Yazar "Kurt, Imran" seçeneğine göre listele
Listeleniyor 1 - 8 / 8
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Comparing performances of logistic regression, classification and regression tree, and neural networks for predicting coronary artery disease(Pergamon-Elsevier Science Ltd, 2008) Kurt, Imran; Ture, Mevlut; Kurum, A. TurhanIn this study, performances of classification techniques were compared in order to predict the presence of coronary artery disease (CAD). A retrospective analysis was performed in 1245 subjects (865 presence of CAT) and 380 absence of CAD). We compared performances of logistic regression (LR), classification and regression tree (CART), multi-layer perceptron (MLP), radial basis function (RBF), and self-organizing feature maps (SOFM). Predictor variables were age, sex, family history of CAD, smoking status, diabetes mellitus, systemic hypertension, hypercholesterolemia, and body mass index (BMI). Performances of classification techniques were compared using ROC curve, Hierarchical Cluster Analysis (HCA), and Multidimensional Scaling (MDS). Areas under the ROC curves are 0.783, 0.753, 0.745, 0.721, and 0.675, respectively for MLP, LR, CART, RBF, and SOFM. MLP was found the best technique to predict presence of CAD in this data set, given its good classificatory performance. MLP, CART, LR, and RBF performed better than SOFM in predicting CAD in according to HCA and MDS. (c) 2006 Elsevier Ltd. All rights reserved.Öğe Comparison of dimension reduction methods using patient satisfaction data(Pergamon-Elsevier Science Ltd, 2007) Ture, Mevlut; Kurt, Imran; Akturk, ZekeriyaIn this study, we compared classical principal components analysis (PCA), generalized principal components analysis (GPCA), linear principal components analysis using neural networks (PCA-NN), and non-linear principal components analysis using neural networks (NLPCA-NN). Data were extracted from the patient satisfaction query with regard to the satisfaction of patients from hospital staff, which was applied in 2005 at the outpatient clinics of Trakya University Medical Faculty. We found that percentages of explained variance of principal components from PCA-NN and NLPCA-NN were highest for doctor, nurse, radiology technician, laboratory technician, and other staff using a patient satisfaction data set. Results show that methods using NN which have higher percentages of explained variances than classical methods could be used for dimension reduction. (C) 2005 Elsevier Ltd. All rights reserved.Öğe Contrast-enhanced MR 3D angiography in the assessment of brain AVMs(Elsevier Ireland Ltd, 2006) Unlu, Ercument; Temizoz, Osman; Albayram, Sait; Genchellac, Hakan; Hamamcioglu, M. Kemal; Kurt, Imran; Demir, M. KemalBackground and purpose: Digital subtraction angiography (DSA) is the current reference standard for the diagnosis, assessment, and management of brain arteriovenous malformations (AVMs). The purpose of this study was to compare the diagnostic utility of three-dimensional (3D) time-off-light (TOF) magnetic resonance angiography (MRA) and contrast-enhanced 3D MRA in patients with intracranial arteriovenous malformations (AVMs) in different sizes and locations. The AVM diagnosis was proved via DSA and almost half of the patients had also hematoma. Materials and methods: Two radiologists, experienced on neurovascular imaging and independent from each other, retrospectively reviewed two MRA techniques and DSA with regard to the assessment of feeding arteries, AVM nidus, and venous drainage patterns on 20 patients with 23 examinations by scoring system. Disagreements were resolved by consensus. Results: An excellent agreement between contrast-enhanced MRA and DSA was found in order to assess the numbers of arterial feeders and draining veins (Spearman r=0.913, P<0.001). The average scores in contrast-enhanced MRA for feeders, nidi, and drainers were respectively 2.26, 2.69, and 2.48, while in TOF-MRA they are 1.96, 1.35, and 0.89, respectively. Conclusion: Compared to TOF-MRA, 3D contrast-enhanced MRA is useful for visualization by subtraction technique of malformation components presented by hematoma or by haem product. On the other hand, for the cases presented by slow or complex flow that is especially in around or nidi or around the venous portion is also advantageous because of the independence from flow-related enhancement. Therapeutic effects were clearly demonstrated in three follow-up patients. A major limitation of this technique is the low spatial resolution. Since there is such a limitation, arterial feeder of a case with micro-AVM is not detected by contrast-enhanced MRA and nidus for the same case was observed retrospectively. In this respect, we believe that 3D contrast-enhanced MRA is a less invasive and inexpensive angiographic tool, but not a safe substitute for DSA. Yet, it can be a beneficial supplement to DSA in patients with cerebral AVMs at both initial diagnosis and at follow-up processes after therapy. (C) 2006 Elsevier Ireland Ltd. All rights reserved.Öğe Efficiency of Cervical Sonography for the Determination of Preterm Birth in Singleton and Twin Pregnancies After 25 Weeks' Gestation(Galenos Yayincilik, 2005) Sayin, N. Cenk; Varol, Fusun G.; Yilmaz, Ozlem; Kurt, ImranObjective: To evaluate the efficiency of cervical sonography in determination of preterm birth in singleton and twin pregnancies. Materials and Methods: A total of 78 women (56 with singleton and 22 with twin pregnancies) with preterm labor who were admitted to our department between 25 and 36 weeks' gestation were enrolled in the study. Cervical length and the presence of cervical funneling were evaluated by transvaginal sonographic scan. For singleton and twin pregnancies reaching >= 37 and >= 34 weeks' gestation respectively, were accepted as good perinatal outcome. The cut-off values for cervical length were set at <35 mm for singleton and <25 mm for twin pregnancies. Results: Sonographic evaluation and delivery weeks for singleton and twin pregnancies were 33 (25-36) and 36 (26-40) weeks, and 31 (25-33) and 34 (31-36) weeks, respectively. A cervical length <35 mm for singleton pregnancies, <25 mm for twins have 67.9% and 57.1% sensitivity, 46.4% and 86.7% specificity, 55.9% and 66.7% positive predictive and 59.1% and 81.3% negative predictive values for the determination of preterm birth, respectively. The presence of cervical funneling for singleton and twin pregnancies determined preterm birth with sensitivity of 83.3% and 55.6%, specificity of 50% and 84.6%, positive predictive value of 44.1% and 71.4%, negative predictive value of 86.4% and 73.3%, respectively. Conclusion: Cervical sonography after 25 weeks' gestation can determine preterm birth with high sensitivity and negative predictive value in twin pregnancies presented with preterm labor. However, for singleton pregnancies the presence of cervical funneling has high sensitivity and negative predictive values for the determination of preterm birth.Öğe High C-reactive protein and low cholesterol levels are prognostic markers of survival in severe sepsis(Elsevier Science Inc, 2007) Memis, Dilek; Gursoy, Olcay; Tasdogan, Muhittin; Sut, Necdet; Kurt, Imran; Ture, Mevlut; Karamanlioglu, BeyhanStudy Objective: To evaluate serum C-reactive protein and cholesterol as a prognostic factor for Survival in patients with severe sepsis. Design: Prospective study. Setting: University hospital. Patients: The study population consisted of 96 patients (age range, 18-75 years; median, 56 years; men/women ratio, 40:56) in whom severe sepsis was diagnosed. Interventions: Patients' serum levels of C-reactive protein and cholesterol were measured upon admission to an intensive care unit, two days later, and on the day of discharge from the intensive care unit or on the day of death. Measurements and Main Results: Cholesterol levels were significantly lower among the nonsurviving patients (day 1, 92.2 +/- 25.1 mg/dL; day 2, 92.1 +/- 21.7 mg/dL; death/discharge day, 92.2 +/- 21.7 mg/dL) than surviving patients (day 1, 175.1 +/- 38.6 mg/dL [P < 0.001]; day 2, 173.0 +/- 39.3 mg/dL [P < 0.001]; death/discharge day, 171.8 +/- 39.6 mg/dL [P = 0.010]). Median C-reactive protein levels were significantly higher among the nonsurvivors (day 1, 32 mg/dL [range, 20.5-64.5 rng/dL]; day 2, 33 mg/dL [range, 22-74.5 mg/dL]; death/discharge day, 30 rng/dL [range, 22-57 mg/dL]) than survivors (day 1, 10 rng/dL [range, 6-14 mg/dL]; day 2, 9 mg/dL [range, 5-10 mg/dL]; death/discharge day, 6 mg/dL [range, 3-9 mg/dL]; P < 0.001). Conclusion: Serum C-reactive protein and cholesterol are a predictor of survival in patients with severe sepsis. Low cholesterol and high C-reactive protein levels appear as a valuable tool for individual risk assessment in severe sepsis patients and for stratification of high-risk patients in future intervention trials. (C) 2007 Elsevier Inc. All rights reserved.Öğe Perinatal Outcome in Women Screened for Gestational Diabetes Mellitus With Normal or With One Elevated Glucose Tolerance Test Value(Galenos Yayincilik, 2007) Sayin, N. Cenk; Varol, Fusun G.; Duran, Ridvan; Acunas, Betul; Kurt, ImranObjective: To investigate the perinatal outcome of women screened for gestational diabetes mellitus (GDM) who had abnormal screening and normal oral glucose tolerance test (OGTT) results or women with a single abnormal test value in OGTT. Materials and Methods: We screened 576 pregnant women with a 50 gr oral glucose challenge test between the 24th and 26th weeks' of gestation. Women with abnormal diabetes screening test results were followed up with standard OGTT with a 100 gr oral glucose load. According to the results, 360 women were with a normal screening test (Group 1), 87 with a positive screening and all normal test values on OGTT (Group 2), 50 with a positive screening but a single abnormal test value on OGTT (gestational impaired glucose tolerance, Group 3) and 79 were diagnosed as with GDM (Group 4). Diet or diet plus insulin therapy was initiated to patients in Groups 3 and 4 as indicated. We compared perinatal outcome between these four groups. Statistical data were calculated with post hoc multiple comparison, Kruskal-Wallis and c 2 tests for comparison of means, medians or for nominal variables. Results: Women with GDM had significantly higher glycosylated hemoglobin levels, large for gestational age infants, macrosomia, neonatal jaundice, neonatal intensive care unit (NICU) admission and neonatal mortality rates but lower gestational age at delivery and 1-minute Apgar scores compared to controls. Also, there were significant differences in neonatal jaundice and NICU admission rates between Groups 1 and 3, or between 2 and 3. There were significantly more macrosomic babies in Groups 3 and 4, compared to Group 2. Discussion: Women with abnormal diabetes screening tests and negative OGTT or with gestational impaired glucose tolerance seem not to be prone to develop severe adverse perinatal outcome.Öğe Prevalence of asthma among preschool children in Edirne, Turkey(Allergy Immunol Soc Thailand, 2007) Yolsal, Guner Emel; Yazicioglu, Mehtap; Ture, Mevlut; Kurt, ImranAllergic diseases generally begin early during childhood, but a late diagnosis is common. This study aimed to evaluate the prevalence of asthma and asthma-related symptoms among kindergarten children in Edirne, Turkey. 873 subjects based on a modified ISAAC questionnaire were included. The prevalence of 'wheezing ever' and 'wheezing during the previous year' was 23.3% and 8.6%, respectively. Prior physician diagnoses existed for 36 of 873 (4.1%) children. The prevalence of children undiagnosed with asthma, but reporting asthma-related symptoms was 3.1%, 51.9% of which had previous beta-agonist prescriptions. However, none of these children received inhaled anti-inflammatory medications. In conclusion, it was found that a large population of preschool children had undiagnosed respiratory symptoms suggestive of asthma. Conducting simple surveys of young children is particularly important, as identification of asthma early in the disease course will facilitate effective prevention and treatment.Öğe Using Kaplan-Meier analysis together with decision tree methods (C&RT, CHAID, QUEST, C4.5 and ID3) in determining recurrence-free survival of breast cancer patients(Pergamon-Elsevier Science Ltd, 2009) Ture, Mevlut; Tokatli, Fusun; Kurt, ImranCurrent evidence supports a clear association between clinical and pathologic factors and recurrence-free survival (RFS) in breast cancer patients. The Cox regression model is the most common tool for investigating simultaneously the influence of several factors oil the survival time of patients. But it gives no estimate of the degree of separation of the different Subgroups. We propose to analyze different decision tree methods (C&RT, CHAID. QUEST, C4.5 and ID3) and use them additionally to the well-known Kaplan-Meier estimates to investigate the predictive power of these methods. Five hundred patients were included to the study. Two hundred and seventy-nine of them had complete data for prognostic factors and median follow-up is about 40.5 months. First, decision tree methods were analyzed for prognostic factors. Then, according to multidimensional scaling method C4.5 (error rate 0.2258 for training set and 0.3259 for cross-validation) performed slightly better than other methods in predicting risk factors for recurrence. Tumor size, age of menarche, hormonal therapy. histological grade and axillary nodal Status arc found that in important risk factors for the recurrence. Eight terminal nodes were found and stratified by Kaplan-Meier survival curves. Larger tumor size (>= 4.4 cm) and receiving no hormonal therapy in a small subgroup of patients were associated with worse prognosis. The five-year RFS is 71.3% in the whole patient population. The sensitivity, specificity and predictive rates calculated by C4.5 method were found 43.8%, 91% and 77.4% respectively. In this study, C4.5 showed a better degree of separation. As a result, we recommend 10 use decision tree methods together with Kaplan-Meier analysis to determine risk factors and effect of this factors oil survival. (C) 2008 Elsevier Ltd. All rights reserved.
