Objective:To screen risk factors for ICU-acquired weakness in patients with mechanical ventilation and construct a predictive model, so as to provide a basis for the health management of patients with mechanical ventilation.Methods:Convenience sampling was used to select 312 ICU patients with mechanical ventilation admitted to the Tenth People's Hospital of Tongji University from October 2019 to August 2020 for the study. Patients were divided into training set ( n=220) and test set ( n=92) in a 7∶3 ratio. Based on machine learning algorithms, decision random forest (DRF), extremely-randomized trees (XRT) and generalized linear model (GLM) were used to construct three ICU-acquired weakness risk prediction models for patients with mechanical ventilation, respectively. The performance of the prediction model was evaluated using the area under the receiver operating characteristic curve ( AUC), the area under the precision-recall curve ( AUPRC), and the root mean square error ( RMSE) . Results:There were 7 predictors of risk of ICU-acquired weakness in patients with mechanical ventilation, including age, gender, braking, duration of mechanical ventilation, blood glucose, lactic acid, and parenteral nutrition. Test set and training set validation showed that AUC and AUPRC of GLM prediction model were greater than those of DRF, XRT prediction model. Test set validation indicated that the RMSE, logarithmic loss of GLM prediction model was less than those of DRF, XRT prediction model. Conclusions:Machine learning algorithm based GLM prediction model has good prediction performance. Healthcare professionals can construct evidence-based decisions for interventions in areas such as braking, duration of mechanical ventilation, and blood glucose management.

Objective:To screen risk factors for ICU-acquired weakness in patients with mechanical ventilation and construct a predictive model, so as to provide a basis for the health management of patients with mechanical ventilation.Methods:Convenience sampling was used to select 312 ICU patients with mechanical ventilation admitted to the Tenth People's Hospital of Tongji University from October 2019 to August 2020 for the study. Patients were divided into training set ( n=220) and test set ( n=92) in a 7∶3 ratio. Based on machine learning algorithms, decision random forest (DRF), extremely-randomized trees (XRT) and generalized linear model (GLM) were used to construct three ICU-acquired weakness risk prediction models for patients with mechanical ventilation, respectively. The performance of the prediction model was evaluated using the area under the receiver operating characteristic curve ( AUC), the area under the precision-recall curve ( AUPRC), and the root mean square error ( RMSE) . Results:There were 7 predictors of risk of ICU-acquired weakness in patients with mechanical ventilation, including age, gender, braking, duration of mechanical ventilation, blood glucose, lactic acid, and parenteral nutrition. Test set and training set validation showed that AUC and AUPRC of GLM prediction model were greater than those of DRF, XRT prediction model. Test set validation indicated that the RMSE, logarithmic loss of GLM prediction model was less than those of DRF, XRT prediction model. Conclusions:Machine learning algorithm based GLM prediction model has good prediction performance. Healthcare professionals can construct evidence-based decisions for interventions in areas such as braking, duration of mechanical ventilation, and blood glucose management.

【Objective】 To analyze the characteristics and mobility of the overlapping population of voluntary blood donation and plasmapheresis donation, so as to provide a scientific basis for the formulation of recruitment and retention strategies for blood donation and plasmapheresis donation, and to further propose a scientific reference for the decision-making of blood banks and plasmapheresis station management in China. 【Methods】 The basic information of blood donors and plasmapheresis donors in two counties in Guangyuan, Sichuan Province, which carried out whole blood collection and plasmapheresis collection from the establishment of the station to July 31, 2021 was statistically compared and analzed using the chi-square test and Post hoc testing test. 【Results】 As of July 31, 2021, a total of 50 658 people participated in blood donation and 63 375 people participated in plasmapheresis donation in Jiange County and Cangxi County, with a total overlap of 6 189 people. In the two regions, 16 458 (35.2%) people aged 40 to 50, and 35 558 people (56.1%) were over 50 years old. Among the overlapping population, 2 496 (40.3%) were 40 to 50 years old, accounted for the largest proportion, and 3 146 (50.8%) were males. Significant differences were noticed in age (P<0.001) and gender (P<0.001). There was a shift in dontion in 5 183, including 2 072 people from plasma to blood and 3 111 people from blood to plasma, among which 2 671 (51.5%) were men and 3 632 (70.1%) were over 50 years old, with significant differences in gender (P<0.05) and age (P<0.001). 【Conclusion】 There were a small number of donors donating both blood and plasma in Jiange and Cangxi, and men aged 40 to 50 were the majority, and people over 50 years old were more likely to shift the donation goals. The vast majority of donors have a single and fixed donation goal (blood or plasma), and are not easy to change.

【Objective】 To analyze the dynamic relationship between the setting up of plasmapheresis station and the volume of voluntary blood donation collected using panel vector autoregressive model, so as to provide scientific reference for the management policies of blood stations and plasmapheresis stations in China. 【Methods】 The data collected from blood stations in seven administrative regions of Guangyuan, Sichuan Province from 2011 to 2021, as well as plasma collection data from two plasmapheresis stations in the region within two years since their operation, were collected. A panel vector autoregressive model was constructed. Impulse response analysis and variance decomposition analysis were used to analyze the impact and time lag effects of simulated plasmapheresis station settings on the collection volume of voluntary blood donation. Covariance analysis was used to explore whether the establishment of plasmapheresis station had an impact on the volume of voluntary blood donation collected after excluding the impact of initial value differences. 【Results】 The pulse response results showed that after the plasmapheresis station was set up, there was a negative impact effect on the voluntary blood donation collection volume at the first stage, and its impact began to rise after the second stage, reached the highest value in the third stage, and then began to decline. After the seventh stage, it tended to be stable. However, within the 10 stage range, the confidence interval for the response strength of voluntary blood donation collection volume always included 0, indicating that the response of blood collection volume to the plasmapheresis station setting in the region was not statistically significant. The results of variance decomposition showed that the contribution of collection volume of voluntary blood donation to their own impact reached 94.3%. In terms of the contribution of plasmapheresis station factors, the number of plasma donors has a relatively greater impact on the volume of voluntary blood donation collected(2.2%). Covariance analysis showed that after removing the initial confounding factors, whether to establish a plasmapheresis station had no significant impact on blood donation volume in the two groups of regions (P>0.05). 【Conclusion】 The establishment of a new plasmapheresis station will have a certain impact on blood collection volume of blood stations in the region in a short term, but in the long term, it may not directly affect the voluntary blood donation collection in the region.

【Objective】 To explore the difference of demographics and influencing factors of motivations between whole blood donors and plasmapheresis donors, so as to provide scientific reference for effective recruitment strategy. 【Methods】 A total of 200 whole blood donors from Guangyuan Blood Center and 200 plasmapheresis donors from Jiange Plasmapheresis Station were selected in August 2021 for on-site questionnaire survey using the method of cross-sectional survey. Statistical analysis was performed by chi-square test, univariate and multivariate logistic regression. 【Results】 There were significant differences in gender, age, occupation, education level and annual family income between whole blood donors and plasmapheresis donors (P<0.05). Males accounted for a large proportion of whole blood donors(124/196, 63.3%), whereas females accounted for a large proportion of plasmapheresis donors(117/198, 59.1%). There was little difference in the number of whole blood donors in different age groups, while the age of plasmapheresis donors was concentrated in 40~59 years old (167/198, 84.3%). In terms of occupation, civil servants (including public institutions) accounted the highest proportion in whole blood donors (41/196, 20.9%), and farmers accounted the highest proportion (152/198, 76.8%) in plasmapheresis donors. The number of whole blood donors increased with the education level, and donors with college/university and above degree accounted the largest proportion (80/196, 40.8%). Plasmapheresis donors with junior middle school education and college/university and above accounted the largest and smallest proportion (49.5% vs 4.5%). The annual family income of whole blood donors ranged from 30 000 to 80 000 yuan accounted the largest proportion (109/196, 55.6%), and the annual family income of plasmapheresis donors less than 30 000 yuan accounted the largest proportion (132/198, 66.7%). 【Conclusion】 There were significant statistical differences in gender, age, education level, occupation and annual family income between whole blood and plasmapheresis donors. Therefore, targeted recruitment strategies should be formulated.

The application of artificial neural network algorithm in pathological diagnosis of gastrointestinal malignant tumors has become a research hotspot. In the previous studies, the algorithm research mainly focused on the model development based on convolutional neural networks, while only a few studies used the combination of convolutional neural networks and recurrent neural networks. The research contents included classical histopathological diagnosis and molecular typing of malignant tumors, and the prediction of patient prognosis by utilizing artificial neural networks. This article reviews the research progress on artificial neural network algorithm in the pathological diagnosis and prognosis prediction of digestive tract malignant tumors.
Humans ; Neural Networks, Computer ; Algorithms ; Prognosis ; Gastrointestinal Neoplasms/diagnosis*

The competency-based medical education has formed a global trend, and puts forward a greater challenge for educational design of resident training. The traditional curriculum cannot meet the goal of competency-based education as the curriculum design is lack of theoretical support. Curriculum design is the core of training content, and serves as a significant contributing factor of training outcome. Based on the six-step approach curriculum design, the theory and practice are integrated to form a curriculum design based on theoretical guidance. Through feedback evaluation, the current curriculum design is continuously improved in order to achieve a higher competency-based training quality. With the 5-year experiences and practice, preliminary reform demonstrates effectiveness. The current study hopes to share the teaching reform experiences of residency training base and provide references for colleagues of medical education.

Objective : To study the guidance effects of fibroblast growth factor 3 ( FGF3) on the prethalamic γ⁃aminobutyric acid (GABA) inhibitory axons, and to explore the effect of FGF3 on the formation of thalamic axonal network. Methods : Expression and localization of FGF3 and prethalamic GABA marker gene in chicken embryonic diencephalon were detected by immunofluorescence. Early orientation of GABA axons was tracked by DiI. Three⁃dimensional gel co⁃culture experiment was carried out to investigate guidance effects of FGF3 on early prethalamic GABA inhibitory axons. Results : Prethalamic GABA cells were adjacent to the FGF3 + hypothalamas, located above the hypothalamus. DiI tracing experiments revealed that prethalamic GABA inhibitory axons had already extended into the thalamus at E6. Compared with the blank control group, FGF3 not only significantly promoted the growth of prethalamic GABA inhibitory nerve fibers in the prethalamus, but also repelled the newborn prethalamic axons to the dorsal thalamus. The number of prethalamic axons was significantly less in proximal section (towards FGF3 beads) than that in distal section (away from FGF3 beads) (P < 0. 01) . Moreover, the guidance effects of FGF3 on prethalamic axons could be blocked by the FGF pathway inhibitor SU5402. Conclusion FGF3, an axon guidance molecule expressed in hypothalamus, exerts guidance effects on the pathway selection of adjacent prethalamic GABA inhibitory nerve fibers.

OBJECTIVE: To compare the curative effect on diarrhea-predominant irritable bowel syndrome (IBS-D) between acupuncture for regulating METHODS: A total of 231 patients with IBS-D were randomized into an acupuncture group (154 cases) and a western medication group (77 cases) at the ratio of 2 to 1. In the acupuncture group, acupuncture was applied to acupoint regimen for regulating RESULTS: After treatment and in follow-up, the total scores of IBS-SSS in the patients of the two groups were all reduced as compared with those before treatment ( CONCLUSION Acupuncture for regulating
Acupuncture Therapy ; Diarrhea/therapy* ; Humans ; Irritable Bowel Syndrome/therapy* ; Quality of Life ; Serotonin Plasma Membrane Transport Proteins/genetics* ; Spleen ; Treatment Outcome

Objective:To study the effect mechanism of Guilu Erxian gum on Alzheimer's disease （AD） from the perspective of regulating perivascular space （PVS），and to explore the scientific connotation of "essence generating marrow". Method:The 80 patients with AD diagnosed by western medicine and kidney deficiency and marrow empty syndrome diagnosed by traditional Chinese medicine were randomly divided into two groups，with 40 cases in each group. Both groups of patients were orally administered with cholinesterase inhibitor Alison，one tablet （5 mg） each time before sleep at night. On this basis，the control group additionally received placebo，while the treatment group was additionally treated with Guilu Erxian gum for 60 days. The Mini-Mental State Examination scale （MMSE）， Wechsler Memory Scale and Activity of Daily Living Scale （ADL） were used before treatment （0 d），as well as 31 d and 61 d after treatment. The number and diameter of PVS in midbrain，basal ganglia，deep insular white matter and semiovale center were counted and their diameters were measured with use of nuclear magnetic resonance imaging （MRI） for head. In addition，the curative effect was evaluated according to MMSE scores on 61 d. Result:There was no significant difference between the two groups 31 d. On 61 d，MMSE and WMS scores increased，while ADL scores decreased as compared with the conditions on 0 d（P<0.01），and there were significant differences in the three indexes and clinical effective rate between two groups （P<0.05，P<0.01） . In addition，there was no significant difference in the number of PVS as compared with the number before treatment and in the comparison between the two groups after treatment，but there was a significant difference in the diameter of PVS（P<0.01）. Conclusion:Guilu Erxian gum is effective in the treatment of AD，and it can improve the PVS diameter in patients，which may be related to the mechanism of "essence generating marrow ".