A Real-Time and Robust Neural Network Model for Low-Measurement-Rate Compressed-Sensing Image Reconstruction.

Compressed sensing (CS) is a popular data compression theory for many computer vision tasks, but the high reconstruction complexity for images prevents it from being used in many real-world applications. Existing end-to-end learning methods achieved real time sensing but lack theory guarantee for robust reconstruction results. This paper proposes a neural network called RootsNet, which integrates the CS mechanism into the network to prevent error propagation. So, RootsNet knows what will happen if some modules in the network go wrong. It also implements real-time and successfully reconstructed extremely low measurement rates that are impossible for traditional optimization-theory-based methods. For qualitative validation, RootsNet is implemented in two real-world measurement applications, i.e., a near-field microwave imaging system and a pipeline inspection system, where RootsNet easily saves 60% more measurement time and 95% more data compared with the state-of-the-art optimization-theory-based reconstruction methods. Without losing generality, comprehensive experiments are performed on general datasets, including evaluating the key components in RootsNet, the reconstruction uncertainty, quality, and efficiency. RootsNet has the best uncertainty performance and efficiency, and achieves the best reconstruction quality under super low-measurement rates.

Efficient Near-Field Radiofrequency Imaging of Impact Damage on CFRP Materials with Learning-Based Compressed Sensing.

Carbon fiber-reinforced polymer (CFRP) is a widely-used composite material that is vulnerable to impact damage. Light impact damages destroy the inner structure but barely show obvious change on the surface. As a non-contact and high-resolution method to detect subsurface and inner defect, near-field radiofrequency imaging (NRI) suffers from high imaging times. Although some existing works use compressed sensing (CS) for a faster measurement, the corresponding CS reconstruction time remains high. This paper proposes a deep learning-based CS method for fast NRI, this plugin method decreases the measurement time by one order of magnitude without hardware modification and achieves real-time imaging during CS reconstruction. A special 0/1-Bernoulli measurement matrix is designed for sensor scanning firstly, and an interpretable neural network-based CS reconstruction method is proposed. Besides real-time reconstruction, the proposed learning-based reconstruction method can further reduce the required data thus reducing measurement time more than existing CS methods. Under the same imaging quality, experimental results in an NRI system show the proposed method is 20 times faster than traditional raster scan and existing CS reconstruction methods, and the required data is reduced by more than 90% than existing CS reconstruction methods.

Pipeline In-Line Inspection Method, Instrumentation and Data Management.

Pipelines play an important role in the national/international transportation of natural gas, petroleum products, and other energy resources. Pipelines are set up in different environments and consequently suffer various damage challenges, such as environmental electrochemical reaction, welding defects, and external force damage, etc. Defects like metal loss, pitting, and cracks destroy the pipeline's integrity and cause serious safety issues. This should be prevented before it occurs to ensure the safe operation of the pipeline. In recent years, different non-destructive testing (NDT) methods have been developed for in-line pipeline inspection. These are magnetic flux leakage (MFL) testing, ultrasonic testing (UT), electromagnetic acoustic technology (EMAT), eddy current testing (EC). Single modality or different kinds of integrated NDT system named Pipeline Inspection Gauge (PIG) or un-piggable robotic inspection systems have been developed. Moreover, data management in conjunction with historic data for condition-based pipeline maintenance becomes important as well. In this study, various inspection methods in association with non-destructive testing are investigated. The state of the art of PIGs, un-piggable robots, as well as instrumental applications, are systematically compared. Furthermore, data models and management are utilized for defect quantification, classification, failure prediction and maintenance. Finally, the challenges, problems, and development trends of pipeline inspection as well as data management are derived and discussed.

Changes of foraging patch selection and utilization by a giant panda after bamboo flowering.

The bamboo flowering leads to the habitat fragmentation and food quality decline of a giant panda. Few empirical research has been conducted about the giant panda's response to the bamboo flowering. Here, we investigated the characteristics of bamboo stands, giant panda's activity, and selection and utilization of bamboo stands by giant panda in Taibaishan National Nature Reserve, China, over a 3-year period (September 2013-May 2016) during the Fargesia qinlingensis flowering period. Our results indicated that the proportion of whole bamboo stands flowering has gradually expanded from 26.7% in 2013 and 33.9% in 2014 to 52.3% in 2015. Although the flowering bamboo has lower crude protein and higher crude fiber than a non-flowering bamboo, the giant panda still fed on flowering bamboo from the evidence of droppings. The giant panda left its feeding sites and moved to the high elevation along river when the proportion of flowering reached 69.2% at elevation of 2350-2450 m in the third year. With the decline of the quality of bamboo stand of Fargesia qinlingensis, the giant panda abandoned its feeding sites when the threshold value of bamboo flowering reached 56.9-69.2%. Flexibility in foraging strategy and spatial behavior can help the giant panda to better adapt to the environment.

Study of in vitro RBCs membrane elasticity with AOD scanning optical tweezers.

The elasticity of red cell membrane is a critical physiological index for the activity of RBC. Study of the inherent mechanism for RBCs membrane elasticity transformation is attention-getting all along. This paper proposes an optimized measurement method of erythrocytes membrane shear modulus incorporating acousto-optic deflector (AOD) scanning optical tweezers system. By use of this method, both membrane shear moduli and sizes of RBCs with different in vitro times were determined. The experimental results reveal that the RBCs membrane elasticity and size decline with in vitro time extension. In addition, semi quantitative measurements of S-nitrosothiol content in blood using fluorescent spectrometry during in vitro storage show that RBCs membrane elasticity change is positively associated with the S-nitrosylation level of blood. The analysis considered that the diminished activity of the nitric oxide synthase makes the S-nitrosylation of in vitro blood weaker gradually. The main reason for worse elasticity of the in vitro RBCs is that S-nitrosylation effect of spectrin fades. These results will provide a guideline for further study of in vitro cells activity and other clinical applications.

[Neonatal outcomes of pregnancy resulting from in vitro fertilization and embryo transfer: analysis of 1274 cases].

OBJECTIVE: To evaluate the neonatal outcomes of pregnancies resulting from in vitro fertilization and embryo transfer (IVF-ET) in relation to the occurrence of congenital malformations of the neonates. METHODS: A total of 1274 infants born after IVF-ET were reviewed. The neonatal outcome was evaluated based on gestational weeks, body weight, congenital malformations, manner of spermatization, maternal age and multiple gestation. RESULTS: IVF-ET resulted in 930 deliveries, giving birth to a total of 1274 newborns. Spontaneous delivery occurred in 115 cases (12.37%), with preterm birth in 224 cases (24.09%). Among these newborns, 363 (28.49%) had very low born weight (VLBW), 13 (1.02%) had congenital malformations, and neonatal mortality occurred in 15 cases (1.18%). CONCLUSION: IVF increases the risks of twin pregnancies, preterm birth and VLBW, but does not increase the rate of congenital malformations and neonatal mortality. Intracytoplasmic sperm injection (ICSI) is at higher risk of congenital malformations. Maternal age and twin pregnancies are not associated with congenital malformations. IVF can be safe for treatment of infertility.

[Clinical application of gonadotropin-releasing hormone antagonist in Chinese women undergoing controlled ovarian hyperstimulation for in vitro fertilization and embryo transfer].

OBJECTIVE: To analyze the clinical characteristics and cycle outcome of Chinese women with gonadotropin-releasing hormone (GnRH) antagonist treatment during controlled ovarian hyperstimulation (COH) for in vitro fertilization and embryo transfer (IVF-ET). METHODS: A retrospective review was conducted in patients who completed 54 consecutive cycles of IVF-ET with GnRH antagonist treatment for luteinizing hormone (LH) surge prevention. Descriptive statistics were recorded for the patients' age, GnRH treatment duration (days) and dose, timing and duration of GnRH antagonist treatment, serum E2 and LH level on the day of antagonist use and hCG injection, number of oocytes retrieved, and clinical pregnancy rate. RESULTS: The clinical pregnancy rate was 46.2% per ET cycle for GnRH antagonist group and 56.8% in GnRH agonist group, showing no significant difference between the two protocols. The age of the patients with GnRH antagonist averaged 35.7-/+3.8 years. Gn and GnRH antagonist treatment lasted for 8.5-/+1.6 and 4.5-/+1.1 days, respectively. On the day of ovulation triggered by hCG, the serum estradiol level was 1616.7-/+721.1 pg/ml, and a mean of 7.4-/+4.6 oocytes was collected per retrieval. The number of the embryos transferred was 2.4-/+0.6, with an implantation rate of 27.7%, resulting in a clinical pregnancy rate of 50.0% in the fixed protocol (antagonist initiation on day 4 or 5 of stimulation) and 37.5% in the flexible protocol (antagonist treatment initiated for a follicle of 12-15 mm, on day 6 to 9 of stimulation). CONCLUSIONS: GnRH antagonists treatment results in good outcomes and can be safe, short, convenient and effective for Chinese women undergoing COH for IVF. GnRH antagonist treatment can be initiated on day 4 to 9 of Gn stimulation to obtain comparable pregnancy rate.

[Comparison of clinical outcomes of four protocols for frozen-thawed embryo transfer cycle].

OBJECTIVE: To compare the clinical outcome of 4 protocols of frozen-thawed embryo transfer cycle to select the optimal endometrial preparation method for frozen-thawed embryos transfer. METHODS: A retrospective analysis of the 4 clinical protocols was conducted including natural cycle, down-regulated hormone replacement treatment (HRT) cycle, hMG cycle and natural cycle+hCG in endometrial preparation for 419 frozen-thawed embryos transfer cycle, and the clinical pregnancy rate, implantation rate, early abortion rate, ectopic pregnancy rate , ongoing pregnancy rate and delivery rate were compared between the 4 protocols. RESULTS: There was no significant difference between the 4 groups with different clinical protocols in age, duration of infertility, reason of infertility, number of embryo transferred and endometrial thickness. The 4 protocols differed little in the implantation rate, clinical pregnancy rate, biochemical pregnancy rate, early abortion rate, ectopic pregnancy rate, ongoing pregnancy rate and delivery rate in the four clinical protocols. CONCLUSION: The 4 clinical protocols for frozen-thawed embryos transfer all have favorable clinical outcome, and choice of a specific protocol should be made according to the a comprehensive consideration of the individual conditions of the patient.

[Combined use of methotrexate and mifepristone for ectopic pregnancy management: a meta- analysis].

OBJECTIVE: To evaluate the clinical effect and safety of combined use of methotrexate and mifepristone for treatment of ectopic pregnancy. METHODS: By searching in the major databases of CNKI, CBMdisk and Pubmed according to the criteria of evidence-based medicine, we collected data of randomized controlled trials pertaining to combined use of methotrexate and mifepristone in the treatment of ectopic pregnancy. RESULTS: Twenty-three randomized controlled trials involving totally 1 706 patients were collected according to the inclusion criteria, and meta-analysis of the data indicated that combined use of methotrexate and mifepristone can be of great value in the management of ectopic pregnancy in comparison with exclusive use of methotrexate [ combined odds ratio (OR) was 2.84 with 95%confidence interval [CI] (2.18, 3.69), Z=7.79, P<0.000 01]. CONCLUSION: The clinical evidence derived from the analysis suggests that the combination of methotrexate and mifepristone for ectopic pregnancy management can be effective with good safety security and minimal side effects, but still, this conclusion needs further verification by randomized, double-blind, and controlled trials with larger sample size and more rigorous trial design.