Abstract: Objective To find out the factors which influence plasma N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and assess whether preoperative plasma NT-proBNP levels can predict postoperative outcomes of cardiac surgery. Methods A total of 120 patients including 83 males and 37 females undergoing various cardiac procedures between December 2008 and May 2009 were included in the study. Their age ranged from 25 to 84 years with an average age of 62.13 years. Through pathological diagnosis, 35 patients had heart valve diseases, 74 had coronary artery diseases, 3 had congenital heart diseases and 8 had aortic aneurysm. NT-proBNP, creatinine, cardiac troponin T (cTnT) and creatine kinase-MB (CK-MB) levels were measured preoperatively and 24 hours after operation. Ventilation time, length of stay in ICU or in hospital, and mortality were closely monitored after operation. The following events were regarded as endpoints: (1) ICU stay timegt;4 d; (2) Ventilation timegt;48 h; (3) Death occurred during the first 30 days after operation. Receiver operating characteristic (ROC) curve was used to analyze the prediction function of NTproBNP on endpoint events. Based on the cutoff value, the patients were divided into the NT-proBNP increasing group and nonincreasing group. Univariate and logistic multifactor analysis were adopted to analyze factors which had an influence on preoperative NT-proBNP level. Results NT-proBNP concentration [CM(159mm]increased significantly from 37.5-30 867.0 pg/ml (1 929.12±3 749.44 pg/ml) preoperatively to 177.7-35 000.0pg/ml(2 950.32±4 006.14 pg/ml) 24 hours after operation (t=-2.599, P=0.012). ROC curve demonstrated that a cutoff value above 867 pg/ml preoperatively could predict endpoint events with a sensitivity of 77.8% and a specificity of 62.7%. Ventilation time and length of stay in hospital for the patients in the NT-proBNP increasing group were significantly longer than those of patients in the nonincreasing group (26.44±32.75 h vs. 14.49±9.23 h, t=2.507, P=0.015; 23.70±24.02 d vs. 16.21±8.11 d, t=2.117,P=0.039). Influencing factors on preoperative NTproBNP level included preoperative atrial fibrillation, heart function classification, left ventricular enddiastolic dimension (LVEDD), ejection fraction (EF), pulmonary artery pressure, preoperative creatinine, cTnT and pathological diagnosis. EF (P=0.007) and preoperative atrial fibrillation (P=0.018) were independently associated with preoperative NT-proBNP level. Preoperative NTproBNP was closely related to ventilation time (P=0.015), and length of stay in hospital (P=0.039). Conclusion Preoperative plasma NT-proBNP level presents a high individual variability in patients undergoing cardiac surgery. Ejection fraction and preoperative atrial fibrillation are independently associated with preoperative NT-proBNP level. Preoperative NT-proBNP is a valuable marker in predicting bad outcome in patients undergoing heart surgery.
Objective To explore the impact of diabetes on coronary artery bypass grafting （CABG） in clinical representations, operative morbidity and mortality in this hospital. Methods Data was collected as a part of prospective registry of CABG through Sep. 2001 to Jul. 2003. Four hundreds and eighty-two patients were recruited. They were divided into diabetic group （n= 135） and non-diabetic group （n=347） depended on if the patients with diabetes or not. All patients were treated with insulin for hyperglycemia. Clinical representations, operative morbidity and mortality in this hospital between two groups were compared by using chi-square tests, t tests and logistic regression. Results Re-exploration in diabetic group was higher than that in non-diabetic group （4.4% vs. 0. 9%; x2= 6. 769, P = 0. 009）. There was no significant difference in the operative morbidity and mortality in hospital between two groups. Multi-variance logistic regression showed that the lower left ventricular ejection fraction （〈 0. 40,OR 15.96）, re-exploration （OR 32. 77） and re-intubation （OR 124.17） were the predictors of perioperative mortality in hospital. Conclusions There are no significant difference in the operative mortality and complication between patients with diabetes and patients with non-diabetes. Strict glucose control in perioperative period would reduce hospital mortality and morbidity.
Computer-aided diagnosis (CAD) systems play a very important role in modern medical diagnosis and treatment systems, but their performance is limited by training samples. However, the training samples are affected by factors such as imaging cost, labeling cost and involving patient privacy, resulting in insufficient diversity of training images and difficulty in data obtaining. Therefore, how to efficiently and cost-effectively augment existing medical image datasets has become a research hotspot. In this paper, the research progress on medical image dataset expansion methods is reviewed based on relevant literatures at home and abroad. First, the expansion methods based on geometric transformation and generative adversarial networks are compared and analyzed, and then improvement of the augmentation methods based on generative adversarial networks are emphasized. Finally, some urgent problems in the field of medical image dataset expansion are discussed and the future development trend is prospected.
Medical image segmentation based on deep learning has become a powerful tool in the field of medical image processing. Due to the special nature of medical images, image segmentation algorithms based on deep learning face problems such as sample imbalance, edge blur, false positive, false negative, etc. In view of these problems, researchers mostly improve the network structure, but rarely improve from the unstructured aspect. The loss function is an important part of the segmentation method based on deep learning. The improvement of the loss function can improve the segmentation effect of the network from the root, and the loss function is independent of the network structure, which can be used in various network models and segmentation tasks in plug and play. Starting from the difficulties in medical image segmentation, this paper first introduces the loss function and improvement strategies to solve the problems of sample imbalance, edge blur, false positive and false negative. Then the difficulties encountered in the improvement of the current loss function are analyzed. Finally, the future research directions are prospected. This paper provides a reference for the reasonable selection, improvement or innovation of loss function, and guides the direction for the follow-up research of loss function.