Jiawei Tian

Probability density difference-based active contour for ultrasound image segmentation

Because of its low signal/noise ratio, low contrast and blurry boundaries, ultrasound (US) image segmentation is a difficult task. In this paper, a novel level set-based active contour model is proposed for breast ultrasound (BUS) image segmentation. At first, an energy function is formulated according to the differences between the actual and estimated probability densities of the intensities in different regions. The actual probability densities are calculated directly. For calculating the estimated probability densities, the probability density estimation method and background knowledge are utilized. The energy function is formulated with level set approach, and a partial differential equation is derived for finding the minimum of the energy function. For performing numerical computation, the derived partial differential equation is approximated by the central difference and non-re-initialization approach. The proposed method was operated on both the synthetic images and clinical BUS images for studying its characteristics and evaluating its performance. The experimental results demonstrate that the proposed method can model the BUS images well, be robust to noise, and segment the BUS images accurately and reliably.

The value of strain ratio in differential diagnosis of thyroid solid nodules.

OBJECTIVE To assess the clinical value of strain ratio in differentiating thyroid solid nodules and explore its distribution characters based on pathological results. MATERIALS AND METHODS The study was approved by the ethic committee and the informed consents were signed. Ninety nine solid thyroid nodules (67 benign and 32 malignant) from 71 female (mean age 46.3 ± 9.8 years) and 28 male (mean age 54.9 ± 11.7 years) patients were evaluated. Five radiologists evaluated the nodules based on a four-degree elastography score system. Strain ratio was calculated on-line. Diagnostic performances of the two evaluations were compared using Receiver Operating Characteristic (ROC) curves. Values of different pathological nodules were compared by one-way ANOVA. RESULTS Areas under the ROC curve (AUC) of the five readers were 0.82, 0.81, 0.79, 0.73 and 0.83, respectively. The AUC of strain ratio evaluation was higher (0.88 vs. 0.79, p < 0.001) than that of the ES score evaluation. Best cut-off points of the two evaluations were 3.5 (82% sensitivity, 72% specificity) and 4.225 (81% sensitivity, 83% specificity), respectively. Both the ES score and strain ratio were higher for malignant nodules than that for benign ones (p < 0.001). CONCLUSIONS Strain ratio was a useful index in differential diagnosis of thyroid solid nodules. It can provide quantitative information on thyroid nodule characterization and improve diagnostic confidence. The best cut-off point for benign and malignant nodules was 4.2.

A novel approach to speckle reduction in ultrasound imaging.

Speckle noise is inherent in ultrasound images, and it generally tends to reduce the resolution and contrast, thereby degrading the diagnostic accuracy of this modality. Speckle reduction is very important and critical for ultrasound imaging. In this paper, we propose a novel approach for speckle reduction using 2-D homogeneity and directional average filters. We have conducted experiments on numerous artificial images, clinic breast ultrasound images and vascular images. The experimental results are compared with that of other methods and the performance is evaluated using several merits, and they demonstrate that the proposed approach can reduce the speckle noise effectively without blurring the edges and damaging the textual information. It will be very useful for computer-aided diagnosis systems using ultrasound images.

Automated Segmentation of Ultrasonic Breast Lesions Using Statistical Texture Classification and Active Contour Based on Probability Distance

Because of its complicated structure, low signal/noise ratio, low contrast and blurry boundaries, fully automated segmentation of a breast ultrasound (BUS) image is a difficult task. In this paper, a novel segmentation method for BUS images without human intervention is proposed. Unlike most published approaches, the proposed method handles the segmentation problem by using a two-step strategy: ROI generation and ROI segmentation. First, a well-trained texture classifier categorizes the tissues into different classes, and the background knowledge rules are used for selecting the regions of interest (ROIs) from them. Second, a novel probability distance-based active contour model is applied for segmenting the ROIs and finding the accurate positions of the breast tumors. The active contour model combines both global statistical information and local edge information, using a level set approach. The proposed segmentation method was performed on 103 BUS images (48 benign and 55 malignant). To validate the performance, the results were compared with the corresponding tumor regions marked by an experienced radiologist. Three error metrics, true-positive ratio (TP), false-negative ratio (FN) and false-positive ratio (FP) were used for measuring the performance of the proposed method. The final results (TP = 91.31%, FN = 8.69% and FP = 7.26%) demonstrate that the proposed method can segment BUS images efficiently, quickly and automatically.

Fractional subpixel diffusion and fuzzy logic approach for ultrasound speckle reduction

Speckle is the dominant source of noise in ultrasound imaging and is a kind of multiplicative noise. It is difficult to design a filter to remove speckle effectively. In this paper, a novel fuzzy subpixel fractional partial difference (FSFPD) for ultrasound speckle reduction is proposed. Euler-Lagrange equation acts as an increasing function of the fractional derivatives absolute value of the image intensity function. The fractional order partial difference is computed in the frequency and fuzzy domain with subpixel precision. We test the proposed method on both synthetic and real breast ultrasound (BUS) images. The comparisons of the experimental results show that the proposed method can preserve edges and structural details of ultrasound images well while removing speckle noise. In addition, the filtered images are assessed and evaluated by radiologists using double blind method. The results demonstrate that the discrimination rate of breast cancers has been highly improved after employing the proposed method.

A novel breast ultrasound image segmentation algorithm based on neutrosophic similarity score and level set

Breast ultrasound (BUS) image segmentation is a challenging task due to the speckle noise, poor quality of the ultrasound images and size and location of the breast lesions. In this paper, we propose a new BUS image segmentation algorithm based on neutrosophic similarity score (NSS) and level set algorithm. At first, the input BUS image is transferred to the NS domain via three membership subsets T, I and F, and then, a similarity score NSS is defined and employed to measure the belonging degree to the true tumor region. Finally, the level set method is used to segment the tumor from the background tissue region in the NSS image. Experiments have been conducted on a variety of clinical BUS images. Several measurements are used to evaluate and compare the proposed methods performance. The experimental results demonstrate that the proposed method is able to segment the BUS images effectively and accurately.

Integrating multi-omics for uncovering the architecture of cross-talking pathways in breast cancer.

Cross-talk among abnormal pathways widely occurs in human cancer and generally leads to insensitivity to cancer treatment. Moreover, alterations in the abnormal pathways are not limited to single molecular level. Therefore, we proposed a strategy that integrates a large number of biological sources at multiple levels for systematic identification of cross-talk among risk pathways in cancer by random walk on protein interaction network. We applied the method to multi-Omics breast cancer data from The Cancer Genome Atlas (TCGA), including somatic mutation, DNA copy number, DNA methylation and gene expression profiles. We identified close cross-talk among many known cancer-related pathways with complex change patterns. Furthermore, we identified key genes (linkers) bridging these cross-talks and showed that these genes carried out consistent biological functions with the linked cross-talking pathways. Through identification of leader genes in each pathway, the architecture of cross-talking pathways was built. Notably, we observed that linkers cooperated with leaders to form the fundamentation of cross-talk of pathways which play core roles in deterioration of breast cancer. As an example, we observed that KRAS showed a direct connection to numerous cancer-related pathways, such as MAPK signaling pathway, suggesting that it may be a central communication hub. In summary, we offer an effective way to characterize complex cross-talk among disease pathways, which can be applied to other diseases and provide useful information for the treatment of cancer.

Is transvaginal elastography useful in pre-operative diagnosis of cervical cancer?

OBJECTIVE To evaluate the clinical value of transvaginal elastography (TVES) in diagnosing cervical malignancies by detecting changes of tissue stiffness. METHODS One hundred and ten consecutive patients with cervical lesions were enrolled. Pathological results were used as the gold standards. TVES was employed to detect the stiffness changes of the cervix. Strain ratio was calculated and compared between the benign and malignant lesions. Depth of invasion into stromas of 56 cases of cervical cancers measured by TVES were recorded and compared with the pathological results. Interclass correlation coefficient (ICC) was used to analyze the reproducibility. RESULTS Strain ratio of malignant lesions were much higher than that of the benign lesions (8.19±5.66 vs. 2.81±2.24, P<0.01). Area under the curve (AUC) was 0.905 with a 95% CI (0.835-0.976). The best cut-off point of strain ratio value was 4.53. Specificity and sensitivity for the best cut-off point were 0.788 and 0.897, respectively. Mean depth of the 56 malignant lesions was 17.8±7.4mm measured by TVES (range 5.4-43.1mm) and 11.5±8.8mm measured by pathological samples (range 3.7-38.4mm). ICC of the 2 methods were 0.87 (95% CI 0.863-0.947) and 0.931 (95% CI 0.902-0.952) for the 2 observers. CONCLUSIONS TVES was a useful technique in confirming the diagnoses of cervical cancer and in estimating the infiltrating region. When the strain ratio of a cervical lesion was higher than 4.53, it is confidential to be diagnosed as malignant.

Identifying ultrasound and clinical features of breast cancer molecular subtypes by ensemble decision.

Breast cancer is molecularly heterogeneous and categorized into four molecular subtypes: Luminal-A, Luminal-B, HER2-amplified and Triple-negative. In this study, we aimed to apply an ensemble decision approach to identify the ultrasound and clinical features related to the molecular subtypes. We collected ultrasound and clinical features from 1,000 breast cancer patients and performed immunohistochemistry on these samples. We used the ensemble decision approach to select unique features and to construct decision models. The decision model for Luminal-A subtype was constructed based on the presence of an echogenic halo and post-acoustic shadowing or indifference. The decision model for Luminal-B subtype was constructed based on the absence of an echogenic halo and vascularity. The decision model for HER2-amplified subtype was constructed based on the presence of post-acoustic enhancement, calcification, vascularity and advanced age. The model for Triple-negative subtype followed two rules. One was based on irregular shape, lobulate margin contour, the absence of calcification and hypovascularity, whereas the other was based on oval shape, hypovascularity and micro-lobulate margin contour. The accuracies of the models were 83.8%, 77.4%, 87.9% and 92.7%, respectively. We identified specific features of each molecular subtype and expanded the scope of ultrasound for making diagnoses using these decision models.

Early Evaluation of Relative Changes in Tumor Stiffness by Shear Wave Elastography Predicts the Response to Neoadjuvant Chemotherapy in Patients With Breast Cancer

Neoadjuvant chemotherapy plays an important role in comprehensive therapy for breast cancer, but response prediction is imperfect. Shear wave elastography (SWE) is a novel technique that can quantitatively evaluate tissue stiffness. In this study, we sought to investigate the application value of SWE for early prediction of the response to neoadjuvant chemotherapy in patients with breast cancer.