No abstract

No abstract

Acquisition of the standard plane is the prerequisite of biometric measurement and diagnosis during the ultrasound (US) examination. In this paper, a new algorithm is developed for the automatic recognition of the fetal facial standard planes (FFSPs) such as the axial, coronal, and sagittal planes. Specifically, densely sampled root scale invariant feature transform (RootSIFT) features are extracted and then encoded by Fisher vector (FV). The Fisher network with multi-layer design is also developed to extract spatial information to boost the classification performance. Finally, automatic recognition of the FFSPs is implemented by support vector machine (SVM) classifier based on the stochastic dual coordinate ascent (SDCA) algorithm. Experimental results using our dataset demonstrate that the proposed method achieves an accuracy of 93.27% and a mean average precision (mAP) of 99.19% in recognizing different FFSPs. Furthermore, the comparative analyses reveal the superiority of the proposed method based on FV over the traditional methods.

Background:Radical nephrectomy with thrombectomy is one of the most difficult and complicated urological operations. But the roles of renal tumor volume and thrombus level in surgical complexity and prognostic outcome are not clear. This study aimed to evaluate the surgical complexity and prognostic outcome between the volume of renal cell carcinoma (RCC) and the level of venous tumor thrombus.Methods:The clinical data of 67 RCC cases with renal vein or inferior vena cava (IVC) tumor thrombus from January 2015 to May 2018 were retrospectively analyzed. Among these 67 cases, 21 (31.3%) were small tumors with high-level thrombus (tumor ≤7 cm in diameter and thrombus Neves Level II–IV), while 46 (68.7%) were large tumors with low-level thrombus group (tumor >7 cm in diameter and thrombus Level 0–I). Clinical features, operation details, and pathology data were collected. Univariable and multivariable logistic regression analyses were applied to evaluate the risk factors for small tumor with high-level thrombus.Results:Patients with small tumors and high-level thrombus were more likely to have longer operative time (421.9 ± 135.1 min vs. 282.2 ± 101.9 min, t = 4.685, P < 0.001), more surgical bleeding volume (1200 [325, 2900] mL vs. 500 [180, 1000] mL, U = 270.000, P = 0.004), more surgical blood transfusion volume (800 [0, 1400] mL vs. 0 [0, 800] mL, U = 287.500, P = 0.004), more plasma transfusion volume (0 [0, 800] mL vs. 0 [0, 0] mL, U = 319.000, P = 0.004), higher percentage of open operative approach (76.2% vs. 32.6%, χ2 = 11.015, P = 0.001), higher percentage of IVC resection (33.3% vs. 0%, χ2 = 17.122, P < 0.001), and higher percentage of post-operative complications (52.4% vs. 19.6%, χ2 = 7.415, P = 0.010) than patients with large tumors and low-level thrombus. In multivariate analysis, decreased hemoglobin (Hb) (odds ratio [OR]: 0.956, 95% confidence interval [CI]: 0.926–0.986, P = 0.005) and non-sarcomatoid differentiation (OR: 0.050, 95% CI: 0.004–0.664, P = 0.023) were more likely to form small tumors with high-level tumor thrombus rather than large tumor with small tumor thrombus. The estimated mean cancer-specific survival times of small tumor with high-level thrombus and large tumor with low-level thrombus were 31.6 ± 3.8 months and 32.5 ± 2.9 months, without statistical significance (P = 0.955). After univariate and multivariate Cox proportional hazard survival regression analyses, only distant metastasis (hazard ratio [HR]: 3.839, P = 0.002), sarcomatoid differentiation (HR: 7.923, P < 0.001), alkaline phosphatase (HR: 2.661, P = 0.025), and severe post-operative complications (HR: 10.326, P = 0.001) were independent predictors of prognosis.Conclusions:The level of the tumor thrombus was more important than the diameter of the primary kidney tumor in affecting the complexity of surgery. In the same T3 stage, neither the renal tumor diameter nor the tumor thrombus level was an independent risk factor for prognosis.

With a focus on service interruptions occurring in optical networks, we propose a failure prognostics scheme based on a bi-directional gated recurrent unit (BiGRU) from the perspective of time-series processing, which leverages actual datasets from the network operator. BiGRU neural networks can capture the temporal features of multi-sourced data and incorporate contextual information. A principal component analysis is introduced to reduce the data dimensionality. Experimental results show that the average accuracy of the prognostics, F1 score, false positive rate, and false negative rate of our method are 99.61%, 99.63%, 0.29%, and 0.84%, respectively, which proves the feasibility of the proposed scheme for failure prognostics of equipment used in optical networks.