency band technique

We have developed a dual-frequency band technique to study frequency-dependent phenomena associated with ultrasonic contrast agents. Our technique uses a superimposed high-frequency (10 MHz) broad-band ultrasound (US) pulse to investigate contrast agent interaction with a low-frequency (e.g., 0.5 MHz) ultrasonic field. Our digitally controlled system has the ability to produce two colinear, confocal US pulses at different center frequencies, to adjust the relative phasing and pulse repetition frequency of each pulse, and to acquire digital backscatter data. A series of experimental studies demonstrated that the high-frequency backscatter signal responded to several phenomena induced in contrast agent particles by the low-frequency beam. These phenomena included radial pulsations, nonlinear oscillations and depletion. Initial results also demonstrated a relative phase shift between the high- and low-frequency signals; this shift is due to a difference in sound velocity at these frequencies, and it may convey information about the contrast agent concentration.     

Classification of hyperspectral images using self-training and a pseudo validation set

It is difficult to acquire sufficient labeled samples to classify remote sensing images. Supervised classification of hyperspectral remote sensing images does not perform well in the absence of an adequate amount of labeled samples of good quality. Self-training is a popular semi-supervised approach to train a classifier using fewer labeled samples and a pool of large unlabeled samples in an iterative manner. However, the traditional self-training approach is designed to use only confident samples. This criteria need not improve the accuracy of the classifier. In this work, the use of informative and confident samples for self-training has been proposed. The technique to selects informative and confident samples using pseudo validation set is a novel attempt to improve the scope of self-training. The pseudo validation set has been extracted from a freshly classified pool of pixels based on neighborhood information. The experimental results with popular benchmark hyperspectral images show that the proposed method has the potential to perform well.     

Histological analysis of ultrasonic images of the prostate: an accurate technique.

Ultrasonography of the prostate furnishes images which still cannot be fully interpreted morphologically. In a cadaver study, the ultrasound images of two groups (21 and 19) prostates, obtained in a water bath, were compared with histology slides taken at corresponding levels. In the first part of the study, using a 4 MHz probe, there was a correlation between hyperechoic lesions and stone formations in 9 out of 15 cases. A relation between hypoechoic lesions and the existence of a carcinoma could also be established in 4 out of 12 cases. In the second part of the study, using a 7 MHz probe, there was a correlation between hyperechoic lesions and stone formations in all cases. Hypoechoic lesions correlated with the presence of a carcinoma in 1 out of 8 cases. The technique used appears to be well suited for the comparative study of ultrasound images and histology. Application of the 7 MHz probe is preferable as, because of a better resolution, smaller lesions can be detected. The results of this study are not very encouraging for the use of transrectal ultrasound for the detection of small prostatic carcinomas.     

S-DetectTM分类技术诊断BI-RADS 4类乳腺肿块

目的 观察S-Detect^TM分类技术鉴别诊断BI-RADS 4类乳腺良恶性肿块的价值。方法 对94例经二维超声诊断为BI-RADS 4类乳腺肿块患者（共104个肿块）行S-Detect^TM分类技术检查，以手术或穿刺活检病理结果作为金标准，评价S-Detect^TM分类技术、BI-RADS分类及二者联合应用诊断乳腺BI-RADS 4类良恶性肿块的价值。结果 104个乳腺肿块，经病理确诊为良性41个、恶性63个。S-Detect^TM分类技术诊断乳腺BI-RADS 4a类乳腺肿块的敏感度（SE）66.67%，特异度（SP）89.29%、阳性预测值（PPV）57.14%、阴性预测值（NPV）92.59%；对乳腺BI-RADS 4b类肿块分别为90.91%、60.00%、88.24%及66.67%；对乳腺BI-RADS 4c类肿块分别为95.83%、66.67%、95.83%及66.67%。S-Detect^TM分类技术联合BI-RADS分类诊断乳腺肿块的SE、SP、准确率明显均高于单独运用（P均<0.05）。结论 S-Detect^TM分类技术判断乳腺BI-RADS 4a类良性肿块、BI-RADS 4b类及BI-RADS 4c类恶性肿块均有较高价值。S-Detect^TM分类技术联合BI-RADS分类可明显提高鉴别BI-RADS 4类乳腺良恶性肿块的效能。     

Computer model of ultrasonic hyperthermia and ablation for ocular tumors using B-mode data.

Computer simulations have been conducted to examine hyperthermia and ablation for treating ocular tumors. An interactive software package has been implemented that permits relevant tissue dimensions to be determined from B-mode data. This package also permits interactive beam positioning, and it provides image displays depicting computed absorbed doses and temperature rises. Results are presented showing how hyperthermia temperature patterns are influenced by beam position, beam geometry and frequency. Images showing ablative temperature rises at various time intervals are also presented. For hyperthermia, geometric models of beam profiles showed that a non-uniform beam pattern (with a central low-intensity region) can produce more uniform heating of small ocular tumors than a beam with a uniform intensity profile. For a given tumor, the uniformity of hyperthermia temperatures was found to be a function of frequency, with 4.75 MHz providing reasonably uniform results for typical tumor heights (near 7 mm). For ablation, diffraction computations were employed to calculate beam intensity profiles; results show an initially rapid rise in temperature levels with subsequent, slower heating beyond the -3-dB limits of the focal volume. The model is now being refined, and additional phenomena, including nonlinear propagation, will be incorporated.     

超声弹性成像应变率比值法在乳腺良恶性肿块中的诊断价值

目的探讨超声弹性成像应变率比值对乳腺肿块良恶性的诊断准确性。方法对162例患者的173个乳腺肿块分别进行常规超声和超声弹性成像超声检查,测量肿块周围正常乳腺组织与肿块组织间的超声弹性成像应变率比值(E2/E1),并将结果与术后病理结果进行比较。结果超声弹性成像应变率比值法诊断乳腺肿块的ROC曲线下面积为0.950(95%CI,0.916～0.983),取约登指数最大值(0.779)时的截断点,确定E2/E1=0.355为诊断良恶性肿块的最佳临界点。良性肿块的E2/E1为0.57±0.15,恶性肿块的E2/E1为0.29±0.08,二者比较差异有统计学意义(P<0.05)。超声弹性成像应变率比值法与常规超声对乳腺恶性肿块的敏感性差异无统计学意义,但联合应用二者较单独应用超声弹性成像应变率比值法敏感性有显著提高(P<0.05)。结论超声弹性成像应变率比值作为一项相对客观诊断乳腺良恶性病变的半定量指标,具有较高的诊断准确性,与常规超声联合应用可提高乳腺恶性肿块的检出率。     

Small calcification depiction in ultrasound B-mode images using decorrelation of echoes caused by forward scattered waves

Purpose  

The purpose of this study is to propose a novel method to depict small calcifications in ultrasound B-mode images using decorrelation of forward scattered waves with no decrease in the frame rate.