Purpose Based on a variety of combinations of cranial MR Dixon images,the deep learning method is used to generate CT images,and the reconstruction efficiency is evaluated by comparing with the corresponding CT images.Materials and Methods A total of 77 cranial CT and MR images were collected retrospectively in Ruijin Hospital,Shanghai Jiaotong University School of Medicine from June to December 2021.The U-Net neural network was used for network training,with 62 cases in the training set and 15 cases in the test set.CT image reconstruction was performed using four kinds of Dixon images and a total of seven models among the various combinations.Mean absolute error,mean squared error,Pearson correlation coefficient and skull area Dice similarity coefficient were used to evaluate the image reconstruction efficiency.Results The generated CT images of the various Dixon image combination models showed strong correlation with the corresponding CT images(R>0.75,P<0.05),and the CT images reconstructed by the four-channel model had the closest value to the actual CT images[mean absolute error=147.516±30.802,mean squared error=(8.648±3.403)×104],the highest correlation coefficient(R=0.796±0.055),and the highest similarity coefficient in the cranial region(Dice similarity coefficient=0.800±0.036).Conclusion Deep learning training through Dixon images can be used to generate CT images,and the combination of four kinds of Dixon contrast images can improve the CT image reconstruction efficiency.

Objective To develop a nomogram-based predictive model for assessing the risk of respiratory distress syndrome(RDS)in premature infants in the high-altitude region of Dali.The predictive performance and clinical applicability of the model will be systematically evaluated to provide evidence-based guidance for the early diagnosis and clinical management of respiratory distress in premature infants.Methods A total of 680 preterm infants admitted to the Dali Maternal and Child Health Hospital between January 2020 and December 2024 were enrolled in the study and randomly divided into a training set(n=476)and a validation set(n=204)at a ratio of 7∶3.Independent predictors were identified through univariate logistic regression and multivariate stepwise regression analyses,and a nomogram model was subsequently developed using R software.The performance of the model,including its discrimination,calibration,stability,and clinical applicability,was evaluated using the receiver operating characteristic curve(ROC),Hosmer-Lemeshow goodness-of-fit test,bootstrap resampling method,and decision curve analysis(DCA).Results The final model incorporated seven independent variables:gestational age,birth weight,Apgar score,blood oxygen saturation,gestational hyperglycemia,prenatal glucocor-ticoid therapy,and maternal history of infection.The areas under the curve(AUCs)for the training and validation sets were 0.88(95%CI:0.84～0.92)and 0.83(95%CI:0.76～0.89),respectively,with all Hosmer-Lemeshow test p-values exceeding 0.05.The bootstrap-corrected AUC was 0.85(95%CI:0.81～0.89).DCA indicated that the model achieved the highest net benefit at a risk threshold range of 10%to 35%.Conclusions This model integrates multiple risk factors associated with the occurrence of RDS in plateau environments,demonstrating robust predictive performance for RDS in preterm infants residing in high-altitude areas such as Dali.It can serve as a valuable tool for risk stratification and clinical decision-making,and may also provide a reference for future multicenter prospective studies.

To address the issue of reduced signal quality of photoplethysmography caused by local fluctuation,an approach called locally orthogonal weighted polynomial fitting(LOWPF)is proposed for signal smoothing.After determining the positions of the fluctuation sequences using the forward-backward difference XOR method,weighted polynomial fitting is applied to these sequences,and the fitted values are used to replace the fluctuation sequences to achieve signal smoothing.By constructing orthogonal basis functions,the condition number of the coefficient matrix is reduced,and the stability of the equation system solution for higher-order fitting is improved.Simulation results demonstrate that the smoothed signal's XOR smoothness of the proposed method surpasses that of the moving average algorithm and the empirical mode decomposition reconstruction algorithm.The smoothing results on 241 sets of measured PPG signals show that LOWPF achieves an efficiency of smoothness of 89.10%,significantly higher than the 78.05%of empirical mode decomposition and the 59.13%of the 5-point moving average algorithm.LOWPF has promising application prospects for smoothing signals with significant local fluctuations.

Objective To develop a nomogram-based predictive model for assessing the risk of respiratory distress syndrome(RDS)in premature infants in the high-altitude region of Dali.The predictive performance and clinical applicability of the model will be systematically evaluated to provide evidence-based guidance for the early diagnosis and clinical management of respiratory distress in premature infants.Methods A total of 680 preterm infants admitted to the Dali Maternal and Child Health Hospital between January 2020 and December 2024 were enrolled in the study and randomly divided into a training set(n=476)and a validation set(n=204)at a ratio of 7∶3.Independent predictors were identified through univariate logistic regression and multivariate stepwise regression analyses,and a nomogram model was subsequently developed using R software.The performance of the model,including its discrimination,calibration,stability,and clinical applicability,was evaluated using the receiver operating characteristic curve(ROC),Hosmer-Lemeshow goodness-of-fit test,bootstrap resampling method,and decision curve analysis(DCA).Results The final model incorporated seven independent variables:gestational age,birth weight,Apgar score,blood oxygen saturation,gestational hyperglycemia,prenatal glucocor-ticoid therapy,and maternal history of infection.The areas under the curve(AUCs)for the training and validation sets were 0.88(95%CI:0.84～0.92)and 0.83(95%CI:0.76～0.89),respectively,with all Hosmer-Lemeshow test p-values exceeding 0.05.The bootstrap-corrected AUC was 0.85(95%CI:0.81～0.89).DCA indicated that the model achieved the highest net benefit at a risk threshold range of 10%to 35%.Conclusions This model integrates multiple risk factors associated with the occurrence of RDS in plateau environments,demonstrating robust predictive performance for RDS in preterm infants residing in high-altitude areas such as Dali.It can serve as a valuable tool for risk stratification and clinical decision-making,and may also provide a reference for future multicenter prospective studies.

To address the issue of reduced signal quality of photoplethysmography caused by local fluctuation,an approach called locally orthogonal weighted polynomial fitting(LOWPF)is proposed for signal smoothing.After determining the positions of the fluctuation sequences using the forward-backward difference XOR method,weighted polynomial fitting is applied to these sequences,and the fitted values are used to replace the fluctuation sequences to achieve signal smoothing.By constructing orthogonal basis functions,the condition number of the coefficient matrix is reduced,and the stability of the equation system solution for higher-order fitting is improved.Simulation results demonstrate that the smoothed signal's XOR smoothness of the proposed method surpasses that of the moving average algorithm and the empirical mode decomposition reconstruction algorithm.The smoothing results on 241 sets of measured PPG signals show that LOWPF achieves an efficiency of smoothness of 89.10%,significantly higher than the 78.05%of empirical mode decomposition and the 59.13%of the 5-point moving average algorithm.LOWPF has promising application prospects for smoothing signals with significant local fluctuations.

Purpose Based on a variety of combinations of cranial MR Dixon images,the deep learning method is used to generate CT images,and the reconstruction efficiency is evaluated by comparing with the corresponding CT images.Materials and Methods A total of 77 cranial CT and MR images were collected retrospectively in Ruijin Hospital,Shanghai Jiaotong University School of Medicine from June to December 2021.The U-Net neural network was used for network training,with 62 cases in the training set and 15 cases in the test set.CT image reconstruction was performed using four kinds of Dixon images and a total of seven models among the various combinations.Mean absolute error,mean squared error,Pearson correlation coefficient and skull area Dice similarity coefficient were used to evaluate the image reconstruction efficiency.Results The generated CT images of the various Dixon image combination models showed strong correlation with the corresponding CT images(R>0.75,P<0.05),and the CT images reconstructed by the four-channel model had the closest value to the actual CT images[mean absolute error=147.516±30.802,mean squared error=(8.648±3.403)×104],the highest correlation coefficient(R=0.796±0.055),and the highest similarity coefficient in the cranial region(Dice similarity coefficient=0.800±0.036).Conclusion Deep learning training through Dixon images can be used to generate CT images,and the combination of four kinds of Dixon contrast images can improve the CT image reconstruction efficiency.

In this study, a new model of "Internet + PACD (namely, theoretical presentation, assimilatin, clinicopathological diagnosis and discussion)" was put forward in the online and offline course construction of surgical diagnostic pathology. and the teaching effect of this teaching model was evaluated through the performance evaluation and questionnaire survey. The results showed that the teaching model of "Internet + PACD" could not only significantly improve the performance of professional courses of students majoring in pathology, but also enhance their learning interest, confidence, competition and cooperation consciousness, which has been affirmed and recognized by students.

The construction of Internet plus and big data pathology medical alliance platform is based on the Clinical Pathological Diagnosis Center of Qiqihar Medical University, on the basis of the digital pathological cloud platform provided by Jiangfeng biological assistance, relying on the medical development plan of Qi Medical Pathology diagnosis Center, cultivating the development concept of high, fine and sharp innovative pathological talents, and absorbing the leading figures of pathological diagnosis in various departments in China. From that, the real-time supervision and control of the diagnostic quality of Primary Pathological Diagnostic Unit can be established, the real-time, timely and accurate all-round training of Basic Pathological Diagnostic Technicians can be realized, and the level of basic pathological diagnosis can be quickly improved. A new training mode for pathology talents can perfectly fit with the big data medicine and artificial intelligence.

Objective:To investigate the current status of multiple leadership styles of head nurses and affective domain ability of nurses, and to explore the influence of multiple leadership styles of head nurses on affective domain ability of clinical nurses.Methods:From June and August 2019, we selected 380 clinical nurses in the Second People's Hospital of Liaocheng with the method of cross-sectional survey and convenience sampling. All nurses were investigated with the General Information Questionnaire, Multifactor Leadership Questionnaire (MLQ) and Nurse Affective Domain Ability Scale. We analyzed the influence of multiple leadership styles of head nurses on affective domain ability of clinical nurses. In this study, a total of 380 questionnaires were distributed and 372 valid questionnaires were recovered with a valid recovery rate of 97.9%.Results:Among 372 clinical nurses, the total score of perceived MLQ of head nurses was (86.30±18.59) and the item average score was (2.70±0.58) ; the total score of the Nurse Affective Domain Ability Scale and the item average score were (97.67±13.35) and (3.91±0.53) respectively. Univariate analysis showed that there were statistical differences in the total score of Nurse Affective Domain Ability Scale among nurses with different nursing ages, professional titles and educational levels ( P<0.05) . Pearson correlation analysis indicated that the total score of MLQ of head nurses had a positive correlation with the total score of the Nurse Affective Domain Ability Scale of nurses with a statistical difference ( r=0.364, P<0.05) . Hierarchical regression analysis showed that nursing ages and transformative leadership style had a predictive effect on clinical nurses' affective domain ability. Conclusions:Head nurses adopt a transformative leadership style to enhance clinical nurses' affective domain ability, so that they devote themselves to nursing work with full enthusiasm and stabilize the development of the nursing talent team.

Objective To investigate the protective effect of Honokiol on the airway inflammation induced by particulate matter 2.5(PM2.5)in the asthmatic mice and its mechanism.Methods Fifty male specific pathogen free (SPF)Balb/c mice were randomly divided into 5 groups.Group A:normal control group;group B:asthmatic model group;group C:PM2.5 exposure asthmatic group;group D:TAK -242 group;group E:Honokiol group. Asthmatic mouse models were established by ovalbumin(OVA)sensitization and challenge.On days 0 and 7,the mice in B-E groups were injected intraperitoneally with injection 100 mg/L OVA and aluminum hydroxide for sensitization;on days 14 to 21,10 g/L OVA solution was given 30 min per day to challenge.During challenge phrase,the mice in C -E groups received intratracheal injection of PM2.5,every other day,4 times totally.On this basis,the mice in group D re-ceived TAK-242 intraperitoneal injection,and the mice in group E received honokiol intragastric administration.Group A was given saline instead of OVA.Animals were sacrificed 24 h after the final inhalation challenge,and the bronchoal-veolar lavage fluid(BALF)of the left lung was used for differential inflammatory cell counts.The expressions of Toll-like receptors 4(TLR4)and nuclear factor(NF)-κB at mRNA level were detected by real-time quantitative PCR. Flow cytometry analysis was performed to measure the levels of Th17 and Treg cells.Results Compared with group A,mice in group B and group C expressed more serious disorders of bronchial epithelial cells,alveolar wall congestion and edema,increased mucus secretion in the airway and infiltration of inflammatory cells in the lung,and those in group C were more obvious than those of group B and group E significantly reduced respiratory inflammation;compared with group A[(4.15 ± 1.35)×108/L,0.012 0 ± 0.002 3],the total number of inflammatory cell counts[(16.79 ± 5.62)×108/L and(24.58 ± 13.46)×108/L],eosinophils proportions(0.113 8 ± 0.022 3 and 0.197 8 ± 0.084 9)in group B and group C,were significantly higher,and the differences were statistically significant(all P<0.05);The total number of inflammatory cell counts and eosinophils proportion in group E(8.56 ± 3.28)×108/L and 0.041 5 ± 0.013 5)were significantly lower than those in group C,and the differences were statistically significant(all P <0.05);The expressions of TLR4 mRNA and NF-κB mRNA in group B and C(1.85 ± 0.56,1.82 ± 0.28 and 2.97 ± 0.41,2.83 ± 0.32)were significantly higher,and the differences were statistically significant(all P <0.05);The expressions of TLR4 mRNA and NF-κB mRNA in group E(1.60 ± 0.28,1.54 ± 0.25)was significantly lower than those in group C,and the differences were statistically significant(all P<0.05);the expressions of Th17 in group B and C[(2.89 ± 0.61)% and(4.96 ± 0.27)%]were significantly higher than those of group A[(1.03 ± 0.35)%] (all P<0.05);The expression of Th17 in group E[(1.83 ± 0.23)%]was significantly lower than that of group C,and the differences were statistically significant(P<0.05);the expressions of Treg in group B and C[(4.96 ± 0.35)%and(2.27 ± 0.41)%]were significantly lower than those of group A[(7.37 ± 0.56)%],and the differences were sta-tistically significant(all P<0.05);The expression of Treg in group E was significantly increased[(6.45 ± 0.38)%] compared with that in group C,and the difference were statistically significant(P<0.05);and those of group D and E were improved remarkably.Conclusions Honokiol can relieve PM2.5 exposure of asthmatic airway inflammation through down-regulating the expression of TLR4 and NF-κB and Th17 and regulating the balance of Th17 and Treg cells.