Seventeen association measures for observer reliability (Interobserver agreement) are verified and calculation formulas are given in a common scoring system. An empirical comparison of 10 of these measures is made during a number of potential background check results. The effect on frequency, error frequency and percentage and correlation error values are analyzed. The question of what is the “best” measure of the Interobserver agreement is debated with regard to the critical issues to be considered by Vet HC, Terwee CB, Knol DL, Bouter LM. use of agreed and insurance measures. J Clin Epidemiol. 2006;59(10):1033–9. The second study focuses on diaschisis in gliomas, where diaschisis is a brain failure far from a focal brain injury. A consecutive series of 14 patients with glioma, referred by our neurosurgery department suspected of cerebral malignancy (assessed by histopathological results from biopsy and MRI samples from the clinical routine) was subjected to FDG-PET/CT examinations from 2012 to 2015. Patients were followed for 1 year or up to death for the duration of treatment, and FDG-PET/CT scans were performed up to five times: 1) at the beginning (before treatment); 2) Post-exploitation; 3), 4) and 5) followed during chemotherapy or no treatment. Each patient was assigned to one of the two scanners (GE Discovery 690 or 710, GE Healthcare, Milwaukee, WI, USA) and 49 FDG-PET/CT scans were collected. The whole of hemispheric glycolysis (GES) has been studied in the ipsilateral and counterlateral hemisphere in the hemisphere relative to the primary tumour, using a special 3D segmenation software (version 2.1.26, ABX GmbH, Radeberg, Germany). Two inexperienced medical students (observers 1 and 2) and an experienced neuro-PET clinician (observer 3) drew ROIs.
GHGs are defined as the product of segmented metabolic volume and average SUV in this volume (cm3 × g/ml) which includes all voxel in a cerebral hemisphere; Iterative thresholds with 40% SUVmax cutoff. Below, only GHG measurements are used in the ipsilateral hemisphere. Diagram of the application of repeated SUVmax measurements for 30 patients (study 1). (EPS 24 kb) The first point will naturally lead to a link between the VCA and Bland-Altman agreements, which in turn are directly related to the term RC: while the Bland-Altman limit values exceed the average of all the differences between the peer-to-peer measures of 1.96 times the standard deviation of these differences (SD), the RC is equal to 2.77 times the standard deviation (Sw); Half the width of the Bland-Altman boundaries corresponds to the RC in simple settings, because the standard deviation within the subject is then synonymous with standard measurement errors (SEM): despite general expectations that analysts are familiar with the various calculations of reliability statistics (BACB, 1995) and declare IOA in their behavioural assessments (Kelly, 1977; Mudford et al., 2009), competing professional responsibilities may exclude appropriate analysis of IOA in the clinical environment.
