ABSTRACT
BACKGROUND: Neuroendocrine dysfunction after traumatic brain injury (TBI) has received increased attention due to its impact on the recovery of neural function. The purpose of this study is to investigate the incidence and risk factors of adrenocortical insufficiency (AI) after TBI to reveal independent predictors and build a prediction model of AI after TBI. METHODS: Enrolled patients were grouped into the AI and non-AI groups. Fourteen preset impact factors were recorded. Patients were regrouped according to each impact factor as a categorical variable. Univariate and multiple logistic regression analyses were performed to screen the related independent risk factors of AI after TBI and develop the predictive model. RESULTS: A total of 108 patients were recruited, of whom 34 (31.5%) patients had AI. Nine factors (age, Glasgow Coma Scale [GCS] score on admission, mean arterial pressure [MAP], urinary volume, serum sodium level, cerebral hernia, frontal lobe contusion, diffuse axonal injury [DAI], and skull base fracture) were probably related to AI after TBI. Three factors (urinary volume [X 4], serum sodium level [X 5], and DAI [X 8]) were independent variables, based on which a prediction model was developed (logit P= -3.552+2.583X 4+2.235X 5+2.269X 8). CONCLUSIONS: The incidence of AI after TBI is high. Factors such as age, GCS score, MAP, urinary volume, serum sodium level, cerebral hernia, frontal lobe contusion, DAI, and skull base fracture are probably related to AI after TBI. Urinary volume, serum sodium level, and DAI are the independent predictors of AI after TBI.
ABSTRACT
Pedestrian detection have been currently devoted to the extraction of effective pedestrian features, which has become one of the obstacles in pedestrian detection application according to the variety of pedestrian features and their large dimension. Based on the theoretical analysis of six frequently-used features, SIFT, SURF, Haar, HOG, LBP and LSS, and their comparison with experimental results, this paper screens out the sparse feature subsets via sparse representation to investigate whether the sparse subsets have the same description abilities and the most stable features. When any two of the six features are fused, the fusion feature is sparsely represented to obtain its important components. Sparse subsets of the fusion features can be rapidly generated by avoiding calculation of the corresponding index of dimension numbers of these feature descriptors; thus, the calculation speed of the feature dimension reduction is improved and the pedestrian detection time is reduced. Experimental results show that sparse feature subsets are capable of keeping the important components of these six feature descriptors. The sparse features of HOG and LSS possess the same description ability and consume less time compared with their full features. The ratios of the sparse feature subsets of HOG and LSS to their full sets are the highest among the six, and thus these two features can be used to best describe the characteristics of the pedestrian and the sparse feature subsets of the combination of HOG-LSS show better distinguishing ability and parsimony.
