Objective To explore the value of CT-based radiomics and machine learning in predicting the severity of community acquired pneumonia(CAP)in children.Methods The clinical and imaging data of 158 patients diagnosed with CAP in children were analyzed retrospectively.All patients were randomly divided into training set(n=110)and validation set(n=48)in a ratio of 7︰3.Radiomics features were outlined and extracted using 3D Slicer software,and feature selection was achieved using maximum relevance and minimum redundancy(MRMR)and least absolute shrinkage and selection operator(LASSO)algorithms.The construction of the nomogram model and the machine learning combined model was performed by combining clinical features and Radiomics score(Radscore),and its performance was evaluated and validated.Results The area under the curve(AUC)of the clinical model,the radiomics model and the nomogram model in the validation set were classified as 0.82,0.86 and 0.91,respectively.The AUC of the combined multi-layer perceptron(MLP),random forest(RF),and adaptive boosting(ADB)models were 0.926,0.934 and 0.917,respectively in the validation set.Conclusion Radiomics combined with clinical data is expected to be a novel predictor of the severity of CAP in children.MLP,RF and ABD machine learning algorithms can further enable model performance.

Objective To explore the value of CT-based radiomics and machine learning in predicting the severity of community acquired pneumonia(CAP)in children.Methods The clinical and imaging data of 158 patients diagnosed with CAP in children were analyzed retrospectively.All patients were randomly divided into training set(n=110)and validation set(n=48)in a ratio of 7︰3.Radiomics features were outlined and extracted using 3D Slicer software,and feature selection was achieved using maximum relevance and minimum redundancy(MRMR)and least absolute shrinkage and selection operator(LASSO)algorithms.The construction of the nomogram model and the machine learning combined model was performed by combining clinical features and Radiomics score(Radscore),and its performance was evaluated and validated.Results The area under the curve(AUC)of the clinical model,the radiomics model and the nomogram model in the validation set were classified as 0.82,0.86 and 0.91,respectively.The AUC of the combined multi-layer perceptron(MLP),random forest(RF),and adaptive boosting(ADB)models were 0.926,0.934 and 0.917,respectively in the validation set.Conclusion Radiomics combined with clinical data is expected to be a novel predictor of the severity of CAP in children.MLP,RF and ABD machine learning algorithms can further enable model performance.

OBJECTIVETo evaluate the effect of techniques of precise liver surgery for donor hepatectomy in living donor liver transplantation.

METHODSEighty-nine donors aged from 19 to 57 years were performed by the same surgical team from June 2006 to December 2013 in Chinese People's Liberation Army General Hospital.Individualized surgical program were developed according to preoperative imaging examination and hepatic functional reserve examination. The evaluation included liver function, liver volume, vascular anatomy and bile duct anatomy. According to the results after the operation, preoperative evaluation accuracy, postoperative donor liver function and postoperative complications were analyzed. ANOVA analysis was used to compare the difference of graft volume by two-dimensional, three-dimensional calculation method and actual postoperative graft weight. Pearson correlation test and linear regression analysis were used to verify the correlation between the estimated graft volume each method and actual graft postoperative weight.

RESULTSAll the 89 cases operation protocol as following, there were 5 cases with left lateral lobe graft, 10 cases with left lobe liver graft, 74 cases with right lobe graft. There were 59 cases with middle hepatic vein (MHV) harvested, and 30 cases without MHV. The mean graft volume by two-dimensional, three-dimensional calculation method and actual postoperative graft weight were (656.2±134.1) ml, (631.7±143.2) ml and (614.5±137.7) ml respectively. ANOVA analysis results showed that there were no statistically significant difference in the three methods (P>0.05). Compared to the actual postoperative graft weight, the average error rate of the two methods were 7.9% and 5.3% respectively. Pearson correlation test showed the graft volume calculated by two-dimensional and three-dimensional methods had a significantly positive correlation with actual graft weight (r=0.821, 0.890, P<0.01) and linear regression analyze showed the R2 were 0.674 and 0.792, respectively. The accuracy rate of preoperative evaluation about portal vein, hepatic vein, hepatic artery and bile duct were 100%, 100%, 97.8% and 95.5%, respectively. The preoperative plan and postoperative practical scheme coincidence rate was 95.5%. Overall donor complication rate was 7.4%. All donors were alive. Sixteen donors received right lobe hepatectomy with gallbladder preserved had a good liver function and gallbladder function.

CONCLUSIONThrough the precise preoperative evaluation, surgical planning, fine operation and excellent postoperative management, precise liver surgery technique can ensure the safety of donor in living donor liver transplantation.

Adult ; Bile Ducts ; Body Weight ; Hepatectomy ; methods ; Hepatic Artery ; Hepatic Veins ; Humans ; Linear Models ; Liver Transplantation ; methods ; Living Donors ; Middle Aged ; Portal Vein ; Postoperative Complications ; Postoperative Period ; Young Adult

Objective To explore the influence of tranexamic acid used in different modes in total hip arthroplasty (T HA ) blood loss by control experiment .Methods 60 patients accepted total hip arthroplasty from orthopaedics in our hospital were se‐lected between March 2010 to August 2013 ,among them femoral neck fracture were 47 cases and 13 cases were osteonecrosis .aged between 45‐82 years old ,and 62 in average .All gave unilateral total hip arthroplasty .All patients were divided into three groups ,A group(contradistinction group) ,B ,C group(experiment group) .each groups include 6 men and 14 women ,no revision surgery pa‐tients .For group A ,100 mL normal saline was dripped both in half on hour before surgery and 3 hours after surgery ;For group B , tranexamic acid diluted in 100 mL normal saline according to 10 mg/kg was dripped half on hour before surgery ,100 mL normal sa‐line was dripped 3 hour after surgery ;For group C ,tranexamic acid diluted in 100 mL normal saline according to 10 mg/kg was dripped both half on hour before surgery and 3 hour after surgery .Compute and record the visible blood cell loss and hidden blood loss ,the comparative analysis was conducted to discuss the effectiveness and safety of tranexamic acid used in the two methods .Re‐sults The visible blood cell loss in each group were (196 .20 ± 44 .45)mL ,(114 .84 ± 35 .21)mL and (104 .47 ± 30 .01)mL ;hidden blood loss in each group were:(614 .50 ± 98 .41)mL ,(425 .74 ± 70 .01)mL and (337 .12 ± 52 .23)mL .Conclusion In the unilateral total hip arthroplasty ,the use of tranexamic acid can significantly reduce a significant amount of visible and hidden blood loss .Com‐pared with tranexamic acid dripped just half one hour before surgery ,dripping tranexamic acid both half one hour before surgery and 3 hour after surgery reduced more hidden blood loss ,decreased transfusion requirement ,and this does not significantly increase the risk of deep venous thrombosis .

OBJECTIVETo investigate the feasibility of no nasogastric intubation and early oral feeding at will after thoracolaparoscopic esophagectomy for patients with esophageal cancer.

METHODSBetween January 2013 and January 2014, the feasibility of no nasogastric intubation and early oral feeding at postoperative day(POD) 1 after thoracolaparoscopic esophagectomy was prospectively investigated in 156 patients (trial group) with esophageal cancer in the Henan Cancer Hospital. One hundred and sixty patients previously managed in the same unit who were treated routinely after thoracolaparoscopic esophagectomy were served as control group.

RESULTSOf 156 patients of trial group, 6(3.8%) patients could not take food early as planned because of postoperative complications. The overall complication rate in trial group was 19.2%(30/156), which was 25.0%(30/160) in control group (P=0.217). The anastomotic leakage in trial group and control group was 2.6%(4/156) and 4.3%(7/160) respectively (P=0.380). Compared with control group, time to first flatus [(2.1±0.9) d vs. (3.3±1.1) d, P<0.001], bowel movement [(4.4±1.3) d vs. (6.6±1.0) d, P<0.001] and postoperative hospital stay [(8.3±3.2) d vs. (10.4±3.6) d, P<0.001] were significantly shorter in trial group.

CONCLUSIONSNo nasogastric intubation and early oral feeding postoperatively in patients with thoracolaparoscopic esophagectomy is feasible and safe. This management can shorten postoperative hospital stay and fasten postoperative bowel function recovery.

Eating ; Esophageal Neoplasms ; surgery ; Esophagectomy ; Fasting ; Feasibility Studies ; Humans ; Intubation, Gastrointestinal ; Postoperative Complications ; Postoperative Period

Objective To investigate the diagnosis and surgical treatment of functional insulinomas.Methods One hundred and thirty two patients from January 2002 to May 2013 were retrospectively reviewed who was confirmed as functional insulinomas by pathology.Results There were 43 males and 89 females.56 cases were located in the head of pancreas,76 body and tail of pancreas.69 patients underwent open enucleation,48 patients had Laparoscopic surgery,5 patients underwent distal pancreatectomy,9 patients had Laparoscopic distal pancreatectomy,1 patients had pancreaticoduodenectomy.The difference of variant imaging examination in Checking insulinomas was statistical significant(P ＜0.01).Laparoscopic treatment of insulinomas was at least as safe as open treatment.Conclusions Imaging examination could improve the accuracy of preoperative diagnosis of insulinomas.Surgery was the most effective methods to cure insulinomas and Laparoscopic surgery was becoming increasingly popular in treating insulinomas.

Objective To study the risk factors of delayed gastric emptying (DGE) after pancreaticoduodenectomy (PD).Methods Between Ja(n)uary 1st 2013 and December 31st 2013,data from 196 consecutive patients who underwent PD at the Chinese PLA General Hospital were studied retrospectively.17 factors were examined.Univariate analysis and multivariate logistic regression analysis were used to determine the relative risks.Results DGE occurred in 71 patients (36.2％).The incidences of grade A,grade B and grade C DGE were 22.4％ (44/196),6.1％ (12/196) and 7.7％ (15/196) respectively.There were three postoperative deaths.The overall mortality rate was 1.5％.BMI,Braun anastomosis,clinically relevant postoperative pancreatic fistula (CR-POPF) and intra-abdominal collection were significantly correlated with DGE on univariate analyses.BMI ≥25 kg/m2,CR-POPF,and intra-abdominal collection were independent risk factors on univariate and multivariate regression analyses.Conclusions Post-operative complications were associated with DGE.Early diagnosis and timely treatment for pancreatic fistula and abdominal collection were helpful to decrease morbidity and to promote recovery of DGE.

Objective To investigate the feasibility of no nasogastric intubation and early oral feeding at will after thoracolaparoscopic esophagectomy for patients with esophageal cancer. Methods Between January 2013 and January 2014, the feasibility of no nasogastric intubation and early oral feeding at postoperative day (POD) 1 after thoracolaparoscopic esophagectomy was prospectively investigated in 156 patients (trial group) with esophageal cancer in the Henan Cancer Hospital. One hundred and sixty patients previously managed in the same unit who were treated routinely after thoracolaparoscopic esophagectomy were served as control group. Results Of 156 patients of trial group, 6 (3.8%) patients could not take food early as planned because of postoperative complications. The overall complication rate in trial group was 19.2%(30/156), which was 25.0%(30/160) in control group (P=0.217). The anastomotic leakage in trial group and control group was 2.6%(4/156) and 4.3%(7/160) respectively (P=0.380). Compared with control group, time to first flatus [(2.1±0.9) d vs. (3.3±1.1) d, P<0.001], bowel movement[(4.4±1.3) d vs. (6.6±1.0) d, P<0.001] and postoperative hospital stay [(8.3 ±3.2) d vs. (10.4 ±3.6) d, P<0.001] were significantly shorter in trial group. Conclusions No nasogastric intubation and early oral feeding postoperatively in patients with thoracolaparoscopic esophagectomy is feasible and safe. This management can shorten postoperative hospital stay and fasten postoperative bowel function recovery.

Objective To investigate the feasibility of no nasogastric intubation and early oral feeding at will after thoracolaparoscopic esophagectomy for patients with esophageal cancer. Methods Between January 2013 and January 2014, the feasibility of no nasogastric intubation and early oral feeding at postoperative day (POD) 1 after thoracolaparoscopic esophagectomy was prospectively investigated in 156 patients (trial group) with esophageal cancer in the Henan Cancer Hospital. One hundred and sixty patients previously managed in the same unit who were treated routinely after thoracolaparoscopic esophagectomy were served as control group. Results Of 156 patients of trial group, 6 (3.8%) patients could not take food early as planned because of postoperative complications. The overall complication rate in trial group was 19.2%(30/156), which was 25.0%(30/160) in control group (P=0.217). The anastomotic leakage in trial group and control group was 2.6%(4/156) and 4.3%(7/160) respectively (P=0.380). Compared with control group, time to first flatus [(2.1±0.9) d vs. (3.3±1.1) d, P<0.001], bowel movement[(4.4±1.3) d vs. (6.6±1.0) d, P<0.001] and postoperative hospital stay [(8.3 ±3.2) d vs. (10.4 ±3.6) d, P<0.001] were significantly shorter in trial group. Conclusions No nasogastric intubation and early oral feeding postoperatively in patients with thoracolaparoscopic esophagectomy is feasible and safe. This management can shorten postoperative hospital stay and fasten postoperative bowel function recovery.

For a matured digestive surgeon,pancreaticoduodenectomy (PD) is regarded as one of the most complicated and technically challenging surgical procedure.Based on the accurate interpretation of patient's preoperative imageologic data,we advocate a novel procedure which is called as G-path pylorus-preserving pancreatoduodenectomy (G-path PPPD).We deen G-path PPPD as a standardized procedure for resectable pancreatic head cancer or periampullary carcinoma,which definitely simplify the procedure,save the operative time,achieve R0 resection through en-bloc resection without interruptedly intraoperative exploration and reduce the risk of iatrogenic tumor metastasis.This article introduced the program of G-path PPPD in detail by taking a patient as an example who suffered from pancreatic head cancer accompanied with obstructive jaundice,and discussed the relevant points.