Sleep electroencephalogram (EEG) is an important index in diagnosing sleep disorders and related diseases. Manual sleep staging is time-consuming and often influenced by subjective factors. Existing automatic sleep staging methods have high complexity and a low accuracy rate. A sleep staging method based on support vector machines (SVM) and feature selection using single channel EEG single is proposed in this paper. Thirty-eight features were extracted from the single channel EEG signal. Then based on the feature selection method F-Score's definition, it was extended to multiclass with an added eliminate factor in order to find proper features, which were used as SVM classifier inputs. The eliminate factor was adopted to reduce the negative interaction of features to the result. Research on the F-Score with an added eliminate factor was further accomplished with the data from a standard open source database and the results were compared with none feature selection and standard F-Score feature selection. The results showed that the present method could effectively improve the sleep staging accuracy and reduce the computation time.
Abstract： Objective To validate the value of Cleveland Clinical Score to predict acute renal failure（ARF） requiring renal replacement therapy （RRT） and in-hospital death in Chinese adult patients after cardiac surgery. Methods A retrospective analysis was conducted for all the patients who underwent cardiac surgery from January 2005 to December 2009 in Renji Hospital of School of Medicine, Shanghai Jiaotong University. A total of 2 153 adult patients, 1 267 males and 886 females，were included. Their age ranged from 18 to 99 years with an average age of 58.70 years. Cleveland Clinical Score was used to predict ARF after cardiac surgery. ARF was defined as the need for RRT. Based on Cleveland Clinical Score, the patients were divided into four risk categories of increasing severity：0 to 2 point（n=979）, 3 to 5 point （n=1 116）, 6 to 8 point（n=54）, 9 to 13 point（n=4）. The rates of ARF, multiple organ system failure （MOSF）, and mortality were compared among the 4 categories. The predictive accuracy of postoperative ARF and hospital mortality was assessed by area under the receiver operating characteristic curve （AUC-ROC）. Results In the four categories, the rate of postoperative ARF was 0.92%, 1.88%, 12.96%, and 25.00%, respectively； MOSF rate was 1.23%, 1.88%, 3.70%, and 25.00%, respectively; mortality was 0.92%, 4.21%, 25.93%, and 50.00%, respectively. There was significant dif ference among the four categories in ARF rate （χ2=55.635, P=0.000），MOSF rate（χ2=16.080, P=0.001）, and mortality （χ2=71.470, P=0.000）. The AUC-ROC for Cleveland Clinical Score predicting ARF rate and hospital mortality was 0.775 （95%CI 0.713 to 0.837, P=0.000）and 0.764（95%CI, 0.711 to 0.817, P=0.000）, respectively. Conclusion Cleveland Clinical Score can accurately predict postoperative ARF and hospital mortality in a large, unselected Chinese cohort of adult patients after cardiac surgery. It can be used to provide evidence for effective preventive measures for patients at high risk of postoperative ARF.
ObjectiveTo analyze the prognosis factors in status epilepticus and confirm the external validity of the Status Epilepticus Severity Score (STESS) prediction functions for outcome in patients with statusepilepticus. MethodsRetrospectively, collecting prognosis factors in status epilepticus of 50 patients. These factors include gender, age, SE aetiology, seizure type at SE onset, history of prior seizures or epilepsy, level of consciousness, duration of SE, albumin and infection. Using STESS score to evaluate the prognosis of patients, then evaluate the effectiveness of the STESS score. Results1. Single factor analysis:age,history of prior seizures or epilepsy, level of consciousness and infection were prognosis factors in status epilepticus (P<0.05). Gender, SE aetiology, seizure type at SE onset and albumin were not prognosis factors (P>0.05). Multivariablelogistic regression models selected two factors:duration of SE (OR3.645), level of consciousness (OR2.877). 2. 28 status epilepticus patient in the STESS 0-2 group were all alive. 10 status epilepticus patients in the STESS 3-6 group were died (45.4%), 12 patient were alive (54.6%). There were significant differences among the prognostic of patienst in different groups (P<0.01). The receiver operating characteristic curve for prediction of death by the STESS Score had an area under the curve of 0.92. The optimal cut-off point is 3. ConclusionAge, history of prior seizures or epilepsy, level of consciousness, duration of SE and infection were prognosis factors in status epilepticus. Level of consciousness and duration of SE were the directly related factors of the prognosis of status epilepticus. STESS score performed reasonable prognositic role on our patients with status epilepticus.
ObjectiveTo study the diagnostic significance of modified Alvarado Score System for different types of acute appendicitis. MethodsClinical data of 1 930 cases of acute appendicitis, who underwent appendicectomy in The First Hospital of Yulin from Jan. 2004 to Jan. 2014, were retrospectively collected. Then the diagnostic significance of Modified Alvarado Score System was detected and evaluated by receiver operating characteristic (ROC) curve, and its correlation with postoperative pathological results was calculated. ResultsThe modified Alvarado score of cases in progressive group was higher than that of simplex group (7.5±1.3 vs. 3.6±0.7, P < 0.001). The areas under ROC curve of modified Alvarado Score System was 0.943 (95% CI:0.929-0.958, P < 0.001). The critically diagnostic points of Modified Alvarado Score System for differential diagnosis of simplex acute appendicitis and progressive acute appendicitis was 6 score, with sensitivity of 92.9%, specificity of 86.3%, positive predictive value of 96.9%, negative predictive value of 72.6%, positive likelihood ratio of 6.8, negative likelihood ratio of 0.1, and Youden index of 91.7%. ConclusionThis Modified Alvarado Score System has a certain significance for the type diagnosis of acute appendicitis.
ObjectiveTo analyze responsiveness of Chinese version of Neck Outcome Score (NOOS-C) and provide a reliable measure to assess intervention effect for patients with neck pain.MethodsCross-cultural adaptation of NOOS was performed according to the Beaton’s guidelines for cross-cultural adaptation of self-report measures. Eighty patients with neck pain were recruited between September 2016 and May 2017. Those patients were assessed using NOOS-C and Chinese version of Neck Disability Index (NDI) before and after intervention. And 71 patients completed those questionnaires. The statistic differences of the score of each subscale and the total scale before and after intervention were evaluated by paired-samples t test. Internal responsiveness was determined by effect size (ES) and standardized response mean (SRM) based on the calculated difference before and after intervention. External responsiveness was analyzed by Spearman correlation coefficient.ResultsThe differences in symptom subscale, sleep disturbance subscale, participating in everyday life subscale, every day activity and pain subscale, and the scale between before and after intervention were significant (P<0.05) except for mobility subscale (P>0.05). The difference of NDI-C before and after intervention was –12.11%±17.45%, ES was 0.77, and SRM was 0.69. The difference of NOOS-C before and after intervention was 13.74±17.22, ES was 0.83, and SRM was 0.80. Spearman correlation analysis revealed that the relativity about NOOS-C and NDI-C before and after intervention were both negative (r=–0.914, P=0.000; r=–0.872, P=0.000).ConclusionNOOS-C’s responsiveness is good.
Objective We studied the change of psychological behavior and quality of life in children with epilepsy, to explore the treatment strategy to improve their psychological behavior and quality of life. Methods Collected forty children with epilepsy from Hebei General Hospital during March 2015 to October 2015 and 40 normal children in this study. "Epilepsy patients quality of life scale", "Daily life ability scale" and "Sense of uncertainty in hospitalized patients disease scale" test were performed to the participants. The difference of daily psychological behavior ability, sense of uncertainty in illness and quality of life between the two groups were analyzed. Results The epilepsy children’s daily life ability and the quality of life are lower than normal children (P<0.05), the disease uncertainty is higher than normal children (P<0.05). Conclusion The epilepsy children had serious psychological and behavior disorders and lower life quality. Some medical intervention should be taken to alleviate the psychological burden, improve the ability of daily life, reduce the uncertainty in illness, and comprehensively improve the quality of life of children with epilepsy.
ObjectiveTo explore the clinical efficacy of percutaneous endoscopic interlaminar discectomy (PEID) and interlaminar fenestration discectomy in the treatment of lumbar (L) 5-Sacral (S) 1 lumbar disc herniation (LDH).MethodsLDH patients were retrospectively included from January 2016 to Januray 2018. And the patients were divided into the PEID group and the fenestration group according to their choice of different surgical methods. The operation time, intra-operative blood loss, and bed rest time in the two groups were recorded. The preoperative and postoperative [1 week, 1 month, 3 months, and last follow-up (>12 months)] Visual Analogue Score (VAS) of the lumbago and leg pain between the two groups were compared; the preoperative and postoperative [1 week, and last follow-up (>12 months)] Oswestry Disability Index (ODI) and also the postoperative [(>12 months)] therapeutic effect (modified MacNab) between the two groups were compared.ResultsA total of 66 patients were included, with 31 in the PEID group and 35 in the fenestration group. There was no significant difference in age, gender and course of disease between the two groups (P>0.05). There were leakage of cerebrospinal fluid and transient lumbago, leg pain and numbness, which were worse than those before operation in the PEID group (1 and 1 patient, respectively) and the fenestration group (2 and 3 patients, respectively). There were statistically significant differences between the PEID group and the fenestration group, in the operative time [(90.65±9.98) vs. (66.23±16.50) minutes], intra-operative blood loss [(51.77±18.64) vs. (184.29±78.38) mL], and bed time [(2.87±0.92) vs. (7.49±1.20) d] (t=−7.365, t’=−9.697, t=−17.374, P<0.001). There was no significant difference in the preoperative VAS score (lumbar-leg pain) and ODI index, and the ODI index at each postoperative time point, between the two groups (P>0.05). VAS score (lumbago) and VAS score (leg pain) in the PEID group at each postoperative time point were lower than those in the fenestration group (P<0.05); VAS scores (leg pain) at other time points were not statistically significant between the two groups (P > 0.05). VAS (lumbar-leg pain) score and ODI index at each postoperative time point were lower than those before the surgery. The was no statistically significant difference in the PEID group (90.32%) and fenestration group (85.71%) in the excellent rate (χ2=0.328, P=0.713).ConclusionsPEID has less surgical trauma, less bleeding, short bed rest, fast recovery, and better relief of postoperative lumbago symptoms. It is worthy of further promotion in clinical work.
ObjectiveTo explore the prognostic value of modified Glasgow Prognostic Score (mGPS) in lung cancer patients.MethodsThe clinical data and follow-up information of patients with lung cancer diagnosed for the first time in West China Hospital of Sichuan University from August 2008 to May 2013 were retrospectively analyzed. Overall survival (OS) of patients with different mGPS were compared by Kaplan-Meier test and log-rank test. Univariate and multivariate Cox proportional hazard analysis were performed, and hazard ratio (HR) and 95% confidence interval (CI) were counted to evaluate the predictive value of different prognostic factors in patients with lung cancer.ResultsA total of 289 patients were included. According to the mGPS score, 127 patients had 0 point, 90 patients had 1 point, and 72 patients had 2 points. The OS of lung cancer patients with mGPS=0 was better than that of patients with mGPS=1 and mGPS=2 (P<0.001). Cox proportional hazards of univariate analysis revealed that age< 65 (P=0.022), stage for Ⅰand Ⅱ (P<0.001), surgery (P<0.001), chemotherapy (P=0.018), and mGPS=0 (1 vs. 0, P=0.008; 2 vs. 0, P<0.001) were the protective factors for lung cancer patients (P<0.05). Multiple-factor analysis showed that age [HR=0.680, 95%CI (0.508, 0.911), P=0.010], stage [HR=0.580, 95%CI (0.359, 0.939), P=0.027], operation [HR=0.254, 95%CI (0.140, 0.459), P<0.001], chemotherapy [HR=0.624, 95%CI (0.435, 0.893), P=0.010], mGPS (1 vs. 0) [HR=1.548, 95%CI (1.101, 2.176), P=0.012] and mGPS (2 vs. 0) [HR=1.425, 95%CI (1.003, 2.024), P=0.048] were independent predictors of OS in patients with lung cancer.ConclusionmGPS could be considered as an independent prognostic factor in lung cancer.
ObjectiveTo evaluate the predictive value of critical illness scores for hospital mortality of severe respiratory diseases in respiratory intensive care unit (ICU).MethodsThe clinical data of the patients who needed intensive care and primary diagnosed with respiratory diseases from June, 2001 to Octomber, 2012 were extracted from MIMIC-Ⅲ database. The Acute Physiology Score (APS) Ⅲ, Simplified Acute Physiology Score (SAPS) Ⅱ, Oxford Acute Severity of Illness Score (OASIS), Logistic Organ Dysfunction System (LODS), Systemic Inflammatory Response Syndrome (SIRS) and Sequential Organ Failure Assessment (SOFA) were calculated according to the requirements of each scoring system. ICU mortality was set up as primary outcome and receiver operating characteristic (ROC) analysis was performed to evaluate the predictive performances by comparing the areas under ROC curve (AUC). According to whether they received invasive mechanical ventilation during ICU, the patients were divided into two groups (group A: without invasive mechanical ventilation group; group B: with invasive mechanical ventilation group). The AUCs of six scoring systems were calculated for groups A and B, and the ROC curves were compared independently.ResultsA total of 2988 patients were recruited, male accounted for 49.4%, median age was 67 (55, 79), and ICU mortality was 13.2%. The AUCs of SAPSⅡ, LODS, APSⅢ, OASIS, SOFA and SIRS were 0.73 (0.70, 0.75), 0.71 (0.68, 0.73), 0.69 (0.67, 0.72), 0.69 (0.67, 0.72), 0.67 (0.64, 0.70) and 0.58 (0.56, 0.62). Subgroup analysis showed that in group A, the AUCs of OASIS, SAPSⅡ, LODS, APSⅢ, SOFA and SIRS were 0.81 (0.76, 0.85), 0.80 (0.75, 0.85), 0.77 (0.72, 0.83), 0.75 (0.70, 0.80), 0.73 (0.68, 0.78) and 0.63 (0.56, 0.69) in the prediction of ICU mortality; in group B, the AUCs of SAPSⅡ, APSⅢ, LODS, SOFA, OASIS and SIRS were 0.68 (0.64, 0.71), 0.67 (0.63, 0.70), 0.65 (0.62, 0.69), 0.62 (0.59, 0.66), 0.62 (0.58, 0.65) and 0.57 (0.54, 0.61) in the prediction of ICU mortality. The results of independent ROC curve showed that the AUC differences between groups A and B were statistically significant in terms of OASIS, SAPSⅡ, LODS, APSⅢ and SOFA, but there were no significant differences in SIRS.ConclusionsThe predictive values of six critical illness scores for ICU mortality in respiratory intensive care are low. Lack of ability to predict ICU mortality of patients with invasive mechanical ventilation should hold primary responsibility.
Objective To explore the predictive value of Composite Congestion Score (CCS) in predicting adverse events within 180 days in patients with acute heart failure (AHF) in emergency intensive care unit (EICU). Methods The patients with AHF who were admitted to EICU of Zigong Fourth People’s Hospital between January 1, 2018 and December 31, 2020 were included consecutively. The patients were followed up for 180 days, and were divided into poor prognosis group and good prognosis group according to whether there were adverse events. Logistic regression equation was used to screen independent risk factors for predicting adverse events in patients with AHF within 180 days after leaving EICU. To compare the discrimination, calibration and clinical usefulness of independent risk factors at EICU discharge and the Acute Physiology and Chronic Health Assessment SystemⅡ (APACHEⅡ) score at EICU admission to predict the occurrence of adverse events of AHF. Results A total of 71 patients were included, including 32 patients with good prognosis and 39 patients with poor prognosis. Except for age, APACHEⅡscore at EICU admission and CCS score at EICU discharge (P<0.05), there was no significant difference in other indicators between the two groups (P>0.05). Logistic regression analysis showed that CCS score at EICU discharge [odds ratio (OR)=2.806, 95% confidence interval (CI) (1.428, 5.512), P=0.003], age [OR=1.086, 95%CI (1.017, 1.159), P=0.013] were independent risk factors for predicting death or returning to hospital within 180 days. Among them, the CCS score at EICU discharge combining with age had a positive improvement ability compared with the CCS score at EICU discharge, the age, and the APACHE Ⅱ score at EICU admission. The calibration curves of the four scoring methods for predicting adverse events within 180 days showed that the CCS score at EICU discharge had the highest calibration and the calibration of age was the lowest. The decision curve showed that the clinical usefulness of age, the CCS score at EICU discharge and the CCS score at EICU discharge combining with age was better than the APACHE Ⅱ score at EICU admission. Conclusions The CCS score of patients with AHF at EICU discharge is closely related to adverse events within 180 days. The CCS score is designed based on clinical variables, simple and practical. The combination of age and the CCS score at EICU discharge will further enhance its clinical application value.