Objective:Middle ear cholesteatoma is a common otolaryngological disease, and traditional diagnostic methods have certain limitations. This study aims to construct a computer-aided diagnosis model for middle ear cholesteatoma based on integrated convolutional neural networks (CNNs) to improve diagnostic accuracy and efficiency.Methods:Firstly, Data were collected from patients who visited the Department of Otorhinolaryngology Head and Neck Surgery at the First People′s Hospital of Yinchuan between January 2020 and December 2021. 8 000 temporal bone CT images were collected, including 5 000 images diagnosed pathologically as middle ear cholesteatoma and 3 000 normal images. A five-fold cross-validation method was used to divide the dataset into training and testing sets. Next, a transfer learning approach was used to initialize model parameters, and the AlexNet, GoogleNet, and ResNet networks were pre-trained to extract deep features from the images. Then, the Softmax classification algorithm was applied to classify the features, resulting in three independent classifiers. These classifiers were combined using an ensemble learning method with a weighted voting approach to obtain the final diagnostic results. Finally, the model was evaluated by comparing the ensemble classifier with individual classifiers to assess its accuracy, precision, sensitivity, specificity, and diagnostic time, and a comparison with low-mid-and high-experience physician groups was conducted to comprehensively evaluate the model′s diagnostic performance.Results:The experimental results showed that the model achieved an accuracy of 88.8%(178/200), precision of 92.9%,(112/120) sensitivity of 89.8%(108/120), and specificity of 88.1%(70/80). The average diagnostic time for individual patient temporal bone CT images was reduced to 2-3 seconds. Compared to the diagnostic results from low-mid-and high-experience physician groups, the model demonstrated significant advantages and effectively assisted clinicians in making rapid and accurate middle ear cholesteatoma diagnoses.Conclusion:The proposed middle ear cholesteatoma diagnostic model based on integrated convolutional neural networks exhibits high recognition accuracy and rapid diagnostic speed, significantly improving clinical diagnostic efficiency, especially in early screening and auxiliary diagnosis, making it of considerable value in clinical practice.

Objective:Middle ear cholesteatoma is a common otolaryngological disease, and traditional diagnostic methods have certain limitations. This study aims to construct a computer-aided diagnosis model for middle ear cholesteatoma based on integrated convolutional neural networks (CNNs) to improve diagnostic accuracy and efficiency.Methods:Firstly, Data were collected from patients who visited the Department of Otorhinolaryngology Head and Neck Surgery at the First People′s Hospital of Yinchuan between January 2020 and December 2021. 8 000 temporal bone CT images were collected, including 5 000 images diagnosed pathologically as middle ear cholesteatoma and 3 000 normal images. A five-fold cross-validation method was used to divide the dataset into training and testing sets. Next, a transfer learning approach was used to initialize model parameters, and the AlexNet, GoogleNet, and ResNet networks were pre-trained to extract deep features from the images. Then, the Softmax classification algorithm was applied to classify the features, resulting in three independent classifiers. These classifiers were combined using an ensemble learning method with a weighted voting approach to obtain the final diagnostic results. Finally, the model was evaluated by comparing the ensemble classifier with individual classifiers to assess its accuracy, precision, sensitivity, specificity, and diagnostic time, and a comparison with low-mid-and high-experience physician groups was conducted to comprehensively evaluate the model′s diagnostic performance.Results:The experimental results showed that the model achieved an accuracy of 88.8%(178/200), precision of 92.9%,(112/120) sensitivity of 89.8%(108/120), and specificity of 88.1%(70/80). The average diagnostic time for individual patient temporal bone CT images was reduced to 2-3 seconds. Compared to the diagnostic results from low-mid-and high-experience physician groups, the model demonstrated significant advantages and effectively assisted clinicians in making rapid and accurate middle ear cholesteatoma diagnoses.Conclusion:The proposed middle ear cholesteatoma diagnostic model based on integrated convolutional neural networks exhibits high recognition accuracy and rapid diagnostic speed, significantly improving clinical diagnostic efficiency, especially in early screening and auxiliary diagnosis, making it of considerable value in clinical practice.

Objective To investigate the relationship between peripheral blood immunoglobulin,erythrocyte sedimentation rate(ESR),homocysteine(Hey)and the severity of central nervous system vasculitis(CNSV)in children,as well as its impact on prognosis.Methods A total of 103 children with CNSV from February 2018 to February 2023 were selected as the study group,and 103 healthy children as the control group.The peripheral blood levels of immunoglobulin A(IgA),immunoglobulin G(IgG),immunoglobulin M(IgM),ESR and Hcy were compared between the 2 groups to evaluate the relationship between each index and the degree of CNSV disease[Birmingham vasculitis disease activity score(BVAS)]and its predictive value for prognosis.Results The levels of peripheral blood IgA,IgG,IgM,ESR and Hcy in the study group were higher than those in the control group of healthy children(P＜0.05);the BVAS scores and the levels of peripheral blood IgA,IgG,IgM,ESR and Hcy in children with active disease were higher than those in children with inactive disease(P＜0.05);the levels of peripheral blood IgA,IgG,IgM,ESR and Hcy were positively correlated with the BVAS scores in children with CNSV(P＜0.05);two cases were lost to follow-up after 6 months.Among the children with CNSV,76 had good prognosis and 25 poor prognosis.The levels of peripheral blood IgA,IgG,IgM,ESR and Hcy in children with poor prognosis were higher than those in children with good prognosis(P＜0.05);before and after correcting for other factors,peripheral blood IgA,IgG,IgM,ESR and Hcy were all independent factors affecting the prognosis of children with CNSV(P＜0.05);the area under curve(AUC)of peripheral blood IgA,IgG,IgM,ESR and Hey for predicting the prognosis of children with CNSV was 0.747,0.808,0.841,0.839,and 0.746,respectively,with optimal cutoff values of 350.58 mg/dL,1513.06 mg/dL,124.84 mg/dL,51.22 mm/h,and 13.66 pmol/L,respectively;the AUC of peripheral blood IgA,IgG,IgM,ESR and Hey for jointly predicting the prognosis of children with CNSV was 0.943(95％CI 0.878-0.979),with a sensitivity of 92.00％and a specificity of 93.42％,which was superior to individual prediction of each indicator.Conclusion Peripheral blood IgA,IgG,IgM,ESR and Hey are positively correlated with the severity of CNSV.Abnormally high expression increases the risk of poor prognosis,and the combined predictive value is reliable.

OBJECTIVE To analyze the clinicopathological features of laryngeal squamous carcinoma tumors and their correlation with prognosis in order to improve the understanding and diagnosis of laryngeal squamous cell carcinoma.METHODS The clinical and pathological data(including gender,age,stage,differentiation,immunohistochemistry,etc.)of 179 patients with laryngeal squamous cell carcinoma[171 males,8 females,aged 30-84(61.53±8.02)years]who were treated in Department of Otolaryngology Head and Neck Surgery,The First and Second Clinical Medical Schools of Ningxia Medical University from January 2015 to December 2022 were retrospectively studied,and the effects of various factors on prognosis were analyzed.RESULTS Among the 179 patients with laryngeal squamous cell carcinoma,the male-to-female ratio was 21.4:1,and the incidence was high in the age group of 60-79 years old(58.7%),and the youngest age of onset was 30 years old.The main clinical manifestations were hoarseness 138 cases(77.1%),sore throat 16 cases(8.9%)and pharyngeal foreign body sensation 13 cases(7.3%).Glottic type was more common in the primary site 135 cases(75.4%),and 31 cases were accompanied by cervical lymph node metastasis(17.3%).The degree of differentiation was more common in the moderately differentiated type 80 cases(44.7%).The positive rates of immunohistochemistry markers p16,EGFR(epidermal growth factor receptor),PD-1/PD-L1 and VEGF(vascular endothelial growth factor)were 20.3%,96.4%,36.4%and 77.3%,respectively.Univariate Kaplan-Meier survival analysis showed that the site of disease,lymph node metastasis,and tumor stage were significantly correlated with disease recurrence.Multivariate Cox regression analysis showed that the clinical stage of the tumor was an independent risk factor for the prognosis of the disease(HR=3.715,95%CI:1.519-9.088,P=0.04).CONCLUSION The stage of laryngeal squamous cell carcinoma,the site of the disease,and the metastasis of the lymph nodes are the main factors affecting the prognosis.The high positive expression rate of immunohistochemistry markers EGFR and VEGF is worth paying attention to Targeted therapy for patients with positive PD-1/PD-L1 testing is a promising research direction.

Objective:To summarize the status and trends of papers published by Chinese scholars on the combination teaching model in the area of medical education in China.Methods:Literature retrieval was performed on CNKI, WanFang, and VIP database, and studies regarding to combination teaching model in medical education were selected and analyzed. BICOMS-2 analysis software was used to extract and sort out keywords, provinces, authors and their organizations and generate co-occurrence matrix and/or word matrix. NetDraw and gCLUTO software were used to draw the network relationship diagram and cluster analysis respectively.Results:Totally 99 articles were included this study, and the number of articles published each year showed an increasing trend. Authors from 15 provinces contributed to these articles, with uneven development and limited cooperation across regions. The authors were divided into 4 main groups, but the communication among these groups was not enough. And there were 3 main research topics.Conclusion:The number of articles on combination teaching model in medical education in China is increasing, covering many authors, teams, institutions and regions. However, the communication and cooperation among these authors and institutions needs to be strengthened; and the research topics should be expanded.

Objective:To evaluate the effects of comprehensive reform of clinical education on cultivating the clinical thinking ability of medical students in internal medicine.Methods:A total of 44 clinical interns on medicine rotation were included in this study and divided into experimental group and control group, and a series of measures for clinical teaching reform, which included undergraduate tutorial system, team-based learning (TBL) mode, combination of electronic medical record and handwritten medical record, as well as formative assessment, were applied to fully promote clinical thinking ability of medical students. SPSS 17.0 software was used to compare the scores of final academic tests of rotation and competition of medical record writing between the control and the experiment groups. Self-designed questionnaires on undergraduate tutors and clinical interns were applied to comprehensively evaluate the effects of this education reform.Results:Independent-sample t tests showed there were statistically significant difference ( P<0.01) in the scores of final academic tests of rotation and competition of medical record writing between the two groups, and the mean score of the experiment group was higher than that of the control group. Questionnaire survey showed that tutors and clinical interns both fully affirmed the positive effects of the education reform on cultivating clinical thinking ability of medical students in internal medicine. Conclusion:Application of comprehensive education reform in clinical teaching of internal medicine, which based on introduction of undergraduate tutorial system and TBL mode, could effectively promote the clinical thinking ability of medical students and the teaching quality.

Objective:To assess the feasibility of free-handed conical craniotomy and drainage guided by novel 16-slice mobile CT-assisted surface projection in patients with supratentorial intracerebral hemorrhage (sICH), and to evaluate the accuracy of catheter placement.Methods:Fifty-two sICH patients received free-handed conical craniotomy and drainage guided by novel 16-slice mobile CT-assisted surface projection in our hospital from January 2018 to December 2019 were chosen (free-handed conical craniotomy group); 30 sICH patients received frameless stereotactic puncture and drainage at the same time period were selected (stereotactic puncture group). The clinical data of these patients were retrospectively analyzed. The CT results were analyzed, and differences of relative error (RE) as the indicator of catheter placement accuracy were compared between the two groups.Results:Mobile CT was successfully performed in all patients from free-handed conical craniotomy group, and sufficient information was provided for surface projection in all patients. The percentages of patients with satisfactory results of catheter placement (RE<1) in the free-handed conical craniotomy group and stereotactic puncture group were 92.3% and 90.0%; one patients from the free-handed conical craniotomy group had repeated puncture. There was no significant difference in postoperative RE between the two groups (0.52±0.33 vs. 0.53±0.29, P>0.05). Subgroup analysis of different hematoma locations and volumes also showed no statistically significant difference in postoperative RE ( P>0.05). Conclusion:Free-handed conical craniotomy and drainage guided by novel 16-slice mobile CT-assisted surface projection is feasible in sICH patients, and the accuracy of catheter placement is similar with frameless stereotactic puncture and drainage.

Objective:To analyze the EGFR mutation profile of lung cancer patients in Yunnan, and to provide evidence for clinical personalized treatment.Methods:Demographic and clinical data of 2 967 lung cancer patients undergoing EGFR identification were collected and analyzed from January 2014 to August 2019 in Yunnan Cancer Hospital.Results:The proportion of EGFR mutation in 2 967 patients with lung cancer was 46.2%. Univariate analysis showed that the proportion of EGFR mutation in women was higher than that in men ( P<0.001) and displayed a downward trend with age ( P=0.03). The mutation rate of ethnic minorities was higher than Han ( P=0.012). Mutation rate in patients without smoking history was higher than those with smoking history ( P<0.001), and patients without drinking history was higher than patients with drinking history ( P<0.001). Mutation rate in patients without family history of lung cancer was higher than those with family history ( P=0.008). The mutation rate of adenocarcinoma was higher than other pathological types ( P<0.001). The mutation rate was different among stages, and it was higher in early patients than that in advanced patients ( P<0.001). The mutation rate of tissue specimens was higher than those of cytology and peripheral blood samples ( P<0.001). The mutation rate of Xuanwei area was lower than that in non-Xuanwei area ( P<0.001). Multivariate analysis showed that gender ( P<0.001), age ( P=0.036), smoking history ( P<0.001), pathological type ( P<0.001), specimen type ( P<0.001), and whether or not Xuanwei area ( P<0.001) were the independent factors of EGFR mutation.The EGFR mutation was more common in female, non-smokers, adenocarcinoma, non-Xuanwei area, tissue specimen and young lung cancer patients.The mutation types of EGFR in 1 370 cases mainly included 19-Del and L858R. The predominant mutation of EGFR in Xuanwei area was L858R, while in non-Xuanwei area was 19-Del.The mutation rates of G719X, G719X+ L861Q, G719X+ S768I, and S768I in Xuanwei were higher while the mutation rates of 19-Del, L858R, and 20-ins were lower than non-Xuanwei area ( P<0.05). The 19-Del mutation rate of ethnic minorities is higher than that of Han ( P<0.001). The combined mutation rate of G719X, L861Q in Han was higher than that of ethnic minorities ( P=0.005). Conclusions:The EGFR mutation rate in lung cancer patients in Yunnan is similar to Asian and Chinese, and higher in female, non-smokers, adenocarcinomas, young and non-Xuanwei area patients. The most common types of EGFR mutation in Yunnan are 19-Del and L858R. The predominant mutation of EGFR in Xuanwei area is L858R, while in non-Xuanwei area is 19-Del. The mutation rates of G719X, G719X+ L861Q, G719X+ S768I and S768I are higher in Xuanwei patients than those in non-Xuanwei patients. The combined mutation rate of G719X and L861Q in Han nationality is higher than that of ethnic minorities.

Objective:To analyze the expressions of non-small-cell lung cancer (NSCLC) driver genes and their mutation distribution characteristics in the Yunnan-Kweichow plateau, and to provide evidences for personalized molecular targeted therapy of lung cancer in high-incidence areas.Methods:A retrospective analysis was performed on the medical records of patients with NSCLC who underwent combined lung cancer 8 gene detection, including epidermal growth factor receptor (EGFR), rat sarcoma viral oncogene (RAS), anaplastic lymphoma kinase (ALK), RET proto-oncogene (RET), v-Raf murine sarcoma viral oncogene homolog (BRAF), ROS proto-oncogene 1 (ROS1), human epidermal growth factor receptor-2 (HER-2), and cellular-mesenchymal to epithelial transition factor (MET), from January 2016 to August 2019 in Yunnan Cancer Hospital. Besides, we analyzed the expressions of NSCLC driver genes and their mutation distributions.Results:The positive rate of NSCLC driver genes in Yunnan was 67.05%(1 508/2 249). The mutation rates in Xishuangbanna (76.92%), Yuxi (72.38%), Xuanwei (71.88%), Qujing (71.24%), and Honghe (71.79%) were significantly higher than other areas. The mutation rates of Hui (84.38%), Hani (85.00%), Zhuang (75.00%), Buyi (100%), Manchu (100%), Tujia (100%) and Achang (100%) are significantly higher than the minority national average. Driver gene mutations were related to gender ( P<0.001), smoking history ( P<0.001), age ( P<0.001), pathological type ( P<0.001), and whether the Xuanwei area ( P=0.027), but not related to the nationality ( P=0.748) and family history of lung cancer ( P=0.676). The mutation rates of EGFR, RAS, BRAF, HER-2 and MET genes were 44.46%, 10.98%, 1.24%, 0.89% and 0.76%, and the rearrangement rates of ALK, RET and ROS1 genes were 4.67%, 1.29% and 0.89%, respectively.The mutation rate of EGFR in females was 56.67%, which was higher than 33.19% in males ( P<0.001). The mutation rate of RAS in males was 12.66%, which was higher than 9.17% in females ( P=0.010). The mutation rate of RAS in the Han was 11.49%, which was higher than 7.17% in the minority ( P=0.032). The rate of RAS mutation in Xuanwei patients was 24.74%, significantly higher than 8.15% in non-Xuanwei area ( P<0.001), and the EGFR mutation rate was 40.63%, which was lower than 45.25% in non-Xuanwei area ( P=0.045). The rate of ALK rearrangement in Xuanwei patients was 1.56%, which was significantly lower than 5.31% in the non-Xuanwei area ( P<0.001), and no HER-2 mutation patients were detected in Xuanwei area. The mutation rate of EGFR in patients with non-smoking history was 51.10%, significantly higher than 29.70% of patients with smoking history ( P<0.001). Meanwhile, the rate of ALK rearrangement with non-smoking history patients was 5.35%, which was also higher than 3.16% of patients with smoking history ( P<0.001). The rate of RAS mutation in patients with non-smoking history was 9.34%, lower than 14.63% of patients with smoking history ( P=0.008). Conclusions:The positive rate of driven gene expression in NSCLC patients from the Yunnan-Kweichow Plateau is slightly lower than the national average. The rates of EGFR and RAS mutations are similar to the domestic average. The rates of ROS1, ALK and RET genes rearrangements and the rates of BRAF, HER2 and MET gene mutations are slightly lower than the national average. EGFR, RAS and ALK genes in the NSCLC patients from Yunnan-Kweichow Plateau have high positive rates, and display different demographic and clinical characteristics, which are of great significance in the selection of targeted therapy populations.

Objective:To analyze the EGFR mutation profile of lung cancer patients in Yunnan, and to provide evidence for clinical personalized treatment.Methods:Demographic and clinical data of 2 967 lung cancer patients undergoing EGFR identification were collected and analyzed from January 2014 to August 2019 in Yunnan Cancer Hospital.Results:The proportion of EGFR mutation in 2 967 patients with lung cancer was 46.2%. Univariate analysis showed that the proportion of EGFR mutation in women was higher than that in men ( P<0.001) and displayed a downward trend with age ( P=0.03). The mutation rate of ethnic minorities was higher than Han ( P=0.012). Mutation rate in patients without smoking history was higher than those with smoking history ( P<0.001), and patients without drinking history was higher than patients with drinking history ( P<0.001). Mutation rate in patients without family history of lung cancer was higher than those with family history ( P=0.008). The mutation rate of adenocarcinoma was higher than other pathological types ( P<0.001). The mutation rate was different among stages, and it was higher in early patients than that in advanced patients ( P<0.001). The mutation rate of tissue specimens was higher than those of cytology and peripheral blood samples ( P<0.001). The mutation rate of Xuanwei area was lower than that in non-Xuanwei area ( P<0.001). Multivariate analysis showed that gender ( P<0.001), age ( P=0.036), smoking history ( P<0.001), pathological type ( P<0.001), specimen type ( P<0.001), and whether or not Xuanwei area ( P<0.001) were the independent factors of EGFR mutation.The EGFR mutation was more common in female, non-smokers, adenocarcinoma, non-Xuanwei area, tissue specimen and young lung cancer patients.The mutation types of EGFR in 1 370 cases mainly included 19-Del and L858R. The predominant mutation of EGFR in Xuanwei area was L858R, while in non-Xuanwei area was 19-Del.The mutation rates of G719X, G719X+ L861Q, G719X+ S768I, and S768I in Xuanwei were higher while the mutation rates of 19-Del, L858R, and 20-ins were lower than non-Xuanwei area ( P<0.05). The 19-Del mutation rate of ethnic minorities is higher than that of Han ( P<0.001). The combined mutation rate of G719X, L861Q in Han was higher than that of ethnic minorities ( P=0.005). Conclusions:The EGFR mutation rate in lung cancer patients in Yunnan is similar to Asian and Chinese, and higher in female, non-smokers, adenocarcinomas, young and non-Xuanwei area patients. The most common types of EGFR mutation in Yunnan are 19-Del and L858R. The predominant mutation of EGFR in Xuanwei area is L858R, while in non-Xuanwei area is 19-Del. The mutation rates of G719X, G719X+ L861Q, G719X+ S768I and S768I are higher in Xuanwei patients than those in non-Xuanwei patients. The combined mutation rate of G719X and L861Q in Han nationality is higher than that of ethnic minorities.