Objective The objective of this study was to investigate the levels of serum miR-21 and miR-34a in patients with advanced gastric cancer before and after Folfox chemotherapy and its relationship with chemotherapy efficacy.Methods Forty-five patients with advanced gastric cancer who underwent Folfox chemotherapy were enrolled from May 2015 to October 2015.The levels of miR-21 and miR-34a were analyzed by Real-time PCR before and after chemotherapy.The relationship between miR-21 and miR-34a levels and chemotherapy efficacy was analyzed.The predictive values of miR-21 and miR-34a on the efficacy of chemotherapy were evaluated by the receiver operating curve(ROC).Results After chemotherapy,6 cases(13.33%)were relieved,10 cases(22.22%)were stable disease(SD),29 cases(64.44%)were disease progress(PD),the objective response rate(ORR)was 13.33%,the disease control rate(DCR)was 35.56%,the median survival time was 8.1 months.The level of miR-21 was down-regulated and up-regulation for miR-34a after chemotherapy.They were statistically significant(P<0.05).The level of miR-21 in the chemotherapy group was lower than that in the ineffective group,and the level of miR-34a was higher than that of the ineffective group(P<0.05).MiR-21 was negatively correlated with the efficacy of chemotherapy(rs=-0.431,P<0.05),while miR-34a was positively correlated with it(rs=0.504,P<0.05).The median survival time was 7.6 months for low-level patients with miR-21 and 6.0 months for high-level patients.The median survival time of miR-34a patients was 5.5 months,compared with 8.3 months for high-level patients,with a statistically significant difference(P<0.05).The Log-Rank test showed that the median survival of patients with low miR-21 was higher than that of patients with high miR-21(P<0.05).The median survival of patients with high miR-34a was higher than that of patients with low miR-34a(P<0.05).The miR-21 combined with miR-34a in predicting chemotherapy efficacy under the curve(AUC)was 0.865,the sensitivity was 80.9% and the specificity was 89.8%.Conclusion miR-21 and miR-34a are correlated with chemotherapy efficacy median survival time of patients with advanced gastric cancer.

Objective:To explore the value of arterial spin labeling (ASL) in detecting epileptogenic zone (EZ) in children with drug-refractory epilepsy (DRE).Methods:From March 2018 to December 2019, 28 children with DRE were collected prospectively in Peking University First Hospital. Structural MRI, ASL sequence, and PET-CT were performed on 28 DRE children. All children underwent surgical treatment. Intraoperative electrocorticogram findings combined with postoperative MRI results were considered the gold standard for locating EZ. A total of 29 EZ were resected in 28 children. Based on the pathological results, the EZ was divided into focal cortical dysplasia (FCD) Ⅰb and Ⅱa group ( n=12), FCD Ⅱ b group ( n=11) and malformation of cortical dysplasia (MCD) group ( n=6). Structural MRI was observed for finding any abnormal changes that could induce epilepsy and was divided into the normal MRI group ( n=13) and the abnormal MRI group ( n=16). The spatial relationship between abnormal areas in the cerebral blood flow (CBF) map and PET images and the gold standard was observed, and the accurate detection rate of EZ was calculated. The region of interest (ROI) on CBF and PET images was drawn. ROIs were defined as EZ, EZ contralateral zone (EZCZ), EZ adjacent zone (EZAZ), EZAZ contralateral zone (EZAZCZ). The CBF and maximum standardized uptake value (SUV max) were measured, and the asymmetry index (AI) value of EZ and EZAZ of CBF and SUV max was calculated respectively. One-way ANOVA was used to compare the difference among 4 regions and 3 pathological types of CBF, SUV max, and AI. The independent sample t-test was used to compare the difference in AI between normal and abnormal MRI groups. Results:In CBF map, the EZ was accurately localized in 89.7% (26/29) of the lesions, in which 24 EZ had decreased perfusion, and 2 EZ had increased perfusion. Among the 24 EZ with decreased perfusion, the CBF of EZ, EZCZ, EZAZ, and EZAZCZ were significantly different( F=8.79, P<0.001). In PET-CT, the EZ was accurately localized in 93.1% (27/29) of the lesions, in which 25 EZ had decreased metabolism, and 2 EZ had increased metabolism. Among the 25 EZ with decreased metabolism, the SUV max of EZ, EZCZ, EZAZ, and EZAZCZ were significantly different ( F=6.40, P=0.001). The AI value of CBF and SUV max of EZ in the abnormal MRI group were larger than those of the normal MRI group, and the difference was statistically significant ( t=3.34, 3.09, P=0.002 , 0.004). There was no statistical difference in the AI values of CBF and SUV max among FCD Ⅰb and Ⅱa group, FCD Ⅱb group and MCD group ( F=2.05, 1.54, P=0.149, 0.234). Conclusions:ASL technology is accurate in detecting EZ. The changes in perfusion and metabolism of normal structural MRI EZ are greater than abnormal structural MRI EZ. There is no obvious difference in CBF and SUVmax changes in different pathological EZ.

Objective: To investigate the clinical value of laparoscopic peritoneal dialysis catheter implantation in peritoneal chemotherapy for gastric cancer with peritoneal metastasis. Methods: From January 2019 to June 2019, the clinical data of 6 patients diagnosed as gastric cancer with peritoneal metastasis were retrospectively analyzed in the Gastrointestinal Surgery Department of Ruijin Hospital Affiliated to Shanghai Jiaotong University School of Medicine. Five were male and 1 was female. The median age was 69.5 (28-77) years. The median body mass index (BMI) was 22.8 (19.6-23.5). All procedures were performed under general anesthesia with endotracheal intubation. The patient′s body position and facility layout in the operating room were consistent with those of laparoscopic gastrectomy. The operator′s position: the main surgeon was located on the right side of the patient, the first assistant stood on the left side of the patient, and the scopist stood between the patient′s legs. Surgical procedure: (1) trocar location: three abdominal trocars was adopted, with one 12 mm umbilical port for the 30° laparoscope (point A). Location of the other two trocars was dependent on the procedure of exploration or biopsy as well as the two polyester cuff position of the peritoneal dialysis catheter: Usually one 5 mm port in the anterior midline 5 cm inferior to the umbilicus point was selected as point B to ensure that the distal end of the catheter could reach the Douglas pouch. The other 5 mm port was located in the right lower quadrant lateral to the umbilicus to establish the subcutaneous tunnel tract, and the proximal cuff was situated 2 cm away from the desired exit site (point C).(2) exploration of the abdominal cavity: a 30° laparoscope was inserted from 12 mm trocar below the umbilicus to explore the entire peritoneal cavity. The uterus and adnexa should be explored additionally for women. Once peritoneal metastasis was investigated and identified, primary laparoscopic peritoneal dialysis catheter implantation was performed so as to facilitate subsequent peritoneal chemotherapy. Ascites were collected for cytology in patients with ascites. (3) peritoneal dialysis catheter placement: the peritoneal dialysis catheter was introduced into the abdominal cavity from point A. Under the direct vision of laparoscopy, 2-0 absorbable ligature was reserved at the expected fixation point of the proximal cuff (point B) for the final knot closure. Non-traumatic graspers were used to pull the distal cuff of peritoneal dialysis catheter out of the abdominal cavity through point B. The 5-mm trocar was removed simultaneously, and the distal cuff was fixed between bilateral rectus sheaths at the anterior midline port site preperitoneally. To prevent subsequent ascites and chemotherapy fluid extravasation, the reserved crocheted wire was knotted. From point C the subcutaneous tunnel tract was created before the peritoneal steath towards the port site lateral to the umbilicus. Satisfactory catheter irrigation and outflow were then confirmed. Chemotherapy regimen after peritoneal dialysis catheterization: all patients began intraperitoneal chemotherapy on the second day after surgery. On the 1st and 8th day of each 3-weeks cycle, paclitaxel (20 mg/m²) was administered through peritoneal dialysis catheter, and paclitaxel (50 mg/m²) was injected intravenously. Meanwhile, S-1 was orally administered twice daily at a dose of 80 mg·m^-2·d^-1 for 14 consecutive days followed by 7-days rest. To observe the patients′ intraoperative and postoperative conditions. Results: All the procedures were performed successfully without intraoperative complications or conversion to laparotomy. No 30 day postoperative complications were observed. The median operative time was 33.5 (23-38) min. The median time to first flatus was 1(1-2) days, and the median postoperative hospital stay was 3 (3-4) days, without short-term complications within 30 days postoperatively. The last follow-up was up to July 10, 2019, and the patients were followed for 4(1-6) months. No ascites extravasation was observed and no death occurred in the 6 patients. There was no catheter obstruction or peritoneal fluid extravasation during and after chemotherapy. Conclusion Laparoscopic peritoneal dialysis catheter implantation was safe and feasible for patients with peritoneal metastasis of gastric cancer. The abdominal exploration, tumor staging and the abdominal chemotherapy device implantation can be completed simultaneously, which could simplify the surgical approach, improve the quality of life for patients and further propose a new direction for the development of abdominal chemotherapy.

Objective:To explore the selection of surgical methods for different sites of symptomatic Rathke's cleft cyst (RCC) and the clinical efficacies of these patients.Methods:Forty-seven patients with symptomatic RCC, admitted to our hospital from January 2016 to December 2019, were chosen in our study; 21 patients with intrasellar symptomatic RCC accepted surgery via unilateral nasal approach at the right side, 19 patients with intra-suprasellar symptomatic RCC accepted surgery via bilateral nasal approach, 3 patients with suprasellar symptomatic RCC accepted endonasal transsphenoidal surgery under endoscope, and 4 patients with suprasellar symptomatic RCC accepted craniotomy via pterion approach. The clinical efficacies and complications of patients accepted different surgical methods were compared. All patients were followed up for 3-36 months to observe the recurrence.Results:The postoperative symptoms of the patients were effectively improved, including headache relief ratio of 27/31, vision loss improvement ratio of 5/5, high prolactin relief ratio of 11/13, pituitary function improvement ratio of 9/18. Complications occurred in 6 patients, presenting as diabetes insipidus. Four patients recurred during follow-up.Conclusion:Intrasellar and intra-suprasellar symptomatic RCC accepted surgery via endoscopic transnasal transsphenoidal approach are safe and effective; selection of surgical methods for suprasellar symptomatic RCC should be determined according to the sizes and growth directions of cysts.

Objective To investigate the characteristics of metabolic foci on 18F-fluorodeoxyglucose (FDG) PET/CT scan in pediatric patients with epilepsy.Methods Twenty-three pediatric patients (15 males,8 females,age range:0.5-13.3 years) with epilepsy were retrospectively reviewed from March 2014 to December 2016.All patients underwent 18F-FDG PET/CT and metabolic foci were found.The visual method and semi-quantitative analysis were used to analyze images.Fourteen of them underwent surgery and were followed up for 3-24 months.Results Glucose hypermetabolism were observed most frequently in the frontal and parietal lobes,with or without surrounding/remote hypometabolism.On the day of PET/CT imaging,8 patients had no seizures,14 patients had seizures,and 1 patient was uncertain.The sites of resection were consistent with the regions of hypermetabolism in 9 patients,among whom the pathological results showed 8 cortical malformations and 1 Rasmussen's syndrome.Follow-up results for the above 9 patients showed that there was 7 Engel class Ⅰ patients and 2 Engel class Ⅲ patients.Conclusion The hypermetabolism may mostly appear in the frontal and parietal lobes of pediatric patients with epilepsy,and malformations of cortical development seem to be the most common pathology results.

Objective:To investigate the application value of three-dimensional (3D) laparoscope in the transanal total mesorectal excision (TaTME).Methods:The retrospective cohort study was conducted. The clinicopathological data of 20 patients with middle and low rectal cancer who underwent TaTME in the Ruijin Hospital Affiliated to Shanghai Jiaotong University School of Medicine from June 2018 to October 2019 were collected. There were 15 males and 5 females, aged from 28 to 81 years, with a median age of 64 years. Of the 20 patients, 10 patients using 3D laparoscopic system for transanal approach of TaTME were divided into 3D group, and 10 patients using two-dimensional (2D) laparoscopic system for transanal approach of TaTME were divided into 2D group. Observation indicators: (1) intraoperative situations and postoperative recovery; (2) postoperative pathological examination; (3) follow-up. Follow-up was conducted by outpatient examination and telephone interview to detect survival of patients and recurrence and metastasis of tumors in patients up to April 2020. Measurement data with skewed distribution were represented as M (range), and comparison between groups was analyzed using the Mann-Whitney U test. Count data were described as absolute numbers, and comparison between groups was analyzed using the Fisher exact propability. Comparison of ordinal data was analyzed using the Mann-Whitney U test. Results:(1) Intraoperative situations and postoperative recovery: patients in the two groups completed surgeries successfully, without tranversion to laparostomy from laparoscopic surgery, transversion to transabdominal surgery from transanal surgery, or intraoperative death. The cases with terminal ileostomy, cases with manual anstomosis or mechanical anastomosis (anastomotic methods), operation time, volume of intra-operative blood loss, duration of postoperative hospital stay, cases with anastomotic leakage or anastomotic hemorrahge (postoperative short-term complications), cases with anastomotic stenosis of the 3D group were 7, 4, 6, 150 minutes (range, 100-220 minutes), 50 mL (range, 30-100 mL), 8.5 days (range, 7.0-16.0 days), 2, 0, 1, respectively, versus 8, 5, 5, 180 minutes (range, 120-250 minutes), 100 mL (range, 30-200 mL), 9.5 days (range, 6.0-17.0 days), 1, 1, 1 of the 2D group. There was no significant difference in the terminal ileostomy, anastomotic methods, postoperative short-term complications, or anastomotic stenosis between the two groups ( P>0.05). There was no significant difference in the operation time, volume of intraoperative blood loss, or duration of postoperative hospital stay between the two groups ( Z=1.909, 1.827, 0.687, P>0.05). Patients with short-term complications in the two groups were improved after conservative treatments. There was 1 patient with anastomotic stenosis in either group, and they were improved after endoscopic balloon dilatation. (2) Postoperative pathological examination: the maximum tumor diameter, distal margin of the tumor, the number of lymph nodes harvested, cases with cancer nodes in the mesentery, cases with complete mesentery or median complete mesentery (the integrity of mesentery), cases in stage Ⅰ, Ⅱ, Ⅲ (postoperative pathological stage) of the 3D group were 3.8 cm (range, 1.8-5.0 cm), 1.0 cm (range, 0.5-2.5 cm), 14.5 (range, 6.0-19.0), 1, 9, 1, 4, 2, 4, respectively, versus 4.8 cm (range, 1.0-8.5 cm), 0.8 cm (range, 0.3-1.5 cm), 15.5 (range, 8.0-18.0), 1, 8, 2, 2, 4, 4 of the 2D group. There was no significant difference in the maximum tumor diameter, distal margin of the tumor, the number of lymph nodes harvested, the integrity of mesentery, or postoperative pathological stage between the two groups ( Z=1.673, 1.772, 0.038, 0.610, 0.482, P>0.05). There was no significant difference in the cases with cancer nodes in the mesentery between the two groups ( P>0.05). Patients in the two groups had negative distal margin and circumferential margin. (3) Follow-up: patients in the 3D group and 2D group were followed up for 11 months (range, 6-16 months) and 13 months (range, 6-21 months), respectively. During the follow-up, there was no local recurrence, distal metastasis, or tumor-related death. Conclusions:3D laparoscope applied in the TaTME can achieve similar clinical efficacy with 2D laparoscope, which may have a positive impact on the operation time and volume of intraoperative blood loss.

Objective:To primarily investigate the application value of glasses-free three-dimensional (3D) laparoscopy system in laparoscopic radical resection of gastrointestinal malignant tumors.Methods:The retrospective cohort study was conducted. The clinical data of 165 patients with gastrointestinal malignant tumors who were admitted to the Ruijin Hospital of Shanghai Jiaotong University School of Medicine between October 2018 and May 2019 were collected. There were 99 males and 66 females, aged from 28 to 86 years, with a median age of 63 years. There were 68 of 165 patients with gastric cancer and 97 with colorectal cancer. Sixteen patients with gastric cancer who underwent laparoscopic radical gastrectomy using the glasses-free 3D laparoscopy system were divided into glasses-free 3D gastric cancer group, and 52 patients with gastric cancer who underwent laparoscopic radical gastrectomy using the polarized glasses 3D laparoscopy system were divided into polarized glasses 3D gastric cancer group. Nineteen patients with colorectal cancer who underwent laparoscopic radical colectomy or proctectomy using the glasses-free 3D laparoscopy system were divided into glasses-free 3D colorectal cancer group, and 78 patients with colorectal cancer who underwent laparoscopic radical colectomy or proctectomy using the polarized glasses 3D laparoscopy system were divided into polarized glasses 3D colorectal cancer group. Observation indicators: (1) operative situations of patients with gastric cancer; (2) postoperative recovery of patients with gastric cancer; (3) postoperative pathological examination results of patients with gastric cancer; (4) operative situations of patients with colorectal cancer; (5) postoperative recovery of patients with colorectal cancer; (6) postoperative pathological examination results of patients with colorectal cancer; (7) follow-up. Follow-up was conducted by outpatient examination or telephone interview to detect complications and survival of patients up to the postoperative 30th day. Measurement data with skewed distribution were represented as M ( P25, P75) or M (range), and comparison between groups was analyzed by the Mann-Whitney U test. Count data were represented as absolute numbers or percentages, and comparison between groups was analyzed using the chi-square test or Fisher exact probability. Results:(1) Operative situations of patients with gastric cancer: all the 68 gastric cancer patients received successfully laparoscopic radical gastrectomy, without intraoperative complication or conversion to laparotomy. Cases with distal gastrectomy or total gastrectomy(surgical methods) , operation time, volume of intraoperative blood loss were 11, 5, 195 minutes(169 minutes, 214 minutes), 20 mL (10 mL, 90 mL) in patients of the glasses-free 3D gastric cancer group, and 31, 21, 196 minutes(173 minutes, 222 minutes), 40 mL(20 mL, 100 mL) in patients of the polarized glasses 3D gastric cancer group, with no significant difference between the two groups ( χ2=0.432, Z=-0.362, -1.065, P>0.05). (2) Postoperative recovery of patients with gastric cancer: the time to first flatus, time to initial semi-fluid diet intake, duration of postoperative hospital stay were 2 days(2 days, 3 days), 6 days(5 days, 7 days), 10 days(9 days, 14 days) in patients of the glasses-free 3D gastric cancer group, and 2 days(2 days, 3 days), 6 days(5 days, 6 days), 11 days(9 days, 14 days) in patients of the polarized glasses 3D gastric cancer group, with no significant difference between the two groups ( Z=-0.163, -1.870, -0.570, P>0.05). The postoperative complication incidence of the glasses-free 3D gastric cancer group was 12.5%(2/16), including 1 case with duodenal stump fistula, 1 case with anastomotic bleeding. The postoperative complication incidence of the polarized glasses 3D gastric cancer group was 17.3%(9/52), including 2 cases with duodenal stump fistula, 2 cases with delayed gastric emptying, 1 case with pulmonary infection, 1 case with abdominal bleeding, 1 case with anastomotic leakage, 1 case with chylous fistula, 1 case with intestinal obstruction. All the patients with complications were cured after symptomatic supportive treatment. There was no significant difference in the complication incidence between the two groups ( χ2=0.209, P>0.05). (3) Postoperative pathological examination results of patients with gastric cancer: the tumor diameter, cases in stage T1, T2, T3, T4 (tumor T staging), cases with vascular invasion, cases with nerve invasion, cases with tumor nodule, cases in stage N0, N1, N2, N3 (tumor N staging), the number of positive lymph node, the number of lymph node dissected, cases with qualified lymph node dissected, cases in stage Ⅰ, Ⅱ, Ⅲ, Ⅳ (TNM clinical staging) were 3.0 cm(2.0 cm, 5.0 cm), 5, 3, 1, 7, 8, 5, 2, 3, 2, 4, 7, 6(1, 15), 28(22, 43), 15, 4, 3, 9, 0 in patients of the glasses-free 3D gastric cancer group, and 3.5 cm(2.0 cm, 6.0 cm), 13, 10, 4, 25, 19, 23, 2, 26, 6, 7, 13, 1(0, 7), 29(21, 39), 43, 21, 10, 20, 1 in patients of the polarized glasses 3D gastric cancer group. There was no significant difference in the tumor diameter, tumor T staging, cases with vascular invasion, cases with nerve invasion, cases with tumor nodule, the number of lymph node dissected, cases with qualified lymph node dissected, TNM clinical staging between the two groups ( Z=-0.570, -0.434, χ2 =0.926, 0.851, 1.655, Z=-0.579, χ2=1.193, Z=-1.134, P>0.05). There were significant differences in the tumor N staging and the number of positive lymph node between the two groups ( Z=-2.167, -2.283, P<0.05). (4) Operative situations of patients with colorectal cancer: all the 97 colorectal cancer patients received successfully laparoscopic radical colectomy or proctectomy, without intraoperative complications or conversion to laparotomy. Cases with radical colectomy or proctectomy (surgical methods), operation time, volume of intraoperative blood loss were 7, 12, 132 minutes(97 minutes, 156 minutes), 20 mL(10 mL, 50 mL) in patients of the glasses-free 3D colorectal cancer group, and 40, 38, 124 minutes(110 minutes, 159 minutes), 25 mL(15 mL, 65 mL) in patients of the polarized glasses 3D colorectal cancer group, with no significant difference between the two groups ( χ2 =1.276, Z=-0.141, -0.863, P>0.05). (5) Postoperative recovery of patients with colorectal cancer: the time to first flatus, time to initial semi-fluid diet intake, duration of postoperative hospital stay were 2 days(1 days, 3 days), 5 days(5 days, 6 days), 8 days(7 days, 10 days) in patients of the glasses-free 3D colorectal cancer group, and 2 days(1 days, 3 days), 5 days(4 days, 6 days), 8 days(6 days, 10 days) in patients of the polarized glasses 3D colorectal cancer group, with no significant difference between the two groups ( Z=-0.678, -1.751, -1.674, P>0.05). The complication incidence of the glasses-free 3D colorectal cancer group was 15.8%(3/19), including 1 case with anastomotic leakage after low anterior proctectomy, 1 case with incision infection, 1 case with urinary tract infection. The complication incidence of the polarized glasses 3D colorectal cancer group was 14.1%(11/78), including 3 cases with anastomotic leakage after low anterior proctectomy, 2 cases with intestinal obstruction, 2 cases with urinary tract infection, 2 cases with incision infection, 1 case with anastomotic bleeding, 1 case with pulmonary infection. One of the 3 cases with anastomotic leakage after low anterior proctectomy in the polarized glasses 3D colorectal cancer group was cured after remedial terminal ileostomy. The other patients with complications were cured after symptomatic supportive treatment. There was no significant difference in the complication incidence between the two groups ( χ2=0.035, P>0.05). (6) Postoperative pathological examination results of patients with colorectal cancer: the tumor diameter, cases in stage T1, T2, T3, T4 (tumor T staging), cases with vascular invasion, cases with nerve invasion, cases with tumor nodule, cases in stage N0, N1-N2 (tumor N staging), the number of positive lymph node, the number of lymph node dissected, cases with qualified lymph node dissected, cases in stage Ⅰ, Ⅱ, Ⅲ, Ⅳ (TNM clinical staging) were 5.0 cm(3.0 cm, 6.0 cm), 3, 2, 7, 7, 3, 2, 1, 8, 11, 0(0, 4), 17(14, 23), 18, 2, 3, 12, 2 in patients of the glasses-free 3D colorectal cancer group, and 4.0 cm(3.0 cm, 5.0 cm), 7, 16, 43, 12, 14, 12, 7, 46, 32, 0(0, 1), 16(13, 19), 74, 14, 17, 40, 7 in patients of the polarized glasses 3D colorectal cancer group, with no significant difference between the two groups ( Z=-0.768, -1.135, χ2 =0.049, 0.292, 0.278, 1.762, Z=-0.694, -1.349, χ2=0.001, Z=-1.011, P>0.05). (7) Follow-up: 165 patients received follow-up, with out short-term reoperation or postoperative death in the postoperative 30 days. Conclusion:There is no significant difference in the efficacy between glasses-free 3D laparoscopic surgery and polarized glasses 3D laparoscopic surgery for radical resection of gastrointestinal malignant tumors, of which the clinical value requires further study.

Objective:To investigate the clinical value of magnetic resonance imaging (MRI) based integrated deep learning model for predicting the times of linear staplers used in double stapling technique for middle-low rectal cancer resection.Methods:The retrospective cohort study was conducted. The clinicopathological data of 263 patients who underwent low anterior resection (LAR) for middle-low rectal cancer in Ruijin Hospital of Shanghai Jiaotong University School of Medicine from January 2018 to December 2022 were collected as training dataset. There were 183 males and 80 females, aged 63(55,68)years. The clinicopathological data of 128 patients with middle-low rectal cancer were collected as validation dataset, including 83 males and 45 females, with age as 65(57,70)years. The training dataset was used to construct the prediction model, and the validation dataset was used to validate the prediction model. Observation indicators: (1) clinicopathological features of patients in the training dataset; (2) influencing factors for ≥3 times using of linear staplers in the operation; (3) prediction model construction; (4) efficiency evaluation of prediction model; (5) validation of prediction model. Measurement data with skewed distribution were represented as M( Q1, Q3), and Mann-Whitney U test was used for comparison between groups. Count data were expressed as absolute numbers, and comparison between groups was conducted using the chi-square test. Wilcoxon rank sum test was used for non-parametric data analysis. Univariate analysis was conducted using the Logistic regression model, and multivariate analysis was conducted using the Logistic stepwise regression model. The receiver operating characteristic (ROC) curve was draw and the area under the curve (AUC) was calculated. The AUC of the ROC curve >0.75 indicated the prediction model as acceptable. Comparison of AUC was conducted using the Delong test. Results:(1) Clinicopathological features of patients in the training dataset. Of the 263 patients, there were 48 cases with linear staplers used in the operation ≥3 times and 215 cases with linear staplers used in the operation ≤2 times. Cases with preoperative serum carcinoembryonic antigen (CEA) >5 μg/L, cases with anastomotic leakage, cases with tumor diameter ≥5 cm were 20, 12, 13 in the 48 cases with linear staplers used ≥3 times in the operation, versus 56, 26, 21 in the 215 cases with linear staplers used ≤2 times in the operation, showing significant differences in the above indicators between them ( χ2=4.66, 5.29, 10.45, P<0.05). (2) Influencing factors for ≥3 times using of linear staplers in the operation. Results of multivariate analysis showed that preoperative serum CEA >5 μg/L and tumor diameter ≥5 cm were independent risk factors for ≥3 times using of linear staplers in the operation ( odds ratio=2.26, 3.39, 95% confidence interval as 1.15-4.43, 1.50-7.65, P<0.05). (3) Prediction model construction. According to the results of multivariate analysis, the clinical prediction model was established as Logit(P)=-2.018+0.814×preoperative serum CEA (>5 μg/L as 1, ≤5 μg/L as 0)+ 1.222×tumor diameter (≥5 cm as 1, <5 cm as 0). The image data segmented by the Mask region convolutional neural network (MASK R-CNN) was input into the three-dimensional convolutional neural network (C3D), and the image prediction model was constructed by training. The image data segmented by the MASK R-CNN and the clinical independent risk factors were input into the C3D, and the integrated prediction model was constructed by training. (4) Efficiency evaluation of prediction model. The sensitivity, specificity and accuracy of the clinical prediction model was 70.0%, 81.0% and 79.4%, respectively, with the Yoden index as 0.51. The sensitivity, specificity and accuracy of the image prediction model was 50.0%, 98.3% and 91.2%, respectively, with the Yoden index as 0.48. The sensitivity, specificity and accuracy of the integrated prediction model was 70.0%, 98.3% and 94.1%, respectively, with the Yoden index as 0.68. The AUC of clinical prediction model, image prediction model and integrated prediction model was 0.72(95% confidence interval as 0.61-0.83), 0.81(95% confidence interval as 0.71-0.91) and 0.88(95% confidence interval as 0.81-0.95), respectively. There were significant differences in the efficacy between the integrated prediction model and the image prediction model or the clinical prediction model ( Z=2.98, 2.48, P<0.05). (5) Validation of prediction model. The three prediction models were externally validated by validation dataset. The sensitivity, specificity and accuracy of the clinical prediction model was 62.5%, 66.1% and 65.6%, respectively, with the Yoden index as 0.29. The sensitivity, specificity and accuracy of the image prediction model was 58.8%, 95.5% and 92.1%, respectively, with the Yoden index as 0.64. The sensitivity, specificity and accuracy of the integrated prediction model was 68.8%, 97.3% and 93.8%, respectively, with the Yoden index as 0.66. The AUC of clinical prediction model, image prediction model and integrated prediction model was 0.65(95% confidence interval as 0.55-0.75), 0.75(95% confidence interval as 0.66-0.84) and 0.84(95% confidence interval as 0.74-0.93), respec-tively. There was significant differences in the efficacy between the clinical prediction model and the integrated prediction model ( Z=3.24, P<0.05). Conclusion:The MRI-based deep-learning model can help predicting the high-risk population with ≥3 times using of linear staplers in resection of middle-low rectal cancer with double stapling technique.