Search

Coronary computed tomographic angiography (CTA) image segmentation plays an important role in many practical applications, such as assisting doctors to judge vascular occlusion, vascular disease diagnosis, etc. In view of the large amount of noise in CTA images and not delicate segmentation results of traditional deep learning algorithms (including FCN, U-Net, V-Net, etc.), this paper proposes a global feature and multi-level feature aggregation network, which includes three network models, global feature module, feature fusion and refined V-shape multi-level feature aggregation module, and deep supervision module. The global feature module can filter the original CTA image and generate the basic features by integrating the early and later feature information and integrating the rich details and semantic information. The refined V-shape module generates refined feature maps of different levels on the basis of basic features, and obtains accurate coronary segmentation images by aggregating the refined feature maps of different levels. In addition, a deep supervision mechanism is added after each refined V-shape module to avoid the problem of gradient disappearing. The results show that the proposed method is superior to the mainstream baseline intuitively and quantitively. The ablation experiments also prove the effectiveness of each module.

Background: Patients with uterine cervical cancer suffer high mortality. Accurate detection of a residual tumor by magnetic resonance imaging (MRI) during and after directed brachytherapy (BCT) is crucial for the success of cancer treatment and is a significant predictor of patient survival.Purpose: To determine the diagnostic significance of MRI in detecting residual tumor tissue after BCT.Material and Methods: The Web of Knowledge, Cochrane Library, and PubMed were systematically searched (January 1997 to December 2016) for post-brachytherapy MRI studies that measured residual tumors in patients with uterine cervical cancer. All data were analyzed using the Meta-Disc 1.4 program.Results: Four clinical studies consisting of 163 patients (147 of whom were included in the present analysis) who were diagnosed with uterine cervical cancer according to the International Federation of Gynecology and Obstetrics (FIGO) staging system were included in the study. All the patients received BCT and underwent MRI detection of residual tumors tissue. In studies where the accuracy of MRI detection was confirmed by histological tests or gynecological tests, the summary estimates of specificity, sensitivity, positive predictive value, negative predictive value, and accuracy were 88.5%, 83.5%, 53.5%, 97.1%, and 84.3%, respectively.Conclusion: MRI-directed BCT is commonly used for cervical cancer patients. Based on our investigation of four independent studies, MRI showed better prediction of positive results than negative results in patients with cervical cancer after BCT. However, more data on the greater numbers of patients are needed to establish the accuracy of MRI detection of cervical cancer after BCT. [ABSTRACT FROM AUTHOR]

The aim of the present study was to explore the association between the parameters of diffusion tensor imaging (DTI), including fractional anisotropy (FA) values, apparent diffusion coefficient (ADC) values and the diffusion tensor tractography (DTT) map, with the Gleason score of prostate cancer (PCa). A retrospective study of 50 cases of PCa confirmed by biopsy or surgical pathology was performed. Conventional magnetic resonance imaging and DTI scans were conducted in these cases. The 50 cases of PCa were divided into three groups, including low, intermediate and high grade, according to the Gleason score. Post-DTI processing was performed using Neuro 3D software, in order to measure the FA and ADC values, and map the prostate fibers. Differences in FA and ADC values among the various PCa groups were examined using analysis of variance, while the correlation of FA and ADC values with the Gleason score was studied using Pearson correlation analysis. The obtained DTT map clearly demonstrated the spatial structure of the prostate fibers. The fibers of the cancer area were dense without interruption in the low-grade group, sparse and disordered in the intermediate-grade group, and were disordered, sparse or even absent in the high-grade group. The FA values were 0.284±0.313, 0.293±0.347 and 0.369±0.347, respectively, with statistically significant differences observed among the three groups (F=234.533; P<0.05) and between each group (P<0.05). In addition, the FA value of PCa was positively correlated with the Gleason score (r=0.884; P<0.05). The ADC values of the low-, intermediate- and high-grade groups were 1.070±0.072x10-3, 0.961±0.081x10-3 and 0.821±0.048x10-3, respectively, which demonstrated statistically significant differences among the three groups (F=49.987; P<0.05) and between each group (P<0.05). Furthermore, the ADC values of PCa were negatively correlated with Gleason score (r=-0.810; P<0.05). In conclusion, there was an association between DTI parameters and Gleason score, which may be used to evaluate the grading and prognosis of PCa. [ABSTRACT FROM AUTHOR]

We live in a 3D world where people interact with each other in the environment. Learning 3D posed humans therefore requires us to perceive and interpret these interactions. This paper proposes LEAPSE, a novel method that learns salient instance affordances for estimating a posed body from a single RGB image in a non-parametric manner. Existing methods mostly ignore the environment and estimate the human body independently from the surroundings. We capture the influences of non-contact and contact instances on a posed body as an adequate representation of the "environment affordances". The proposed method learns the global relationships between 3D joints, body mesh vertices, and salient instances as environment affordances on the human body. LEAPSE achieved state-of-the-art results on the 3DPW dataset with many affordance instances, and also demonstrated excellent performance on Human3.6M dataset. We further demonstrate the benefit of our method by showing that the performance of existing weak models can be significantly improved when combined with our environment affordance module.