Objective To evaluate the predictive value of lactate dehydrogenase (LDH) to albumin (Alb) ratio (LAR) in the prognosis of severe pneumonia patients complicated with DIC. Methods A total of 312 patients with severe pneumonia hospitalized in the intensive care unit (ICU) of the Affiliated Changzhou No.2 People's Hospital with Nanjing Medical University from January 1, 2018 to March 1, 2023 were retrospectively collected. The clinical parameters, such as gender, age, underlying diseases, and lactate dehydrogenase, albumin etc. l of the first test on admission were collected. LAR, sequential organ failure assessment (SOFA) and acute physiology and chronic health evaluation Ⅱ(APACHEⅡ) within 24 hours were calculated. The firstly endpoint of the study was the incidence of disseminated intravascular coagulation (DIC), the secondary endpoint was the 30-day in-hospital mortality in severe pneumonia patients with DIC. Univariate and multivariate logistic regression were used to analyze the risk factors of severe pneumonia with DIC. The receiver operating characteristic curve (ROC curve) was drawn and the area under the ROC curve (AUC) was calculated to evaluate the predictive value of LAR for the incidence of DIC in patients with severe pneumonia. Results The level of LAR was higher in the severe pneumonia patients with DIC than the severe pneumonia patients without DIC [LAR median ratio 12.72 (8.72, 21.89) vs. 7.23 (5.63, 10.90), P<0.001]. Multiple logistic regression analysis showed that LAR [OR=1.071, 95%CI 1.038 - 1.106, P<0.001] was the independent risk factor of the incidence of DIC in the patients with severe pneumonia. ROC curve analysis showed that the AUC for LAR to predict the incidence of DIC was 0.723, 95%CI 0.650 - 0.796, P<0.001. When the LAR cut-off value was 8.08, the sensitivity was 79.7% and the specificity was 56.1%. Kaplan-Meier survival analysis curve showed that the patients in the above LAR cut-off value group had a significantly lower 30-day survival rate than those in the below LAR cut-off value group (P<0.001). In the subgroup analysis and numerical variable transformed analysis, LAR was still the risk factor of DIC. Conclusion The increased LAR is a high risk factor of the incidence of DIC and mortality in patients with severe pneumonia, which is useful for predicting prognosis of patients with severe pneumonia.
The segmentation of organs at risk is an important part of radiotherapy. The current method of manual segmentation depends on the knowledge and experience of physicians, which is very time-consuming and difficult to ensure the accuracy, consistency and repeatability. Therefore, a deep convolutional neural network (DCNN) is proposed for the automatic and accurate segmentation of head and neck organs at risk. The data of 496 patients with nasopharyngeal carcinoma were reviewed. Among them, 376 cases were randomly selected for training set, 60 cases for validation set and 60 cases for test set. Using the three-dimensional (3D) U-NET DCNN, combined with two loss functions of Dice Loss and Generalized Dice Loss, the automatic segmentation neural network model for the head and neck organs at risk was trained. The evaluation parameters are Dice similarity coefficient and Jaccard distance. The average Dice Similarity coefficient of the 19 organs at risk was 0.91, and the Jaccard distance was 0.15. The results demonstrate that 3D U-NET DCNN combined with Dice Loss function can be better applied to automatic segmentation of head and neck organs at risk.