Search

Objective To explore the clinical application of quantitative electroencephalography (qEEG) in predicting the early prognosis of acute ischemic stroke after mechanical thrombectomy. Methods A total of 71 patients who underwent acute ischemic stroke mechanical thrombectomy from April 2021 to September 2022 were enrolled in He'nan Provincial People's Hospital, all of them performed qEEG after surgery. Modified Rankin Scale (mRS) was used to evaluate the prognosis on the 90 d after the surgery. Univariate and multivariate backward Logistic regression analyses were used to screen risk factors for prognosis in acute ischemic stroke after mechanical thrombectomy. Receiver operating characteristic (ROC) curve was used to predict the value of qEEG for the prognosis of neural function. Results Patients in dismal prognosis group (mRS score 3-5, n = 28) had higher NIHSS score at admission (t = 2.686, P = 0.009), higher proportion of thrombectomy performed more than 3 times (χ2 = 4.201, P = 0.040), higher values of DTABR on the affected side (t = 2.183, P = 0.032), higher ratio of DTABR on affected side/ unaffected side (t = 6.230, P = 0.000), and higher proportion of immediately Thrombolysis in Cerebral Infarction (TICI) < 2b after surgery (χ2 = 5.420, P = 0.020) compared to favorable prognosis group (mRS score 0-2, n = 43). Logistic regression analysis showed higher NIHSS score at admission (OR = 1.542, 95%CI: 1.368-1.725；P = 0.012), DTABR on affected side/unaffected side (OR = 3.428, 95%CI: 2.673-7.314; P = 0.008), percent alpha variability (PAV) grade Ⅱ (OR = 2.983, 95%CI: 2.625-4.682; P = 0.003) and grade Ⅲ (OR = 4.088, 95%CI: 3.825-5.349; P = 0.002), amplitude electroencephalography (aEEG) grade Ⅱ (OR = 2.536, 95%CI: 1.942-5.287; P = 0.005) and grade Ⅲ (OR = 3.924, 95%CI: 2.012-6.378; P = 0.003) were risk factors of dismal prognosis in acute ischemic stroke after mechanical thrombectomy. ROC curve showed the area under the curve (AUC) of NIHSS score was 0.676 (95%CI: 0.554-0.782, P = 0.000), DTABR on affected side/unaffected side was 0.887 (95%CI: 0.789-0.950, P = 0.000), the PAV was 0.760 (95%CI: 0.643-0.853, P = 0.000), aEEG was 0.778 (95%CI: 0.664-0.868, P = 0.000), and the combined indicators classification was 0.943 (95%CI: 0.861-0.984, P = 0.000). Among them, the predictive power of the combined indicators is higher than that of the NIHSS score at admission (Z = 4.150, P = 0.000), PAV (Z = 4.006, P = 0.000) and aEEG (Z = 3.462, P = 0.001). Conclusions qEEG is an effective method to predict the early prognosis of acute ischemic stroke after mechanical thrombectomy.

In order to improve the cooperative ability of multiple unmanned combat aircraft vehicle(UCAV) in air combat, a cooperative air combat maneuver decision method based on multi-agent dual delay depth deterministic policy gradient algorithm (MATD3) is proposed. Firstly, the air combat environment is constructed based on the three degree of freedom dynamic model of UCAV, and the action space represented by the change of control quantity is designed based on the pilot’s control mode. Secondly, the design of state space and reward function is optimized to improve the accuracy of maneuvering decision and the safety of maneuvering flight. The relative relationship between friendly aircraft and enemy aircraft is introduced into state space, the continuous reward function is established according to the relative angle, relative distance and other air combat situation factors, and the flight constraints are integrated into the discrete type reward function. Training techniques such as phased training, heuristic guidance, dual exploration mechanism, and alternating freezing game are adopted to improve the convergence speed of the algorithm and the robustness of the maneuvering strategy.Finally, a two-to-one air combat simulation scenario is constructed, and the results show that our two aircraft can show obvious cooperative behavior, which improves the perception and control of air combat situation.

Objective: This study aimed to investigate the effects of 6 kinds of probiotics and their different adding ways on the production of rice wine. Methods: The probiotic rice wine was prepared by adding Lactobacillus rhamnosus BV-77, Lactobacillus fermentans TF-331, Lactobacillus saliva AP-32, Lactobacillus johnsoni MH-68, Lactobacillus acidophilus TYCA-06 and Lactobacillus casei CS-773. To compare the effects of saccharifying enzyme alone, saccharifying enzyme and yeast at the same time and adding methods （probiotic bran koji and probiotic powder） on alcohol content, reducing sugar, total acid and sensory quality of rice wine. Results: Yeast can promote the fermentation of probiotic rice wine and significantly improve the alcohol content and reducing sugar content of rice wine. The rice wine was fermented at 30 ℃ for 3 days by adding 200 U/g saccharifying enzyme and 0.04% yeast, with the highest alcohol content and reducing sugar content of 7.20%vol and 270.00 mg/mL, respectively, and the lowest total acid content of 8.40 mL/L. The rice wine is relatively rich in taste. Conclusion: Lactobacillus fermentans is the most suitable probiotics for rice wine brewing among 6 kinds of probiotics.

Reproducible build is the ability to recreate binary artifacts in a predefined build environment.Due to the role of reproducible build in ensuring the security of software construction environment and improving the efficiency of software construction and distribution,many open source software repositories(such as Debian) have carried out software reproducible build practice.However,due to the lack of sufficient judgment information and the complexity and diversity of source files,it is still a time-consuming and laborious challenge to determine why software can not be built reproducibly.In order to overcome this challenge,this paper studies the classification and detection of software unreproducible build causes based on machine learning.This paper stu-dies four typical reasons for unreproducible build,namely timestamp,fileordering,randomness and locale.This method uses the word vector generated by word2vec to represent the text log,and then cooperates with the logistic regression model to learn and train the text corpus combined with the difference log and the build log,so as to realize the automatic classification of the causes of unreproducible build.In this paper,the algorithm is implemented and tested on 671 unreproducible build Debian software packa-ges.Experimental results show that our method achieves a macro average precision of 80.75% and a macro average recall of 86.07%,which are better than other commonly used machine learning algorithms.In addition,we also analyze the relevance and importance of difference log and build log.Result indicates that both of them are significant for the classification of unreproducible build causes.This method provides a reliable research basis for automatic classification of unreproducible build causes.

Application programming interface(API) benefits to effectively improve software development efficiency by reusing existing software frameworks or libraries.However,many constraints must be satisfied to correctly use APIs,such as call order,exception handling.Violation of these constraints will cause API misuse,which may result in software crashes,errors,or vulnerabilities.Although many API misuse detection techniques have been proposed,these techniques still face two challenges:1) the acquisition of API usage specification is difficult,and 2) the detection of many different types of API misuse at the same time is difficult.To address the above challenges,a decision tree algorithm-based API misuse detection method is proposed.First,the API usage source code is converted into an API usage graph,and the API usage specification is mined from the graph to effectively solve the first challenge.Second,an API usage decision tree is constructed based on the obtained API specification information,and the generalization ability of the API usage decision tree is improved by incorporating pruning strategies.Finally,a combination of coarse-grained and fine-grained detection is proposed in the detection phase to improve the detection capability of the API usage decision tree,which effectively solves the second challenge.Experimental results show that the proposed approach can rea-lize detection of API misuse defects to a certain extent.

The cost of software maintenance increases as the continuous iterative development of software.To reduce this cost,automated software refactoring is proven to be an effective solution.One of the most typical automated software refactoring approaches is the search-based refactoring.This approach refactors software according to some predefined objectives.Existing stu-dies show that the selection of objectives plays a decisive role in the search process.Although many quality objectives and non-quality objectives are proposed,there is no research to analyze that in the same evaluation framework which objectives can guide the software refactoring to achieve results that meet developers’ expectations.Meanwhile,whether the combination of quality objectives and non-quality objectives will have better results is not analyzed.To answer these questions,this paper designs a search-based multi-objective software refactoring evaluation framework.Under the framework,this paper investigates the impact of the combination of seven different objectives on software refactoring with six open-source software projects in different scales;the software quality before and after refactoring is evaluated using a variety of indicators,and the effectiveness of different objective combinations is also analyzed.Experimental results show that the combination of quality objectives and non-quality objectives can better improve the effectiveness of software refactoring than using one single type of objectives,and the combination of the ‘qua-lity’ objective and the ‘similarity with recorded code changes’ objective is important to the improvement of software refactoring effectiveness.

Objective: A microbial enhancer with high-yield glucoamylase was developed from medium-high temperature Daqu and applied to the brewing of Daqu Baijiu for the yield increasing. Methods: Strains with high yield of glucoamylase was isolated from medium-high temperature Daqu and used to make Miqu. Miqu was employed as an enhanced starter in the brewing of Daqu Baijiu, and the yield of Baijiu were tested. Results: A strain M-1 with higher glucoamylase production was isolated and identified as Rhizopus oryzae. The results of the optimized conditions for Miqu making was as follows: bran and rice (2∶8), moisture content 45%, incubation time 72 h, drying temperature 40 ℃. Under the control of these conditions, the glucoamylase activity of Miqu is as high as 864.50 U/g. When appling the Miqu to the brewing of Daqu Baijiu, the results showed that when the addition amount of by, and the yield increased to 43.17% with the addition of 2% (weight of grain) M-1 Miqu. Conclusion: This study established a method for preparing Miqu from Rhizopus oryzae to improve the glucoamylase activity of medium-high temperature Daqu.

Objective To analyze the surgical strategy, efficacy and long⁃term safety of chronic neuropathic pain (NP) after simple cauda equina injury. Methods The clinical data of patients with chronic NP after cauda equina injury treated by dorsal root entry zone (DREZ) derogation and spinal cord stimulation (SCS) in Xuanwu Hospital, Capital Medical University from September 2011 to December 2020 were analyzed. The DREZ derogation or SCS was selected according to the lesion location, injury degree and pain distribution. The long⁃term effect was evaluated according to the preoperative and postoperative Visual Analog Scales (VAS) scores, pain relief rate ≥ 50% was considered as effective, and the complications related to the surgery were followed up. Results Among 24 cases, 14 cases were treated with SCS, 9 cases had pain relief rate 60%-80%, and permanent pulse generator (IPG) was implanted; the pain relief rate was more than 60% in 2 cases, but it did not meet expectation so that the IPG was not implanted; no pain relief in 3 cases, DREZ derogation was performed later, the pain was relieved in 2 cases, and one case was still ineffective. DREZ derogation was firstly performed in 10 cases, and the pain disappeared in 7 cases; 2 cases had residual pain in some areas; one case was ineffective, and the pain was still unrelieved after SCS. After long⁃term implantation of IPG, 2 cases had been taken out of the stimulator due to fluid accumulation at the implantation site. After DREZ derogation, 3 cases had pain at the upper boundary of the preoperative pain area, which gradually disappeared within 2 weeks to one month. In one patient, the pain disappeared in early stage, which recurred after 2 years, and disappeared after the second DREZ derogation. There was no significant difference in the effective rate between 2 groups (9/13 vs. 11/15, Fisher's exact probability: P=1.000). However, DREZ derogation had high pain improvement rate than SCS [100% (0, 100%) vs. 25% (0, 75%); Z=0.441, P=0.030]. Conclusions For chronic neuropathic pain after simple cauda equina injury with clear etiology, both SCS and DREZ derogation may receive satisfactory long⁃term effect. SCS is the first choice for patients with lower limb movement, urinary and bowel function in pain area. DREZ derogation is the first choize for patients without lower limb movement, urinary and bowel. The pain improvement rate of DREZ derogation was better than that of SCS.

Pituitary incidentalomas (PIs) are defined as previously unexpected pituitary lesions which are accidentally discovered during imaging examinations performed for unrelated reasons, including pituitary adenoma, craniopharyngioma, Rathke's cyst, pituitary hyperplasia and other sellar lesions. With the wide application of imaging techniques, especially MRI, more and more pituitary incidentalomas have been detected. However, there are still many problems regarding the definition of pituitary incidentaloma, natural course of disease, intervention indication at the first diagnosis, follow⁃up plan, and so on, resulting in misdiagnosis and overtreatment. This review attempts to discuss the definition, relevant factors for predicting the natural course of tumor, diagnosis, surgical indications, and individualized follow ⁃up plan, and so on, in order to attract clinical attention and conduct related research to further clarify the clinical unsolved problems.

Objective To explore the preoperative factors affecting the immediate remission of Cushing's disease after transsphenoidal surgery. Methods A total of 420 patients with Cushing's disease (led by Dr. Feng Ming) in the Department of Neurosurgery of Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College from January 2014 to August 2020 were included. Univariate and multivariate Logistic regression analyses were performed to find risk factors for immediate remission after operation of Cushing's disease, and receiver operating characteristic curve (ROC) was used to predict the cut⁃off value of immediate remission after operation. Results Among 420 patients with Cushing's disease, 318 patients had immediate remission after operation, and the immediate remission rate was 75.71%. Logistic regression analysis showed that non⁃primary surgery (OR = 3.904, 95%CI: 2.199-6.930; P = 0.000) and high adrenocorticotropic hormone level (OR = 0.995, 95%CI: 0.991-0.999; P = 0.027) were risk factors for immediate remission after opration in patients with Cushing's disease. ROC showed the area under the curve (AUC), sensitivity and specificity of 85.45 pg/ml adrenocorticotropic hormone level (best threshold) for predicting immediate remission after operation were 0.586 (95%CI: 0.524-0.648, P = 0.009), 0.510 and 0.640. Conclusions Preoperative adrenocorticotropic hormone and the first operation are important factors affecting the immediate remission of patients with Cushing's disease after transsphenoidal operation. Endocrinologists and neurosurgeons should pay attention to the factors affecting the prognosis of patients during the preoperative evaluation of patients with Cushing's disease and make good communication between doctors and patients.

Nervous system diseases are associated with complexity and diversity. As a result, the prognosis of patients deeply depends on physicians' clinical experience and their diagnosis and treatment level. With the exponential growth of medical big data and the development of statistical analysis tools, more researchers discover the pathogenesis of diseases and assist clinical process through the mining of big data, so as to improve the diagnosis, treatment, and prognosis of nervous system diseases. This paper elaborates the application of three kinds of medical big data in nervous system diseases, including structured data, radiomics, and bioinformatics. doi：10.3969/j.issn.1672⁃6731.2021.03.002

With the development of molecular biology technology, the era of precision medicine is gradually coming. Standardized clinical database can provide a platform for clinicians to integrate clinical and molecular data, so as to make diagnosis and treatment plans more effectively. This paper summarizes the research progress of pituitary adenoma databases at home and abroad in recent years, expounds the management and application of database, and prospects the development of big data of pituitary adenoma. doi：10.3969/j.issn.1672⁃6731.2021.03.003

With rapid aging worldwide, stroke disease has become the second leading cause of death in people over 60 years old, following the cancer disease. The large⁃scale medical database provides epidemiological data by systematically integrating medical data and transforming research hypotheses into evidence⁃based medicine evidence to guide clinical practice, widely used in clinical medical research. This article describes different stroke databases to explain their application and research progress in stroke diseases, and describes the application of artificial intelligence (AI) in database construction, aiming to promote the construction of stroke databases and conduct high⁃quality stroke research. doi：10.3969/j.issn.1672⁃6731.2021.03.004

Objective To retrospectively analyze the clinical characteristics of patients with Cushing's syndrome (CS) complicated with perioperative pulmonary embolism (PE). Methods Depending on Chinese Pituitary Disease Registry Network (CPDRN) and Pituitary Tumor Database in Peking Union Medical College Hospital, 20 patients with CS complicated perioperative PE were collected from January 1, 2013 to September 30, 2020, and the clinical manifestations and laboratory examination data of the patients were investigated. The clinical characteristics were compared between ectopic adrenocorticotropic hormone syndrome (EAS) and CS with PE during the perioperative period. In addition, clinical characteristics were also studied in these patients before and after the operation. Results According to etiology, they were divided into CS (7 cases), EAS (9 cases) and adrenal adenoma (4 cases). The course of CS was longer than EAS (t = ⁃ 2.938, P = 0.011). According to the occurrence time of PE, the patients were divided into preoperative PE (13 cases) and postoperative PE (7 cases). The patients with postoperative PE had a lower body mass index (t = 2.427, P = 0.026) and a higher fibrinogen (t = ⁃ 2.162, P = 0.047). The patients with PE before surgery included 5 cases of CS, 5 cases of EAS, and 3 cases of adrenal adenomas, while those who with PE after surgery included 2 cases of CS, 4 cases of EAS, and one case of adrenal adenoma. Among the 7 patients with PE after surgery, 5 cases did not get remission. Compared with preoperative examination, the erythrocyte count (t = ⁃ 2.032, P = 0.042) and hemoglobin (t = 9.807, P = 0.001) were decreased after operation, the prothrombin time (t = ⁃ 2.847, P = 0.047) and international normalized ratio (t = ⁃ 3.059, P = 0.038) were increased after operation. Conclusions Longer course of disease and no remission hypercortisolemia after surgery may be the risk factors for perioperative PE in CS patients. Patients with elevated fibrinogen before surgery and reduced red blood cell count and hemoglobin after surgery should be alert to the risk of PE. doi：10.3969/j.issn.1672⁃6731.2021.03.009

Objective To summarize the clinical features of adrenocorticotropic hormone (ACTH)⁃dependent Cushing's syndrome (CS) with multiple endocrine neoplasia (MEN). Methods Based on the pituitary tumor database of Peking Union Medical College Hospital of China Pituitary Disease Register Network (CPDRN), 18 patients with ACTH ⁃ dependent CS complicated with MEN from January 1984 to September 2020 were collected. The clinical manifestations, involved other endocrine glands, laboratorial and imageological examinations were analyzed. Results Among 18 patients, 13 cases were Cushing's disease and 5 cases were ectopic ACTH syndrome (EAS); 15 cases were MEN1 type, 2 cases were MEN2 type and one case was MEN4 type. The clinical manifestations were subclinical CS in 6 cases and overt CS in 12 cases. All cases had osteopenia or osteoporosis. Urolithiasis accounted for 11/18 cases. Regarding the involved endocrine glands, pituitary ACTH adenoma accounted for 13/18 cases, prolactinoma accounted for 4/18 cases, hyperparathyroidism accounted for 15/18 cases, pancreatic neuroendocrine tumors accounted for 13/18 cases and adrenal adenomas and hyperplasia accounted for 10/18 cases. Laboratorial examinations showed the average morning serum cortisol was (652.90 ± 254.90) nmol/L, the morning ACTH was (13.38 ± 8.73) pmol/L and the 24 h urinary free cortisol was 446.35 (233.08, 531.10) µg. Serum prolactin was elevated in 2 cases, parathyroid hormone was elevated in 13 cases and serum calcium was elevated in 11 cases. In imageological examinations, there were pituitary microadenoma in 12 cases, pituitary macroadenoma in 3 cases, adrenal adenoma in 8 cases, thickened adrenal in 4 cases, adrenal pheochromocytoma in 2 cases, and paraganglioma in one case. Genetic test showed 3 cases had MEN1 gene mutation including c.1183C ＞ T, c.1579C ＞ T and c.1174G ＞ T. Conclusions ACTH⁃dependent CS with MEN is a rare disease with complex manifestations. The most common symptom is osteopenia/osteoporosis. The most common cause of CS is pituitary microadenoma. MEN1 is the most common MEN type, and primary hyperparathyroidism is the most common type of affected endocrine glands, followed by pancreatic neuroendocrine tumors. doi：10.3969/j.issn.1672⁃6731.2021.03.008

Objective To explore the imaging data of spontaneous intraparenchymal hemorrhage (sIPH) extracted from the Chinese Intracranial Hemorrhage Imaging Database (CICHID), and analyze the distribution of characteristics and features of hematoma expansion (HE) in patients with more than one non⁃contrast CT within 24 h. Methods There were 831 patients who received recurrent head CT scans involved the inclusion criteria analyzed from January 2016 to June 2020. The cutoff of HE was defined a 6 ml, and the patients were divided into HE group (≥ 6 ml) and non ⁃HE group (＜ 6 ml) to explore the characteristics of patients and their imaging. To explore more features, the HE group was subdivided into groups. According to the cutpoints of the expansion volume in hematoma volume, the groups were subdivided into group A (reduce/no change), group B (＜ 3 ml), group C (3-5.99 ml), group D (6-12.49 ml) and group E (≥ 12.50 ml). According to the expansion rate in hematoma volume, there were group F (＜ 26%), group G (26%-32%) and group H (≥ 33%). Results Among 831 patients, there were 155 patients (18.65%) with hematoma expansion. Compared to non ⁃ HE group, the patients in HE group had larger initial hematoma volume (t = ⁃ 6.148, P = 0.000), higher rate of hematoma breaking into ventricle (χ2 = 20.812, P = 0.000) and shorter interval time between CTs (t = 2.306, P = 0.021). When subdivided as 3, 6 and 12.50 ml, there were significant differences among the groups, which including initial hematoma volume (H = 97.129, P = 0.000), rate of hematoma breaking into ventricle (χ2 = 50.043, P = 0.000) and interval time between CTs (t = 2.306, P = 0.021), the initial hematoma volume (P = 0.014, 0.005, 0.000) and the rate of hematoma breaking into ventricle (P = 0.000, 0.000, 0.000) in 3-5.99 ml group, 6-12.49 ml group and ≥ 12.50 ml group were higher than ＜ 3 ml group, the interval time between CTs in ≥ 12.50 ml group was shorter than ＜ 3 ml group (P = 0.000). And subdivided as 26% and 33%, only the rate of hematoma breaking into ventricle showed significant difference among groups (χ2 = 15.649, P = 0.001). Conclusions In the studies, there was 18.65% sIPH patients occurred HE within 24 h and the risk factors might be initial hematoma volume, rate of hemorrhage breaking into ventricle and interval time between CTs. doi：10.3969/j.issn.1672⁃6731.2021.03.010

Objective To validate the agreement among the convolution neural network segmentation algorithm, Tada formula and manual segmentation for subdural/epidural hemorrhage volume. Methods A total of 129 cases with 352 subdural/epidural hemorrhage CT scans were extracted from Chinese Intracranial Hemorrhage Image Database (CICHID) from January 2017 to June 2019. All CT scans were measured by three methods including manual segmentation, algorithm segmentation and Tada formula. The manual segmentation was regarded as the "golden standard" and the agreement test among three methods was performed. We explored the influence factors in different measurement methods, such as the shape or boundary of hematoma. Results Compared with the Tada formula method, the percentage error of segmentation algorithm was small (23.62%), and the agreement between algorithm and the manual reference was strong, which 94.89% (334/352) of the data was within the 95% limits of agreement (95%LoA), however, the 95%LoA was broad. And the performance of segmentation algorithm showed better in asymmetry (P = 0.000) and clear boundary hematoma (P = 0.000). Conclusions The segmentation algorithm based on convolution neural network has a certain application prospect, but need to be validated in large sample research. doi：10.3969/j.issn.1672⁃6731.2021.03.011

doi：10.3969/j.issn.1672⁃6731.2021.03.001

Raman spectroscopy is well-known for its capability of detecting the physical properties, level of defects and number of layers for graphene-based materials, but far more than that, it's proven to be a versatile and promising tool for characterizing graphene-based polymer composites. This work focuses on the applications of Raman spectroscopy in the field of graphene-based polymer composites. The feature bands in Raman spectroscopy enable direct 2D and 3D imaging of graphene-based nanofillers in the polymer matrix, and even out of other carbonaceous materials. In addition, shifts in the vibrational frequencies of the characteristic bands induced by the strain of graphene could be utilized for analyzing the interactions between graphene-based nanofillers and polymer molecules, for calculating the effective moduli as well as for determining the spatial orientation of graphene-based nanofillers in the matrix. In the meantime, the recent progress of applications of Raman spectroscopy in the field of graphene-based polymer composites is introduced, such as the analysis of micromechanics of graphene-based nanocomposites, the investigation of stress transfer efficiency between the nanofillers and the matrix, and the reveal of key factors affecting material behavior. The Raman spectroscopy researches of graphene-based polymer composites currently are mainly focused on model composites, and the fluorescent effect of the matrix polymer as well limits the further applications of Raman spectroscopy. In order to address such problem, the amplified razor power are often adopted, and the resulted nonlinear effects are capable of increasing the intensity of the Raman signal, thus the Raman spectroscopy will be more widely applied in the field of graphene-based polymer composites.

Objective To establish an automatic segmentation algorithm using convolution neural network, and to validate the consistency between the algorithm and manual segmentation. Methods One hundred andforty⁃six CT scans of intracerebral hemorrhage (ICH) were included from Chinese Intracranial Hemorrhage Image Database (CICHID). They were randomly divided into training set (n=90), testing set (n=26) and validation set(n=30). All CT scans were manual segmentation. Training set and testing set were used for algorithm training. The validation set was measured by four methods including manual segmentation, algorithm segmentation, accurate Tada formula and traditional Tada formula. The consistency test was performed. Results Compared with the Tada formula methods, the percentage error ofalgorithm values was the smallest 15.54 (8.41,23.18) %, and algorithm agreement with the manual reference was the strongest (correlation coefficient 0.983). Bland⁃Altman analysis showed that 93.33% of the data was within the 95% limits of agreement (95%LoA), and 95%LoA was narrow (⁃ 6.46-5.97 ml). No significant differences were found in size and shape (P > 0.05, for all). Conclusions The algorithm using convolutional neural network has a certain application prospect, but it needs still more validation in large sample research. DOI：10.3969/j.issn.1672⁃6731.2020.07.005

The basic principles and methods of infrared radiation and infrared stealth technology were discussed in this paper. It was found that flexible regulation of multi-band spectrum was the key to realize infrared stealth compatible with radar stealth, laser stealth and visible light stealth. The research status of traditional infrared stealth materials were summarized and the bottleneck of their development was clarified. Furthermore, the design idea, research basis and development advantages of infrared stealth metamaterials with spectral tailoring function were proposed. We reviewed the research status and development trend of the infrared stealth compatible with radar stealth, infrared stealth with infrared radiation window and infrared stealth compatible with laser and visible light.

Previous selection of diagnosis and treatment methods of pituitary adenoma are mainly based on the clinical experience of neurosurgeons, so the prognosis of patients mainly depends on the diagnosis and treatment level of doctors. In recent years, with the development and progress of artificial intelligence technology such as machine learning and radiomics, clinicians can make diagnosis and treatment plans with the help of machines, so as to achieve more consistent diagnostic accuracy and better efficacy. This paper summarizes the application of artificial intelligence in the diagnosis and treatment of pituitary adenoma, so as to provide references for clinicians to understand various methods of artificial intelligence technology, and to comprehensively evaluate its advantages and disadvantages, so as to make rational use of it in medical research. DOI：10.3969/j.issn.1672⁃6731.2020.03.003

Objective Metastatic carcinoma in sellar region is a rare disease. This article is to discuss the metastatic origin, clinical manifestations, diagnosis and differential diagnosis of metastatic carcinoma in sellar region, and to summarize the experience in diagnosis and treatment of disease. Methods and Results From January 2014 to October 2019, six patients with metastatic carcinoma in sellar region were confirmed by pathological examinations after endoscopic or microscopic transsphenoidal surgery. Four patients presented with mass effect in sellar region as the first manifestation, and 2 patients were associated with a history of malignant neoplasm. The common clinical manifestations were headache and visual disturbance. The lesions mainly involved cavernous sinus and pituitary gland. Three patients whose lesions involved adenohypophysis presented with pituitary hypofunction, of which 2 patients were diagnosed as non⁃functional pituitary adenoma on admission. Conclusions Presentations associated with metastatic carcinoma in sellar region can be the first manifestation of malignant neoplasm. Compared with pituitary adenomas, metastatic carcinoma in sellar region is more prone to present with diabetes insipidus and pituitary hypofunction. Relying solely on imaging diagnosis is relatively limited, and differential diagnosis should be considered in multiple aspects to reduce the misdiagnosis rate. DOI：10.3969/j.issn.1672⁃6731.2020.03.009

Objective To explore the characteristics, risk factors of postoperative infections for patients with growth hormone (GH)⁃secreting pituitary adenoma after endoscopic transsphenoidal surgery, and to share the experience of diagnosis and treatment of these patients. Methods A total of 122 patients with GH⁃secreting pituitary adenoma who underwent endoscopic transsphenoidal surgery were included from January 2016 to October 2019, and the data of postoperative body temperature, postoperative infection and related possible risk factors were analyzed. Results The incidence of postoperative infections in patients with acromegaly was significantly higher than that with other types of pituitary adenomas [9.84% (12/122) vs. 3.77% (8/212), P=0.025]. Central nervous system infection (7 cases) and bacteremia (4 cases) were the most common types of infections. Gram ⁃ negative bacilli were predominant (10 cases). The postoperative peak body temperature of patients with infections (12 cases) was significantly higher (P = 0.000) and appeared later than non ⁃infectious group (110 cases, P= 0.000). Multivariate Logistic analysis showed that intraoperative cerebrospinal fluid leakage (OR = 5.520, 95% CI:1.193-25.551; P = 0.029) and female (OR = 7.804, 95% CI: 1.088 - 55.948;P = 0.041) patients were major risk factors for postoperative infections. Conclusions For patients with GH ⁃ secreting pituitary adenoma who underwent endoscopic surgery, postoperative infections were uncommon but serious complication, especially for females, for those who experienced intraoperative cerebrospinal fluid leakage and postoperative peak body temperature delayed. Early identification and timely treatment are very important. Empirical anti ⁃ infective treatment should cover Gram ⁃ negative bacteria. Repair using autologous fat fascia and nasoseptal flap may reduce the incidence of postoperative infections in patients who suffered severe cerebrospinal fluid leakage. DOI：10.3969/j.issn.1672⁃6731.2020.03.013

Background Headache is a common complaint in growth hormone (GH) ⁃ secreting pituitary adenoma and considered to be relieved after tumor removed. However, some headache could be resistant after surgery, which is rare and intractable. This article summarized the clinical characteristics and diagnosis and treatment of intractable headache combined with GH ⁃ secreting pituitary adenoma. Methods and Results Four intractable headache combined with GH ⁃ secreting pituitary adenoma patients hospitalized in Peking Union Medical College Hospital from Febuary 2014 to July 2017 were reviewed. All the 4 patients received transsphenoidal surgery again and took long ⁃ acting somatostatin analogs (SSTAs) to control headache symptoms after admission. The serum GH, insulin ⁃ like growth factor ⁃ 1 (IGF ⁃ 1) and GH in Oral Glucose Tolerance Test (OGTT) before surgery were (5.80 ± 2.86) , (644 ± 249) and (3.12 ± 1.37) μg/L, (3.50 ± 2.91), (362.25 ± 160.11) and (3.28 ± 2.99) μg/L in 7 days after surgery, and (3.10 ± 2.14), (357 ± 165) and (1.54 ± 1.24) μg/L 3 months after surgery. Numeric Rating Scale(NRS) was (7.50 ± 0.58), (1.75 ± 0.50) and (3.75 ± 1.50) score in different timepoints. Conclusions Long ⁃ acting SSTAs can be effective to intractable headache combined with GH⁃secreting pituitary adenoma, further studies are needed. DOI：10.3969/j.issn.1672⁃6731.2020.03.010

Because of the complex anatomic relationship and difficult treatment of skull base tumors, a multidisciplinary diagnosis and treatment mode (neurosurgery, otorhinolaryngology, head and neck surgery, oral and maxillofacial surgery, ophthalmology, plastic surgery, etc.) are needed to achieve the purpose of effectively removing tumors, improving function of the nerve and the quality of life. Skull base oncology is regarded as a challenging major in neurosurgery. In order to promote the development of skull base tumor surgery in China, this paper reviews the current situation and existing problems, and looks forward to the future development direction. DOI：10.3969/j.issn.1672⁃6731.2020.03.002

According to the high incidence of cerebral hemorrhage, the burden of disease is heavy. Both domestic and overseas built various diagnosis and treatment system. However, according to the lack of precious diagnosis and treatment in the traditional system, the patient's prognosis is quite different. Thank to the advances in technology such as artificial intelligence, it is possible to achieve the precision medicine in the diagnosis, evaluation, prediction and treatment of hemorrhagic stroke. The present paper intends to review the current technological progress, looking forward to achieve precision, individualization in hemorrhagic stroke. DOI：10.3969/j.issn.1672⁃6731.2019.09.003

Both manifestations and treatments of intracranial hemorrhage (ICH) are varied and the effect meets bottlenecks. Recently, the artificial intelligence (AI) technology has developed rapidly. This review aims to help clinicians understand AI technology regarding its application in ICH by systematically reviewing the historical and current examples. Hope to stimulate the AI progress and enhance the level of treatment in ICH in the future. Ultimately, the treatment of ICH would be precision and individualization. DOI:10.3969/j.issn.1672-6731.2019.09.004

DOI：10.3969/j.issn.1672⁃6731.2020.03.001

The study aimed to explore the nutritional value of sea buckthorn by-products and the antibacterial effects of their extracts on Escherichia coli. The experiment employed the sodium nitrite-aluminum nitrate method to determine flavonoid content, while alcohol extraction and water extraction methods were used to extract active substances from sea buckthorn by-products. The antibacterial effects of sea buckthorn by-product extracts on Escherichia coli were investigated by the filter paper method, perforation method, and broth dilution method. The results showed that the dry matter (DM) content of sea buckthorn by-products was 97.98%, ash content was 3.41%, crude protein (CP) content was 12.87%, and crude fat (EE) content was 22.48%. With the increase of sea buckthorn by-product addition, gas production showed a decreasing trend. Among alcohol extraction and water extraction methods, the constant broth dilution method exhibited the best antibacterial effect. In the alcohol extraction method, the filter paper method showed good antibacterial effect with an inhibition zone of 14.6 mm. In the water extraction method, the filter paper method did not exhibit a significant inhibition zone, and the perforation method was not significant in both alcohol extraction and water extraction methods. The study suggests that sea buckthorn by-products have rich nutritional value, and alcohol extraction is more effective in inhibiting Escherichia coli compared to water extraction. [ABSTRACT FROM AUTHOR]

Despite considerable achievement in the past few decades, medical system still faces critical challenges in China, including uneven distribution of medical resources, discrepancy in disease management, limited continuing education and insufficient research platform, which restrict the further improvement of healthcare in China. The 5th-generation mobile network (5G) has the characteristics of high speed and low latency. Taking the advantage of top-level design and leading technology, 5G may facilitate thee ffective settlement of the above mentioned problems and promote the development of healthcare system in China. DOI: 10.3969/j.issn.1672-6731.2019.09.002

In order to solve the problem of redundant time slot allocation for single node sectors, too large overhead and excessively long time spent for beamforming training under mobile scenarios in existing directional MAC protocols, this paper proposes high slot utilization rate and fast directional MAC protocol(HUFD-MAC) for Terahertz wireless personal area networks. The HUFD-MAC protocol uses redundant single node sectors to transmit downlink data and reduces the number of transmitted beam training frames based on time information. This effectively improves the utilization of time slots, reduces network overhead, and decreases the time required for beamforming. The simulation results show that compared with the IEEE 802.15.3c and ACAP-MAC protocols, the network overhead of HUFD-MAC is reduced by about 3.23%, and the MAC layer throughput is increased by about 8.50%.

The terahertz wireless personal area network is a new type of central network without infrastructure, supporting data transmission rates of 10 Gb/s to 1 Tb/s, and has broad application prospects. The existing terahertz wireless personal area network(MAC) protocol and the efficient directional MAC protocol(ED-MAC) have the problem of large overhead and long time, so a fast and efficient terahertz wireless personal area network oriented MAC protocol FE-MAC(Fast and Efficient MAC protocol) is proposed in this paper. Through simulation verification, the proposed FE-MAC protocol is compared with the 802.15.3c-based terahertz MAC protocol and ED-MAC protocol, which can effectively reduce the data transmission delay between nodes and improve network throughput.

The existing opportunistic network routing algorithms based on bargaining game has excessive control overhead caused by the excessive interaction process of nodes, the additional overhead caused by requesting for useless messages，the additional cost and the delay caused by the low probability of the transaction between the two parties. Besides, when the number of remaining messages in the SV list drops to 1，this condition will result in overhead as well. In response to the above problems, this paper proposes an efficient opportunity network routing algorithm——EORB. The algorithm reduces redundant overhead by adopting adaptive streamlined packet digest，adaptively merging SVDP messages and purchase messages, and comprehensively considering the game strategy of buyers′ and sellers′ benefits. It accelerates the message forwarding rate and improves the arrival rate of messages. The simulation results show that the algorithm effectively improves the rate of message delivery,reduces system overhead and reduces average end-to-end delay of messages.

Effect of low-melting-point curing agent (MOEA) on property of epoxy resin was studied. The properties of epoxy resin (MOEA40) with MOEA as the curing agent were characterized and compared with the epoxy resin (DDM30) with 4, 4'-diaminodiphenyl methane (DDM) as the curing agent. The viscosity-temperature property, curing behavior, mechanical properties and thermal properties were characterized by rotational rheometer, differential scanning calorimetry (DSC), universal material testing machine, dynamic thermomechanical analyzer (DMA) and thermal gravimetric analyzer (TGA), respectively. The results show that MOEA40 has better viscosity-temperature property than that of DDM30. MOEA shows low viscosity (0.2-3.5Pa·s) in the temperature range of 60-140℃. MOEA40 and DDM30 are cured with the same curing progressing. Cured MOEA40 shows outstanding mechanical properties with bending strength about 147MPa and tensile strength about 89 MPa, which are 9.7% and 11.2% higher than that of cured DDM30 (bending strength 134MPa, tensile strength 80MPa). In addition, cured MOEA40 shows higher glass transition temperature (168℃) and higher thermal stability with 5% mass loss temperature about 367℃ in nitrogen atmosphere.

BACKGROUND Current X-ray Diffraction analysis of clay minerals is used to detect the clay mineral assemblages and contents of suspended solution with a particle size of less than 2 μm, but the content data is sometimes inconsistent with other test data. The clay minerals with grain sizes larger than 2 μm in the Ordos Basin were observed by Polarizing Microscope and Scanning Electron Microscope (SEM). The kaolinite has grain sizes ranging from 3 μm to 10 μm, which would result in imprecise content data of clay minerals. OBJECTIVE To evaluate the influence of large particle clay minerals on the relative content of clay minerals by designing a comparative test of extracts from different particle size. METHODS 267 samples were selected randomly, and each sample was prepared by extracting two suspension solutions according to the standards of 10 μm and 2 μm, respectively. The clay mineral assemblages and contents of 10 μm and 2 μm extracts were obtained using the analytical method in the SY/T 5163-2010 industry standard. RESULTS Through comparison, it is found that large-grained clay minerals have a significant effect on the relative contents of various types of clay minerals. The clay minerals in 10 μm extracts have a higher content of kaolinite than the 2 μm extracts. Illite/montmorillonite interlayer mineral contents decrease significantly. The change in the illite content is small and irregular. The overall change in chlorite content is small, and the chlorite content in minor samples deviates significantly, which is related to the occurrence of large-particle kaolinite. The standard for extracting grain size should be carefully selected in combination with the geological conditions. CONCLUSIONS It is proved by this experiment that large particle kaolinite has a significant influence on the relative content of clay minerals. Furthermore, the extraction of particle size standards for clay minerals should be carefully chosen by the actual geological conditions and the purpose of the work. It is also suggested that the industry standard SY/T 5163-2010 should be revised.

Graphene is widely used in the preparation of polymer matrix composites because of its excellent properties. 3D printing, as an emerging technology, is increasingly applied to the shaping and processing of graphene/polymer composites. In this paper, three kinds of main preparation methods of graphene/polymer composites which are solution mixing, melt blending, and in-situ polymerization were introduced. 3D printing methods and the advantages and disadvantages of inkjet printing, fused deposition modeling (FDM), stereolithography (SLA) and selective laser sintering (SLS) as well as their characteristics were summarized. Besides, the applications of 3D-printed graphene/polymer composites in electronics, energy, biomedical and aerospace were reviewed. Finally, it was pointed out that the further research in this area would be focused on the fabrication of graphene/polymer composites which have good printability, homogeneous dispersed graphene, and excellent functional characteristics.

To resolve the problem that the standard aggregation frame mechanism and slot application method of the existing IEEE802.15.3c and High Efficient Fairness MAC(HEF-MAC) protocol applied to the terahertz wireless personal area network is not perfect, a high time slot utilization MAC protocol for terahertz wireless personal area networks(HTSU-MAC) is proposed. By using the new mechanism such as on-demand retransmission of frame aggregation and utilization of no-acknowledged time slot, HTSU-MAC makes the channel slot resources used efficiently and the network throughput greatly improved. The theoretical analysis proves the superiority of the protocol, and the simulation results show that the HTSU-MAC protocol has better network performance than other protocols.

Because the biological behavior of pituitary adenoma varies greatly, and the prognosis of patients is closely related to treatment plan and experience and level of the surgeon, how to make appropriate diagnosis and treatment of patients with pituitary adenoma has become a problem that must be solved by surgeons. It has become an important issue for neurosurgeons in many hospitals to learn and master surgical methods and techniques of pituitary adenoma resection to improve the surgical effect. This article will elaborate on several issues that should be paid attention to in pituitary adenoma resection through transsphenoidal approach, hoping to be helpful to neurosurgeons. DOI: 10.3969/j.issn.1672-6731.2019.04.002

DOI: 10.3969/j.issn.1672-6731.2019.03.001

Objective To investigate how economic development and health resource input impact infant mortality in Shanghai. Methods Data on GDP per capita, the number of health facility beds per 10,000 population, and the infant mortality rate in Shanghai from 1991 to 2020 were selected and analyzed using the vector error correction model. Results A long-term equilibrium relationship existed between infant mortality, GDP per capita, and the number of health beds per 10,000 persons, with the elasticity coefficient of infant mortality being-2. 83 (95% CI:-3. 86-1. 80) for GDP per capita and-3. 02 (95% CI: -5. 03-- 1. 01) for the number of health beds per 10,000 persons. The maximum value is reached in the 5th and 6th year respectively and has a long-term negative impact. Conclusion The improvement of infant health cannot be achieved without the two-way promotion of socio-economic and health resource investment. It is necessary to promote the effective combination of investment in health resources and economic development, improve the construction of the maternal and child healthcare service system, strengthen the leading role of the Government, and build an efficient mechanism for the investment of health resources. [ABSTRACT FROM AUTHOR]

DOI: 10.3969/j.issn.1672-6731.2019.09.001

To solve the problems of high transmission delay and low forwarding efficiency in the Routing Algorithm based on Daily Routines（RADR）in opportunistic social networks,an Efficient Community-based Routing Algorithm is proposed（ECRA）.The algorithm only selects the neighbor nodes which are in the same community with message destination node to calculate the node’s important degree.The algorithm uses topological connections to intercept the encounter node.If there exists node which has higher node’s importance than the encounter nodes neighbor,the message is transmitted to the encounter node neighbor nodes by the encounter node.Theoretical analysis and simulation results show that ECRA outperforms existing RADR and the correlation algorithms in terms of delivery ratio,average end-end delivery delay,relay ratio and average storage time.

DOI: 10.3969/j.issn.1672-6731.2019.04.001

Objective To introduce the experience of diagnosing and treating one case of pituitary carcinoma with distinct hormone expressions in primary and metastatic lesions and to improve understanding of this disease. Methods Retrospective study was performed to analyze the clinical manifestations, imaging characteristics, histopathologic findings, and treatment information of the patient. Immunohistochemical staining was done to both primary and metastatic lesions. Results The patient presented with eye pain and discomfort 4 months posterior to pituitary adenoma surgery. Head MRI showed multiple abnormal intracranial signals, irregular pituitary contour, and abnormal enhancements of the sellar region. PET-CT scan showed multiple hypermetabolic lesions. After the first surgery, histological study of the pituitary tumor showed disseminated tumor cells. The cells were round-shaped or polygonal, with mild atypia, moderate amount of eosinophilic plasma and round-shaped nuclei with fine chromatin and unconspicuous nucleoli; mitosis was abundant, while necrosis was absent. The tumor cells expressed P53, chromogranin A (CgA), with scattered expression for growth hormone (GH) and a Ki-67 index of 80% by immunohistochemistry. The first pathologic diagnosis was atypical pituitary adenoma. The parietal tumor cells infiltrated parenchymal after the tumor recurrence. Immunohistochemistry findings were different from the first one. The tumor cells expressed GH diffusely, with a decreased Ki-67 index of 75%. The second pathologic diagnosis was metastatic pituitary carcinoma. Conclusions Pituitary carcinoma is a rare malignant pituitary tumor. Diagnosis relies on radiology and pathology. Surgical resection and radiochemotherapy are the current treatment of choice but yield poor response. General prognosis of the disease is poor. DOI: 10.3969/j.issn.1672-6731.2015.01.013

Resident training of neurosurgical specialization will be tried and carried out in all over China. From the point of view of training base hospitals, a majority of 3A level hospitals have sufficient patients and good equipments which will ensure the success of training process; however, division of subspecialty, teaching motivation and teaching method still have a great potential to improve. In order to establish and improve training bases for residents of specialization, supervision from national administrative department should be strengthened; professional society also plays an important role in standardizing and controlling the training quality. Considering our nation's situation, integration of postgraduate education and resident training is worth of discussion. DOI: 10.3969/j.issn.1672-6731.2015.07.015

To evaluate and select the mode for the EPC project consortium, and to solve the problems of multiple evaluation indicators and insufficient evaluation, on the basis of the analysis on the consortium mode, from the perspective of helping the owner to control the EPC project, six first-level evaluation index systems are built and the evaluation methods based on the TOPSIS method, the gray relational degree, and the combination of the two are employed to evaluate and select the consortium mode for large as well as small and medium EPC projects. The analysis and evaluation results show that, for large EPC projects, the results of the three methods are consistent. The tight consortium led by the design party is the best choice, followed by the partnership led by the design party. For small and medium EPC projects, the results are consistent with the large-scale EPC projects when the TOPSIS method and the combined evaluation method are used. But there are differences in the ranking results when the grey relational degree is used to evaluate. By introducing different evaluation methods and selecting appropriate preference coefficients, it can help the owner to make a more rational judgment on the evaluation and selection of the EPC project consortium model. [ABSTRACT FROM AUTHOR]

Aiming at the problems of normal channel time allocation period (N-CTAP) existing in cross personal area network (PAN) communication in terahertz wireless personal area network, such as the success rate of node communication affected by interference in data transmission of some conventional nodes, an efficient adaptive reporting approach of terahertz network (EARA-T) is proposed. In this method, the conventional node is subdivided into normal node and edge node, and the node type information is timely and efficiently reported to the piconet coordinator (PNC) of the pan before and after the node enters the network, so that PNC can take effective measures to reduce or even eliminate the impact of inter network interference on data transmission in time slot allocation. Compared with FDP protocol and FCP protocol, the reduction of information reporting control overhead and delay of the proposed method is no less than 14. 1% and 3. 7% respectively, and the success rate of information reporting is improved by 0. 4%. [ABSTRACT FROM AUTHOR]

In order to understand the effects of lucerne cropping rotation on the bacterial community of loess soil, a long-term field experiment was conducted in rain-fed agricultural area of Loess Plateau. The cropping systems included continuous lucerne (Medicago sativa, LC), lucerne removed and rotated with spring wheat (Triticum aestivum, LFW), lucerne removed and rotated with corn (Zea mays, LFC), lucerne removed and rotated with potato (Solanum tuberosum, LP), and lucerne removed and rotated with continuous millet (Panicum miliaceum, LM ). Based on 16S rRNA high-throughput sequencing technology, we investigated soil bacterial community structure and diversity in different cropping systems, and predicted ecological function using PICRUSt method. The results showed that the dominant phyla of loess soil bacteria were Actinomycetes (20.3% - 32.0%), Proteobacteria (19.2%-23.0%), Acidobacteria (12.4%-14.2%) and Chloroflexus (11.0%-12.7%). The dominant genus was Bacillus (1.9%) in lucerne-corn system and Pseudarthrobacter (2.5%) in other treatments. Rotation with annual crops decreased the relative abundance of Actinobacteria and increased that of Chloroflexi and Firmicutes. Redundancy analysis showed that the main soil factors driving soil bacterial community structure were nitrate, ammonium, and total nitrogen. PICRUSt function prediction results showed that metabolism (78.6%-79.1%) was the main function of soil bacterial communities in loess soil. Rotation with continued annual crops significantly decreased the abundance of soil bacterial carbohydrate metabolism functional genes, and significantly increased the abundance of functional genes for soil bacterial cofactors and vitamin metabolism, neurodegenerative diseases, and immune system. In conclusion, lucerne removed and rotated with continuous annual crops changed soil bacterial community structure and ecological functions. This study provided theoretical reference to explore succession characteristics of soil bacteria and to select succeeding crops for alfalfa in loess soil. [ABSTRACT FROM AUTHOR]