Wos İndeksli Açık & Kapalı Erişimli Yayınlar
Permanent URI for this communityhttps://hdl.handle.net/11727/10751
Browse
6 results
Search Results
Item Searching For The Urine Osmolality Surrogate: An Automated Machine Learning Approach(2022) Topcu, Deniz Ilhan; Bayraktar, Nilufer; https://orcid.org/0000-0002-1219-6368; https://orcid.org/0000-0002-7886-3688; 000819864400001; E-3717-2019; Y-8758-2018Objectives Automated machine learning (AutoML) tools can help clinical laboratory professionals to develop machine learning models. The objective of this study was to develop a novel formula for the estimation of urine osmolality using an AutoML tool and to determine the efficiency of AutoML tools in a clinical laboratory setting. Methods Three hundred routine urinalysis samples were used for reference osmolality and urine clinical chemistry analysis. The H2O AutoML engine completed the machine learning development steps with minimum human intervention. Four feature groups were created, which include different urinalysis measurements according to the Boruta feature selection algorithm. Method comparison statistics including Spearman correlation, Passing-Bablok regression analysis were performed, and Bland Altman plots were created to compare model predictions with the reference method. The minimum allowable bias (24.17%) from biological variation data was used as the limit of agreement. Results The AutoML engine developed a total of 183 ML models. Conductivity and specific gravity had the highest variable importance. Models that include conductivity, specific gravity, and other urinalysis parameters had the highest R-2 (0.70-0.83), and 70-84% of results were within the limit of agreement. Conclusions Combining urinary conductivity with other urinalysis parameters using validated machine learning models can yield a promising surrogate. Additionally, AutoML tools facilitate the machine learning development cycle and should be considered for developing ML models in clinical laboratories.Item Serum growth differentiation factor-15 analysis as a malnutrition marker in hemodialysis patients(2021) Turgut, Didem; Topcu, Deniz Ilhan; Alperen, Cemile Cansu; Baskin, Esra; 0000-0002-1219-6368; 0000-0001-7474-5927; 34247467; E-3717-2019Background/aim: Growth differentiation factor (GDF)-15 is related to inflammation and mortality in many conditions. We aimed to determine if an elevated serum GDF-15 level is related to nutritional status of patients on hemodialysis (HD) and mortality. Materials and methods: Routine HD patients (n = 158) were included in the study and followed for 18 months. Some malnutrition/ inflammation scoring indexes (malnutrition/inflammation score (MIS), controlling nutritional status (CONUT) score, and prognostic nutritional index (PNI)), biochemical parameters, and GDF-15 were used to build Cox regression multivariate models to study the association with mortality. Results: Among the patients, 90 (57 %) had a high MIS ( _8), which associates with worse status. The serum GDF-15 level was higher in the same group (p = 0.003). The serum GDF-15 level differentiated malnutrition/inflammation according to the MIS (p = 0.031). Age, GDF15, and C-reactive protein (CRP) were significantly associated with higher all-cause mortality risk. Patients with both age and GDF-15 above the mean had a hazard ratio of 2.76 (p = 0.006) when compared with those both < mean. Conclusion: In HD patients, the GDF-15 level is increased in worse nutritional status. Beyond the MIS, age, GDF-15 and CRP would be used together to estimate the worse clinical outcome in these patients.Item Detection of COVID-19 by Machine Learning Using Routine Laboratory Tests(2021) Cubukcu, Hikmet Can; Topcu, Deniz Ilhan; Bayraktar, Nilufer; Gulsen, Murat; Sari, Nuran; Arslan, Ayse Hande; 0000-0002-1219-6368; 0000-0002-7886-3688; 34791032; E-3717-2019; Y-8758-2018Objectives The present study aimed to develop a clinical decision support tool to assist coronavirus disease 2019 (COVID-19) diagnoses with machine learning (ML) models using routine laboratory test results. Methods We developed ML models using laboratory data (n = 1,391) composed of six clinical chemistry (CC) results, 14 CBC parameter results, and results of a severe acute respiratory syndrome coronavirus 2 real-time reverse transcription-polymerase chain reaction as a gold standard method. Four ML algorithms, including random forest (RF), gradient boosting (XGBoost), support vector machine (SVM), and logistic regression, were used to build eight ML models using CBC and a combination of CC and CBC parameters. Performance evaluation was conducted on the test data set and external validation data set from Brazil. Results The accuracy values of all models ranged from 74% to 91%. The RF model trained from CC and CBC analytes showed the best performance on the present study's data set (accuracy, 85.3%; sensitivity, 79.6%; specificity, 91.2%). The RF model trained from only CBC parameters detected COVID-19 cases with 82.8% accuracy. The best performance on the external validation data set belonged to the SVM model trained from CC and CBC parameters (accuracy, 91.18%; sensitivity, 100%; specificity, 84.21%). Conclusions ML models presented in this study can be used as clinical decision support tools to contribute to physicians' clinical judgment for COVID-19 diagnoses.Item The effect of jet-lag on serum concentrations of thyroid stimulating hormone and prolactin: A case report(2020) Gungoren, Merve Sibel; Topcu, Deniz Ilhan; Zungun, Cevdet; 0000-0002-1219-6368; 32063733; E-3717-2019Introduction: This case report is about the importance of sleeping status for analysis of thyroid hormone stimulating hormone (TSH) and prolactin (PRL) which arose from discordant results of a patient who was referred for serum TSH and PRL testing within 12-hour period after an intercontinental flight. Case description: An adult male patient was admitted to our laboratory for serum TSH and PRL tests and came back questioning the accuracy of his previous results. Further investigations: A new analysis with a new sample was offered. His new results were not consistent with his previous results. What happened: It was revealed that the night before the first sampling, he travelled back to Turkey from The United States of America and came to testing within 12 hours after the arrival. Discussion: Sleeping status is one of the factors that can affect laboratory results. Intercontinental flights causing jet-lag can alter the secretions of TSH and PRL which are predominantly modulated by thyrotropin-releasing hormone (TRH). Main lesson: Travel history and sleeping status are important factors to be evaluated prior sampling for hormone analysis. Patients must be informed about the importance of sampling timing.Item Peripheral Block Education and Level of Competency: A Survey of Turkish Anesthesiologists(2020) Selvi, Onur; Tulgar, Serkan; Senturk, Ozgur; Tas, Zafer; Kose, Halil Cihan; Topcu, Deniz Ilhan; Ozer, ZelihaBACKGROUND/AIMS Presently, in Turkey, there is no work being conducted on the period of regional anesthesia education, and there are no statistics available on the frequency with the administration of basic extremity blocks. The present survey was conducted on anesthesia doctors throughout Turkey to explore the personal knowledge and competency of the participants. Simultaneously, this survey aimed to evaluate the information sources for regional anesthesia training in Turkey. MATERIAL and METHODS The present study surveyed 377 anesthesia doctors in March 2017 through a questionnaire formulated on the Delphi platform. The voluntary participants were recruited from professional organizations, hospital portals, and Turkey's anesthesia departments. Descriptive analyses were conducted for statistical evaluation. RESULTS Of the total participants, 3.2% were professors, 3.2% were associate professor doctors, 7.4% were assistant professors, 64.5% were specialist doctors, and 21.8% were residents. In the segment concerning skill and information level self-evaluation for blocks, the most commonly performed act was infraclavicular block (26.8%). With regard to their "information source," 40% of the participants chose "from my colleagues" for upper extremity blocks. The specialist training was the least popular information source for ultrasound-guided interscalene block, while the digital visual information sources were often consulted for the same. CONCLUSION Various sources have been identified as the source of information as much as the specialty training, more so in some blocks. A significant majority of the participants avoided block applications. In the future, it would be useful to collect data on a more comprehensive national scale to overcome the limitations of the present issue.Item The Impact of Serum 25-Hydroxyvitamin D3 Levels on Allergic Rhinitis(2019) Coban, Kubra; Oz, Isilay; Topcu, Deniz Ilhan; Aydin, Erdinc; 0000-0002-1219-6368; 31569970; E-3717-2019We aimed to clarify the relation between allergic rhinitis and the serum levels of 25-hydroxivitamin D in the adult population. The study group consisted of 86 patients with allergic rhinitis who were diagnosed with the help of history of allergy, positive signs for allergy, blood samples, and positive skin prick tests; while the control group included 43 age- and sex-matched healthy volunteers with negative skin prick tests. The demographic data, medical history, findings in the physical examinations, serum levels of total immunoglobulin E (IgE) and 25-hydroxyvitamin D, and skin prick test results of the groups were noted. A total of 129 patients fulfilling the necessary criteria were enrolled. The median serum 25-hydroxyvitamin D levels in the study group were significantly lower compared to the control group (P= .014). In the study group, median serum vitamin D levels were significantly higher in men, compared to women (P= .03). There was a significant negative correlation between IgE and vitamin D levels in the allergic rhinitis group (P= .028,r= -0.246). This study showed that patients with allergic rhinitis might be more vulnerable to have lower serum levels of vitamin D. Thus, vitamin D supplementation as an adjunctive therapy may be considered in those patients.