ABSTRACT
Intensive care unit (ICU) is a very special unit of a hospital, where healthcare professionals provide treatment and, later, close follow-up to the patients. It is crucial to estimate mortality in ICU patients from many viewpoints. The purpose of this study is to classify the status of patients with acute kidney injury (AKI) in ICU as early mortality, late mortality, and survival by the application of Classification and Regression Trees (CART) algorithm to the patients' attributes such as blood urea nitrogen, creatinine, serum and urine neutrophil gelatinase-associated lipocalin (NGAL), alkaline phosphatase, lactate dehydrogenase (LDH), gamma-glutamyl transferase, laboratory electrolytes, blood gas, mean arterial pressure, central venous pressure and demographic details of patients. This study was conducted 50 patients with AKI who were followed up in the ICU. The study also aims to determine the significance of relationship between the attributes used in the prediction of mortality in CART and patients' status by employing the Kruskal-Wallis H test. The classification accuracy, sensitivity, and specificity of CART for the tested attributes for the prediction of early mortality, late mortality, and survival of patients were 90.00%, 83.33%, and 91.67%, respectively. The values of both urine NGAL and LDH on day 7 showed a considerable difference according to the patients' status after being examined by the Kruskal-Wallis H test.
ABSTRACT
AIM: To compare the results of midazolam-ketamine-propofol sedation performed by an endoscopy nurse and anaesthetist during colonoscopy in terms of patient satisfaction and safety. METHODS: American Statistical Association (ASA) I-II 60 patients who underwent colonoscopy under sedation were randomly divided into two groups: sedation under the supervision of an anaesthetist (SSA) and sedation under the supervision of an endoscopy nurse (SSEN). Both groups were initially administered 1 mg midazolam, 50 mg ketamine and 30-50 mg propofol. Continuation of sedation was performed by the anaesthetist in the SSA group and the nurse with a patient-controlled analgesia (PCA) pump in the SSEN group. The total propofol consumption, procedure duration, recovery times, pain using the visual analogue scale (VAS) and satisfaction score of the patients, and side effects were recorded. In addition, the patients were asked whether they remembered the procedure and whether they would prefer the same method in the case of re-endoscopy. RESULTS: Total propofol consumption in the SSEN group was significantly higher (P < 0.05) than that in the SSA group. When the groups were compared in terms of VAS score, recovery time, patient satisfaction, recall of the procedure, re-preference for the same method in case of re-endoscopy, and side effects, there were no significant differences (P > 0.05) between the two groups. No long-term required intervention side effects were observed in either group. CONCLUSION: Colonoscopy sedation in ASA I-II patients can be safely performed by an endoscopy nurse using PCA pump with the incidence of side effects and patient satisfaction levels similar to sedation under anaesthetist supervision.
ABSTRACT
We investigated the association between the textural features obtained from 18F-FDG images, metabolic parameters (SUVmax, SUVmean, MTV, TLG), and tumor histopathological characteristics (stage and Ki-67 proliferation index) in non-small cell lung cancer (NSCLC). The FDG-PET images of 67 patients with NSCLC were evaluated. MATLAB technical computing language was employed in the extraction of 137 features by using first order statistics (FOS), gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM), and Laws' texture filters. Textural features and metabolic parameters were statistically analyzed in terms of good discrimination power between tumor stages, and selected features/parameters were used in the automatic classification by k-nearest neighbors (k-NN) and support vector machines (SVM). We showed that one textural feature (gray-level nonuniformity, GLN) obtained using GLRLM approach and nine textural features using Laws' approach were successful in discriminating all tumor stages, unlike metabolic parameters. There were significant correlations between Ki-67 index and some of the textural features computed using Laws' method (r = 0.6, p = 0.013). In terms of automatic classification of tumor stage, the accuracy was approximately 84% with k-NN classifier (k = 3) and SVM, using selected five features. Texture analysis of FDG-PET images has a potential to be an objective tool to assess tumor histopathological characteristics. The textural features obtained using Laws' approach could be useful in the discrimination of tumor stage.
