Search

Ultrasound is a critical non-invasive test for preoperative diagnosis of ovarian cancer. Deep learning is making advances in image-recognition tasks; therefore, we aimed to develop a deep convolutional neural network (DCNN) model that automates evaluation of ultrasound images and to facilitate a more accurate diagnosis of ovarian cancer than existing methods.In this retrospective, multicentre, diagnostic study, we collected pelvic ultrasound images from ten hospitals across China between September 2003, and May 2019. We included consecutive adult patients (aged ≥18 years) with adnexal lesions in ultrasonography and healthy controls and excluded duplicated cases and patients without adnexa or pathological diagnosis. For DCNN model development, patients were assigned to the training dataset (34 488 images of 3755 patients with ovarian cancer, 541 442 images of 101 777 controls). For model validation, patients were assigned to the internal validation dataset (3031 images of 266 patients with ovarian cancer, 5385 images of 602 with benign adnexal lesions), external validation datasets 1 (486 images of 67 with ovarian cancer, 933 images of 268 with benign adnexal lesions), and 2 (1253 images of 166 with ovarian cancer, 5257 images of 723 benign adnexal lesions). Using these datasets, we assessed the diagnostic value of DCNN, compared DCNN with 35 radiologists, and explored whether DCNN could augment the diagnostic accuracy of six radiologists. Pathological diagnosis was the reference standard.For DCNN to detect ovarian cancer, AUC was 0·911 (95% CI 0·886-0·936) in the internal dataset, 0·870 (95% CI 0·822-0·918) in external validation dataset 1, and 0·831 (95% CI 0·793-0·869) in external validation dataset 2. The DCNN model was more accurate than radiologists at detecting ovarian cancer in the internal dataset (88·8% vs 85·7%) and external validation dataset 1 (86·9% vs 81·1%). Accuracy and sensitivity of diagnosis increased more after DCNN-assisted diagnosis than assessment by radiologists alone (87·6% [85·0-90·2] vs 78·3% [72·1-84·5], p0·0001; 82·7% [78·5-86·9] vs 70·4% [59·1-81·7], p0·0001). The average accuracy of DCNN-assisted evaluations for six radiologists reached 0·876 and were significantly augmented when they were DCNN-assisted (p0·05).The performance of DCNN-enabled ultrasound exceeded the average diagnostic level of radiologists matched the level of expert ultrasound image readers, and augmented radiologists' accuracy. However, these observations warrant further investigations in prospective studies or randomised clinical trials.National Key Basic Research Program of China, National Sci-Tech Support Projects, and National Natural Science Foundation of China.

Background Oxidative stress and inflammation play important roles in the neuronal injury caused by intracerebral hemorrhage (ICH). Uric acid (UA), an important natural antioxidant, might reduce the neuronal injury caused by ICH. Delineating the relationship between UA and ICH will enhance our understanding of antioxidative mechanisms in recovery from ICH. Methods We conducted a retrospective study of 325 patients with acute supratentorial ICH to investigate the relationship between serum UA levels and hematoma volumes and prognosis. A hematoma volume of ≥30 mL was defined as a large hematoma. An unfavorable outcome was defined as a modified Rankin scale score of 4–6 on day 30. Results The serum UA level was significantly lower in the patients with a large hematoma volume (median, 306 μmol/L; 25th to 75th percentile, 243–411 μmol/L) than in those with a small hematoma volume (median, 357 μmol/L; 25th to 75th percentile, 271–442 μmol/L; P = 0.012). Similarly, the unfavorable outcome group had had lower serum UA levels (median, 309 vs. 363 μmol/L; P = 0.009) compared with the favorable outcome group. The results of the multivariate logistic analysis indicated that a lower serum UA level was associated with a larger hematoma volume (odds ratio, 0.996; P = 0.006) and an unfavorable outcome (odds ratio, 0.997; P = 0.030). Conclusions The results from the present study have indicated that in patients with acute supratentorial ICH, a low serum UA level might indicate that the patient has a large hematoma volume and might be a risk factor for a poor day 30 functional prognosis.

Wastewater treatment plants (WWTPs) play an important role within the urban water cycle in preventing the natural water environment and human health from being negatively affected by human activities. However, wastewater treatment processes such as the discharge of effluent and indirect emissions resulting from energy or chemical production, also negatively affect the environment. Footprints have been used to track human influence on the environment through different areas of concern and have been applied in assessing the sustainability of WWTPs. A comprehensive review of footprint assessment for evaluating the wastewater treatment process in WWTPs was investigated in this article. The review showed that the carbon footprint and the water footprint were the two of the most popular footprints for evaluating the sustainability of WWTPs, and other footprint (such as nitrogen and phosphorus footprints) assessment applications were also introduced to assess the eutrophication of water bodies. To promote the application of footprint assessment on WWTP, this article regulated the study goals, frameworks, system boundary, data treatment methods, resultant interpretation process. The pros and cons of footprint assessments were discussed and investigated in detail, several suggestions were come up with for the improvement of footprint assessments. The analyzation of footprint assessments under different WWTPs revealed that wastewater treatment technologies and scales significantly impacted the footprints. Moreover, the research hotspots identified by way of a keyword network diagram showed that researchers tended to propose new indicators and models for the grey water footprint and the carbon footprint respectively, and water-carbon-energy nexus was a promising direction for future studies. Finally, two possible application scenarios, element tracing through modified footprints and management improvement through mathematical optimization, were proposed to optimize the operation strategies of WWTPs. Overall, footprint assessment research for WWTPs is still at the initial stage and needs further exploration and improvement, e.g. introducing new tools, exploring system variation under various scenarios, combining other footprints, and implementing the nexus assessment.

Tensile stress occurs in the vicinity of upper surface of the specimen in the severe plastic deformation zone, which increases the cracking and fracture tendency of the specimen and impedes the further ECAP processing. In this paper, the conventional ECAP die (Ψ = 16° and Φ = 90°) was modified to eliminate the tensile stress and enhance the compressive stress in the severe plastic deformation zone, therefore reducing the cracking and fracture tendency of the specimen. Finite element analysis demonstrated that the stress state changes from tensile to strongly compressive when using the modified die. A modified die was made and employed to extrude the commercially pure aluminum to verify its effectiveness experimentally. The billet was successfully extruded for 20 passes without obvious surface defects with the modified die, compared to 13–14 passes at most for the conventional die. Consequently, much more fine and uniform microstructure was obtained with the average grain size of 200–300 nm, while the average grain size is ∼500 nm in the case of using the conventional die.