This article reviews the challenges faced by research teams in China, and drawing on advanced management modes both in China and abroad, proposes a "dual optimization of management modes, stable and efficient platform support, and flexible and precise talent development" mode for research team construction and student cultivation. Specifically, this mode includes promoting flexible team management through enhanced two-way feedback between mentors and students, improving shared experimental platforms to increase resource utilization efficiency, and designing tailored training programs that align with individual student needs. The aim is to enhance team cohesion, improve the communication, collaboration, and innovation in basic medical research teams, and ultimately raise the quality of talent development in universities.

Objective:To develop a deep learning-based system for real-time recognition and classification of portal hypertensive gastropathy (PHG) and evaluate its ability to assist junior endoscopists.Methods:A total of 2 848 gastroscopy images from 832 patients with liver cirrhosis were selected from Digestive Endoscopy Center databases of Renmin Hospital of Wuhan University, Wuhan Hospital of Traditional Chinese and Western Medicine, and the Second Hospital of Jingzhou from January 2015 to October 2023. This system referred to 3 endoscopic features of Baveno Ⅱ scoring system. Three models were developed respectively for gastric antral vascular ectasia (GAVE), mosaic-like pattern (MLP), and red marks (RM). The specific classification references were as follows: (1) GAVE model: 0 no, 1 yes; (2) MLP model: 0 no, 1 mild, 2 severe; (3) RM model: 0 no, 1 isolated, 2 fused. The classification results for endoscopic characteristics of PHG of 3 endoscopy experts were taken as the gold standard. The yolov8-m model was used for training. The training dataset, validation dataset, and test dataset were allocated at a ratio of 8∶1∶1. The test dataset was used to evaluate the performance of models and their auxiliary effects on endoscopists. The accuracy, recall, precision, specificity and Kappa coefficient were calculated. Results:The accuracy, recall, specificity of GAVE model were 96.0% (48/50), 87.5% (7/8) and 97.6% (41/42). There was no significant difference between its accuracy and the gold standard ( χ2=316.226, P=1.000). The precision of GAVE1 and GAVE0 were 87.5% (7/8) and 97.6% (41/42) respectively. The accuracy of MLP model was 84.1% (132/157), and there was no significant difference compared with the gold standard ( χ2=3.286, P=0.193). The precision and recall of MLP2 were 88.2% (15/17) and 75.0% (15/20). The precision and recall of MLP1 were 77.9% (60/77) and 88.2% (60/68). The precision and recall of MLP0 were 90.5% (57/63) and 82.6% (57/69). The accuracy of RM model was 87.9% (123/140), and there was no significant difference compared with the gold standard ( χ2=2.891, P=0.409). The precision and recall of RM2 were 94.7% (18/19) and 78.3% (18/23). The precision and recall of RM1 were 72.2% (26/36) and 81.3% (26/32). The precision and recall of RM0 were 92.9% (79/85) and 92.9% (79/85). The mean accuracy of the three junior endoscopists, with and without the assistance of the GAVE model, MLP model, and RM model, respectively increased from 95.3% to 99.3%, from 83.9% to 91.9%, and from 81.9% to 83.1%. The overall consistency analysis of the 3 junior endoscopists with the gold standard indicated that the consistency of the GAVE model before and after assistance was extremely strong (both an overall Kappa of 1.000); the consistency before assistance of the MLP model was moderate (with an overall Kappa of 0.601), which increased to extremely strong after assistance (with an overall Kappa of 0.964); and the consistency of the RM model before and after assistance was also relatively strong (with an overall Kappa of 0.792 before and 0.798 after). Conclusion:The deep learning system accurately identifies and classifies PHG features and significantly enhances diagnostic performance of junior endoscopists.

This article reviews the challenges faced by research teams in China, and drawing on advanced management modes both in China and abroad, proposes a "dual optimization of management modes, stable and efficient platform support, and flexible and precise talent development" mode for research team construction and student cultivation. Specifically, this mode includes promoting flexible team management through enhanced two-way feedback between mentors and students, improving shared experimental platforms to increase resource utilization efficiency, and designing tailored training programs that align with individual student needs. The aim is to enhance team cohesion, improve the communication, collaboration, and innovation in basic medical research teams, and ultimately raise the quality of talent development in universities.

Objective:To develop a deep learning-based system for real-time recognition and classification of portal hypertensive gastropathy (PHG) and evaluate its ability to assist junior endoscopists.Methods:A total of 2 848 gastroscopy images from 832 patients with liver cirrhosis were selected from Digestive Endoscopy Center databases of Renmin Hospital of Wuhan University, Wuhan Hospital of Traditional Chinese and Western Medicine, and the Second Hospital of Jingzhou from January 2015 to October 2023. This system referred to 3 endoscopic features of Baveno Ⅱ scoring system. Three models were developed respectively for gastric antral vascular ectasia (GAVE), mosaic-like pattern (MLP), and red marks (RM). The specific classification references were as follows: (1) GAVE model: 0 no, 1 yes; (2) MLP model: 0 no, 1 mild, 2 severe; (3) RM model: 0 no, 1 isolated, 2 fused. The classification results for endoscopic characteristics of PHG of 3 endoscopy experts were taken as the gold standard. The yolov8-m model was used for training. The training dataset, validation dataset, and test dataset were allocated at a ratio of 8∶1∶1. The test dataset was used to evaluate the performance of models and their auxiliary effects on endoscopists. The accuracy, recall, precision, specificity and Kappa coefficient were calculated. Results:The accuracy, recall, specificity of GAVE model were 96.0% (48/50), 87.5% (7/8) and 97.6% (41/42). There was no significant difference between its accuracy and the gold standard ( χ2=316.226, P=1.000). The precision of GAVE1 and GAVE0 were 87.5% (7/8) and 97.6% (41/42) respectively. The accuracy of MLP model was 84.1% (132/157), and there was no significant difference compared with the gold standard ( χ2=3.286, P=0.193). The precision and recall of MLP2 were 88.2% (15/17) and 75.0% (15/20). The precision and recall of MLP1 were 77.9% (60/77) and 88.2% (60/68). The precision and recall of MLP0 were 90.5% (57/63) and 82.6% (57/69). The accuracy of RM model was 87.9% (123/140), and there was no significant difference compared with the gold standard ( χ2=2.891, P=0.409). The precision and recall of RM2 were 94.7% (18/19) and 78.3% (18/23). The precision and recall of RM1 were 72.2% (26/36) and 81.3% (26/32). The precision and recall of RM0 were 92.9% (79/85) and 92.9% (79/85). The mean accuracy of the three junior endoscopists, with and without the assistance of the GAVE model, MLP model, and RM model, respectively increased from 95.3% to 99.3%, from 83.9% to 91.9%, and from 81.9% to 83.1%. The overall consistency analysis of the 3 junior endoscopists with the gold standard indicated that the consistency of the GAVE model before and after assistance was extremely strong (both an overall Kappa of 1.000); the consistency before assistance of the MLP model was moderate (with an overall Kappa of 0.601), which increased to extremely strong after assistance (with an overall Kappa of 0.964); and the consistency of the RM model before and after assistance was also relatively strong (with an overall Kappa of 0.792 before and 0.798 after). Conclusion:The deep learning system accurately identifies and classifies PHG features and significantly enhances diagnostic performance of junior endoscopists.

Objective:To explore the efficacy and adverse reactions of nimotuzumab combined with induction chemotherapy (IC) based on albumin-bound paclitaxel plus cisplatin (TP regimen) for patients with locally advanced head and neck squamous cell carcinoma (LA-HNSCC).Methods:Clinical data were collected from 65 patients with LA-HNSCC (stages Ⅲ/Ⅳ A/Ⅳ B; excluding nasopharyngeal carcinoma) who received 2-3 cycles of IC followed by concurrent chemoradiotherapy (CRT) in the Second Hospital of Dalian Medical University from January 2018 to June 2022. Based on the IC regimen, these patients were categorized into a nimotuzumab combined with TP (Nimo-TP) group ( n = 34) and a TP group ( n = 31), and their short-term efficacy [i.e., the objective response rate (ORR)], survival outcomes [e.g., overall survival (OS), progression-free survival (PFS), local recurrence-free survival (LRFS), and distant metastasis-free survival (DMFS)], and adverse reactions were compared. Additionally, factors affecting their survival outcomes were analyzed. Results:There were statistically significant differences in 1- and 2-year DMFS between both groups (90.4% vs. 69.5%, 90.4% vs. 66.0%, χ2=1.81, P < 0.05), so did the ORRs after IC and CRT of both groups (after IC: 67.6% vs. 41.9%, χ2=4.34, P = 0.037; after CRT: 88.2% vs. 67.7%, χ2=4.03, P = 0.045). However, there was no statistically significant difference in the 2-year OS, PFS, and LRFS between both groups ( P > 0.05). Multivariate analysis revealed that nimotuzumab combined with TP-based IC served as an independent prognostic factor for DMFS ( HR = 0.27, 95% CI: 0.07-0.97, P = 0.045), while complete/partial response after IC acted as an independent prognostic factor for both PFS and local relapse-free survival ( HR = 0.36, 95% CI: 0.17-0.76, P = 0.008; HR = 0.28, 95% CI: 0.11-0.69, P = 0.006). Notably, adding nimotuzumab did not aggravate the adverse reactions in the patients during IC and CRT( P > 0.05). Conclusions:Nimotuzumab combined with TP-based induction chemotherapy followed by CRT significantly improved the DMFS of LA-HNSCC patients, exhibiting high safety. However, such therapy failed to significantly improve their OS, PFS, and LRRFS, and, thus, further research is required.

Objective To observe the value of pseudo-continuous arterial spin labeling(PCASL)MRI for evaluating renal function in patients with renal occupying lesions.Methods Totally 56 patients with single renal occupying lesion were retrospectively enrolled.The left and right side kidneys were divided into normal renal function group(normal group,30 ml/min)and damaged renal function group(damaged group,＜30 ml/min)according to glomerular filtration rate(GFR)measured with 99Tcm-DTPA dynamic renal imaging,respectively.The total renal blood flow(tRBF)and cortical renal blood flow(cRBF)were calculated using total nephrometry and cortical nephrometry based on PCASL MRI,respectively,then GFR,tRBF and cRBF were compared between groups on the same side.Receiver operating characteristic curve was drawn,and the area under the curve(AUC)was calculated to evaluate the efficacy of tRBF and cRBF for assessing unilateral renal injury.Pearson correlation analysis was performed to observe the correlations of tRBF and cRBF with GFR.Results GFR,tRBF and cRBF in left/right damaged group were all significantly lower than those in ipsilateral normal group(all P＜0.05).AUC of tRBF and cRBF for assessing left renal injury was 0.823 and 0.813,respectively,being not significantly different(P＞0.05).AUC of tRBF and cRBF for assessing right renal injury was 0.940 and 0.922,respectively,being not significantly different(P＞0.05).No obvious correlation of bilateral tRBF nor cRBF with GFR was found(all P＞0.05).Conclusion PCASL MRI could effectively evaluate renal function in patients with renal occupying lesion,and the efficacy of total nephrometry was comparable to that of cortical nephrometry.

Objective To discuss the safety and clinical efficacy of modified traction method using endoloop and clip for endoscopic submucosal dissection(ESD).Methods Fifty patients who underwent ESD at Renmin Hosptial of Wuhan University between August 2016 and February 2017 were randomly divided into two groups, including 25 patients in modified ESD group and 25 patients in conventional ESD group as control. The therapeutic conditions, dissection time and incidence of complications were compared between the two groups. Results In the modified ESD group, the dissection time of submucosal exposure to the full dissection (19. 9±6. 5 min VS 26. 4±9. 2 min, P＝0. 001), total dissection time (27. 5±8. 1 min VS 35. 1± 10. 7 min, P＝0. 003), and dissection time per unit area (2. 4±1. 1 min/cm2VS 3. 3±1. 3 min/cm2, P＝0. 009) were significantly shorter compared with those in the control group. There were 1 case of delayed bleeding in the modified group and 2 cases in the control group with no significant difference ( P＝0. 248). No perforation occurred. Conclusion The modified traction method using endoloop and clip for ESD is safe and effective with a shorter operation time.

Objective To compare the effectiveness and safety of endoscopic submucosal dissection ( ESD) with endoscopic piecemeal mucosal resection ( EPMR) for early esophageal cancer and precancerous lesions with length more than 5 cm. Methods A retrospective analysis was performed on data of 85 patients diagnosed as early esophageal cancer and precancerous lesions with length more than 5 cm in Fujian Medical Association of Early Esophageal Carcinoma from January 2012 to July 2017. The patients were divided into ESD group (52 cases) and EPMR group (33 cases), and the effectiveness and safety between the two groups were compared. Results There was no significant difference on the complete resection rate between the two groups[86. 5％ (45/52) VS 87. 9％ (29/33), P>0. 05]. The operative time (58. 53±30. 50 min VS 32. 06±9. 12 min), postoperative fasting time (4. 18±1. 30 d VS 3. 67±0. 96 d), postoperative hospital-stay time (7. 45±2. 44 d VS 6. 54±1. 73 d), and postoperative antibiotics using time (3. 48±2. 33 d VS 1. 96±2. 20 d) in ESD group were higher than those in EPMR group (all P<0. 05). There were no significant difference in the rate of intraoperative complication and short-term postoperative complication, such as fever, chest pain, and postoperative bleeding, between the two groups ( all P>0. 05 ) . But the postoperative stricture rate of ESD group was higher than that of EPMR group[23. 1％ (12/52) VS 6. 1％(2/33), P<0. 05]. During the follow-up of 3-63 months, 5 cases recurred in ESD group and 1 case in EPMR group, with no significant difference ( P>0. 05). Conclusion ESD and EPMR have equivalent efficacy and safety on the treatment of early esophageal cancer and precancerous lesion. EPMR has a shorter operative time, lower rate of post-operative stricture, and is easier to master.

[Summary] A more conservative approach for endometrial cancer ( EC ) is desirable in women who have requirement to preserve fertility, without compromising their survival .In this article, we tried to formulate clinical recommendations of preserving fertility based on suggestions of European Society of Gynaecological Oncology ( ESGO) in 2015 and relevant research progress .

Objective To investigate the clinical effect of "pre-purse-string" suture on submucous gastric lesions of external growth type in endoscopic patients. Methods Clinical data of 27 patients with gastric lesions of external growth type admitted to the digestive endoscopy center of Renmin Hospital of Wuhan University from August 2016 to October 2016 for endoscopic submucosal dissection ( ESD ) were analyzed. Patients were randomly divided into two groups, the experiment group receiving the "pre-purse-string" suture, the control group receiving purse-string suture after ESD. Mean size of the lesions, procedure time, withdrawal time of gastric decompression, and hospitalization time were compared. Results In the 15 patients of the experiment group, there were 3 cases whose lesions were located in gastric antrum, 6 in gastric body and 6 in gastric fundus. In the 12 cases of the control group, 3 lesions were located in gastric antrum, 5 in gastric body and 4 in gastric fundus. All procedures were successfully completed in endoscopy. No obvious postoperative complications occurred. The mean size of lesions was 2. 5 cm × 3. 2 cm in the experiment group,and 2. 4 cm×3. 0 cm in the control group (P=0. 63). The operation time was 28. 0-43. 0 min (35. 8±6. 1 min) in the experiment group, and 45. 6-68. 8 min (48. 8±5. 3 min) in control group ( P<0. 01) . Conclusion "Pre-purse-string" technique is of clinical effect on gastric submucosal tumor of external growth.