The use of radiomics and machine learning for the differentiation of chondrosarcoma from enchondroma.

PURPOSE: To construct and compare machine learning models for differentiating chondrosarcoma from enchondroma using radiomic features from T1 and fat suppressed Proton density (PD) magnetic resonance imaging (MRI). METHODS: Eighty-eight patients (57 with enchondroma, 31 with chondrosarcoma) were retrospectively included. Histogram matching and N4ITK MRI bias correction filters were applied. An experienced musculoskeletal radiologist and a senior resident in radiology performed manual segmentation. Voxel sizes were resampled. Laplacian of Gaussian filter and wavelet-based features were used. One thousand eight hundred eighty-eight features were obtained for each patient, with 944 from T1 and 944 from PD images. Sixty-four unstable features were removed. Seven machine learning models were used for classification. RESULTS: Classification with all features showed neural network was the best model for both readers' datasets with area under the curve (AUC), classification accuracy (CA), and F1 score of 0.979, 0.984; 0.920, 0.932; and 0.889, 0.903, respectively. Four features, including one common to both readers, were selected using fast correlation based filter. The best performing models with selected features were gradient boosting for Fatih Erdem's dataset and neural network for Gülen Demirpolat's dataset with AUC, CA, and F1 score of 0.990, 0.979; 0.943, 0.955; 0.921, 0.933, respectively. Neural Network was the second-best model for FE's dataset based on AUC (0.984). CONCLUSION: Using pathology as a gold standard, this study defined and compared seven well-performing models to distinguish enchondromas from chondrosarcomas and provided radiomic feature stability and reproducibility among the readers.

Planned Obsolescence in the Context of a Holistic Legal Sphere and the Circular Economy.

Businesses may take advantage of rapid technological developments to increase sales by designing products with a short lifespan and encouraging consumers to buy a replacement more quickly than they otherwise might have to, which is commonly known as planned obsolescence. While employing a holistic approach and exploring planned obsolescence from three different angles-the demand side, supply side and environmental side-the article argues that the current measures in the fields of unfair competition and consumer protection law, competition law and environmental law are quite insufficient to deal with planned obsolescence. Therefore, there is a need for an EU measure outlawing planned obsolescence in the context of the circular economy.

[Epidemiological and microbiological evaluation of nosocomial infections caused by Candida species]. / Candida Türlerine Bagli Nozokomiyal Enfeksiyonlarin Epidemiyolojik ve Mikrobiyolojik Açidan Degerlendirilmesi.

Nosocomial infections due to candida species are increasingly gaining importance. In this study, we aimed to determine the risk factors associated with nosocomial candida infections and also to determine the species and antifungal sensitivity of candida strains. Candida spp. strains isolated from the clinical specimens of 92 patients who were hospitalized in Ankara Research and Training Hospital between May 2008 and September 2009 were prospectively included in the study. The demographic features, clinical and laboratory findings of the patients were recorded. Identification and antifungal sensitivities of the candida species were determined by VITEC 2 Compact System (BioMeriéux, France). Invasive candida infection was determined in 79 (85.9%) and candida colonization was found in 13 (14.1%) of the patients. Types of candida infections were as follows; urinary tract infection in 57 (72.1%), candidemia in 26 (32.9%), surgical site infection in 6 (7.5%) and intraabdominal infection in 3 (%3.8) patients. According to the univariate analysis, being female gender, duration and rate of hospitalization in the intensive care unit, use of linezolid, prolonged duration of urinary catheterization and total parenteral nutrition and the duration of central venous catheterization were found significantly higher in the urinary tract infection group. In candidemia group, carbapenem use, rates and periods of central venous catheterization and total parenteral nutrition were detected higher which were statistically significant. In multivariate analysis for urinary tract infections female gender (Odds ratio: 3; 95% CI: 1.25-7.17) and prolonged stay (p= 0.031) in intensive care unit; for candidemia use of total parenteral nutrition (OR: 16.25; 95% CI: 2.07-127.35) were found as independent risk factors. A total of 114 Candida spp. Strains were isolated from the clinical specimens. The most common isolated species were C.albicans (n= 62, 54.4%), C.glabrata (n= 16, 14%) and C.tropicalis (n= 13, 11.4%). While the resistance rates of Candida spp. To fluconazole and flucytosine were 2.6% and 1.7% respectively, there was no resistance to voriconazole. The rate of isolates with amphotericin B MIC values > 1 µg/ml was 4.4%. Consequently, the restricted use of central venous catheter and shortening the length of stay in intensive care unit may prevent the development of invasive candida infections. The local surveillance of antifungal resistance in candida strains might aid to the decision of empirical antifungal therapy.