Yüksel Balcı, Anıl Özgür, Sermin Tok Umay, Kaan Esen, Semra Erdoğan, Gülhan Temel

Stiloid Proseslerin İki Farklı Zamanda İki Değerlendirici Tarafından Alınan Üç Boyutlu Bilgisayarlı Tomografi Ölçümlerinin Uyumunun Test Edilmesinde Alternatif Uyum Yaklaşımları

Amaç: Bu çalışmada üç boyutlu bilgisayarlı tomografi (BT) görüntüleri üzerinde iki deneyimli radyolog tarafından alınan stiloid proses (SP) ölçümlerinin değerlendiriciler arası ve değerlendirici içi uyumunu, uyum istatistik yöntemleri olan konkordans korelasyon katsayısı (concordance correlation coefficient—CCC) ve birey uyum katsayısı (coefficient of individual agreement—CIA) metotları kullanılarak değerlendirilmesi amaçlanmıştır. Gereç ve Yöntemler: Haziran 2015—Aralık 2015 döneminden, 68 hastaya ait kontrastlı karotis BT anjiyografi görüntüleri geriye dönük olarak incelendi. SP temporal kemiğe tutunma noktası ve distal ucu arasındaki mesafe, 64 kesitli bir BT cihazı ile gerçekleştirilen üç boyutlu BT görüntülerinde ölçüldü. Radyologlar arasındaki ve radyologların kendi tekrarlı ölçümleri arasındaki uyum, CIA ve CCC uyum indekslerinden yararlanılarak hesaplandı. Bulgular: Uyum istatistikleri olarak CIA kullanıldığında, bireylerin sağ ve sol ölçümlerinde iki radyolog arasında uyum bulunmadı. CCC uyum istatistiği kullanıldığında ise iki radyoloğun tekrarlı ölçümleri arasında mükemmel bir uyum bulundu. Tartışma ve Sonuç: Her bir değerlendiriciden iki tekrarlı ölçüm alındığında, denekler arası ve denekler içi varyans değerlerinin dikkate alınması gerekmektedir. Eğer bu değerler birbirinden oldukça büyük farklılıklar gösteriyorsa, CCCtotal değerleri gerçek değerden daha büyük bir değer olarak hesaplanacak ve elde edilen bu sonuçlar araştırmacıları yanıltacaktır. Böyle bir durumla karşılaşıldığında, CIA istatistiğinin çok daha kararlı olduğu ve denekler arası ve denekler içi varyans değerlerinin, her bir değerlendiricinin iki tekrarlı ölçümleri arasındaki farklılıklar üzerinden hesaplanması gerektiği araştırmacılara önerilir. Böylece, 2 2B W oranı mutlakadikkate alınmalıdır.

Alternative Agreement Approaches in Testing the Agreement of Three-dimensional Computerized Tomography Measurements of Styloid Processes by Two Observers at Two Different Times*

Aim: This study aimed to evaluate the intraobserver and interobserver agreement of styloid process(SP) measurements on three-dimensional (3D) computerized tomography (CT) images between twoexperienced radiologists by using the concordance correlation coefficient (CCC) and coefficient ofindividual agreement (CIA) agreement statistical methods.Materials and Methods: Contrast-enhanced carotid CT angiography images of 68 patients, performed between June 2015 and December 2015, were evaluated retrospectively. The length betweenthe attachment point of the SP to the temporal bone and the distal end was measured on 3D CTimages performed with a 64-slice CT scanner. Both the agreements between the two radiologistsand the replicated measurements of each radiologist were calculated with the help of CIA and CCCagreement indexes.Results: When CIA was used for agreement statistics, the two radiologists disagreed in the right andleft measurements of the individual. When the CCC agreement statistic was used, there was a perfectagreement between the measurements of the two radiologists.Discussion and Conclusion: The variance values of between-subject and within-subject should betaken into consideration for each observer in cases of two replicated measurements. If these valuesshow very large differences from each other, CCCtotal values may have been calculated larger thanthe real value. However, the CIA statistics value is more stable and when such a case is encountered,researchers are advised that between-subject and within-subject variances should be calculated dueto the differences between the two replicated measurements of each observer. Therefore, the 2 2B Wrate should absolutely be taken into consideration.

PDF

___

1. Gao J, Pan Y, Haber M. Assessment of observer agreement for matched repeated binary measurements. Computational Statistics & Data Analysis. 2012;56(5):1052–60.
2. Barnhart HX, Haber MJ, Lin LI. An overview on assessing agreement with continuous measurements. J Biopharm Stat. 2007;17(4):529–69.
3. Barnhart HX, Lokhnygina Y, Kosinski AS, Haber M. Comparison of concordance correlation coefficient and coefficient of individual agreement in assessing agreement. J Biopharma Stat. 2007;17(4):721–38.
4. Bland J, Altman D. Measuring agreement in method comparison studies. Stat Methods Med Res. 1999;8(2):135–60.
5. Bland J, Altman D. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17(4):571–82.
6. Başekim CC, Mutlu H, Güngör A, Silit E, Pekkafali Z, Kutlay M, et al. Evaluation of styloid process by three-dimensional computed tomography. Eur Radiol 2005;15(1):134–9.
7. Kosar MI, Atalar MH, Sabancioğullari V, Tetiker H, Erdil FH, Cimen M, et al. Evaluation of the length and angulation of the styloid process in the patient with prediagnosis of Eagle syndrome. Folia Morphol (Warsz). 2011;70(4):295–9.
8. Oztunç H, Evlice B, Tatli U, Evlice A. Cone-beam computed tomographic evaluation of styloid process: a retrospective study of 208 patients with orofacial pain. Head Face Med. 2014;10(5):1–7.
9. Okur A, Ozkırış M, Serin HI, Gencer ZK, Karaçavuş S, Karaca L, et al. Is there a relationship between symptoms of patients and tomographic characteristics of styloid process? Surg Radiol Anat. 2014;36(7):627–32.
10. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. International Journal of Nursing Studies. 2010;47(8):931–6.
11. Van Stralen KJ, Dekker FW, Zoccali C, Jager KJ. Measuring agreement, more complicated than it seems. Nephron Clin Pract. 2012;120(3):c162–7.
12. Lin, L, Hedayet AS, Wu W. A unified model for continuous and categorical data. In: Statistical Tools for Measuring Agreement, 1. ed. New York: Springer; 2012.
13. Vangeneugden T, Laenen A, Geys H, Renard D, Molenberghs G. Applying linear mixed model to estimate reliability in clinical trials with repeated measurement. Controlled Clinical Trials. 2004;25:13–30.
14. Molenbergs G, Vangeneugden T, Laenen A. Estimating reliability and generalizability from hierarchical biomedical data. J Biopharm Stat. 2007;17(4):595–627.
15. King TS, Chinchilli VM, Carrasco JL, Wang K. A class of repeated measures concordance correlation coefficients. J Biopharm Stat. 2007;17(4):653–72.
16. Quiroz J. Assessment of equivalence using a concordance correlation coefficient in repeated measurement design. J Biopharm Stat. 2005;15:913–28.
17. Chen CC, Barnhart HX. Assessing agreement with intraclass correlation coefficient and concordance correlation coefficient for data with repeated measures. Comput Statist Data Anal. 2013;60:132–45.
18. Haber M, Barnhart HX. A general approach to evaluating agreement between two observers or methods of measurement. Stat Methods Med Res. 2008;17(2):151–69.
19. Haber M, Barnhart HX, Song J, Gruden J. Observer variability: a new approach in evaluating interobserver agreement. J Data Sci. 2005;3:69–83.
20. Lin L, Hedayat AS, Wu W. A unified approach for assessing agreement for continuous and categorical data. J Biopharm Stat. 2007;17(4):629–52.
21. Vieira EM, Guedes OA, Morais SD, Musis CR, Albuquerque PA, Borges AH. Prevalence of elongated styloid process in a Central Brazilian population. J Clin Diagn Res. 2015;9(9):ZC90–2.
22. Jung T, Tschernitschek H, Hippen H, Schneider B, Borchers L. Elongated styloid process: when is it really elongated? Dentomaxillofac Radiol. 2004;33(2):119–24.
23. Scaf G, Freitas DQ, Loffredo Lde C. Diagnostic reproducibility of the elongated styloid process. J Appl Oral Sci. 2003;11(2):120–4.
24. Nayak DR, Pujary K, Aggarwal M, Punnoose SE, Chaly VA. Role of three-dimensional computed tomography reconstruction in the management of elongated styloid process: a preliminary study. J Laryngol Otol. 2007;121(4):349–53.
25. Erdoğan S, Orekici Temel G, Selvi H, Ersöz Kaya I. An alternative method used in evaluating agreement among repeat measurement by two raters in education. Educational Sciences: Theory & Practice. 2017;17:217–36.
26. Pan Y, Haber M, Gao J, Barnhart HX. A new permutation-based method for assessing agreement between two observers making replicated quantitative readings. Stat Med. 2012;31(20):2249–61.