Estudi de la fiabilitat de la escintigrafía miocardial 123 I - MIBG en pacients caucàsics amb malaltia de Parkinson

Abstract

[cat] El propòsit d’aquest treball és l’estudi de les mesures preses per un conjunt de metges per a avaluar la fiabilitat de la tècnica d’imatge 123 I − MIBG i concloure si és reproduı̈ble. Els ı́ndexs estimats i interpretats per a l’obtenció dels resultats són els ı́ndex de fiabilitat o Intraclass Corre- lation Coefficient (ICC) i l’ı́ndex de concordança o Concordance Correlation Coefficient (CCC), ambdues mesures basades en l’estudi de la variabilitat de les dades. La mostra d’estudi consta d’un total de 27 individus i està conformada per un conjunt de pacients patològics i un altre de pacients control. Sobre aquests i a partir de la tècnica d’imatges s’extreuen mesures sobre un ı́ndex en dos moments temporals diferents i seràn les variables d’estudi al llarg del treball. Un cop finalitzat l’anàlisi de la variabilitat, es comparen els resultats obtinguts amb un anàlisi que es va realitzar sobre les mateixes dades i amb el mateix objectiu però fent ús d’una metodologia subòptima. Finalment, s’avaluen les conseqüències d’una metodologia inadequada.

[eng] The purpose of this paper is to study the measurements taken by a group of doctors in order to analyze the reliability of the image technique 123 I − MIBG and conclude whether if it is a reprodu- cible methodology. The estimated indices to obtain results are the Intraclass Correlation Coefficient (ICC) and the Concordance Correlation Coefficient (CCC), both of them based on the study of the variability within and between clusters of the data. The sample of study is composed of 27 cauca- sian individuals and can be segregated in diagnosed group and control group. Measurements are taken from the images obtained using the technique of study in two different moments, 15 minutes and 4 hours post-injection. This information is taken by three doctors, one of them realizes the same measurement 3 times. Once the variability study is finished, the results of the investigation are compared with those of an original study made years ago using a different methodology, therefore reporting different insights. Lastly, the negative effects of the applycation of a suboptimal methodology are exposed. To carry out the analysis, the statistical software ”R”and the packages related to modelling and visualization of data will be used.The purpose of this paper is to study the measurements taken by a group of doctors in order to analyze the reliability of the image technique 123 I − MIBG and conclude whether if it is a reprodu- cible methodology. The estimated indices to obtain results are the Intraclass Correlation Coefficient (ICC) and the Concordance Correlation Coefficient (CCC), both of them based on the study of the variability within and between clusters of the data. The sample of study is composed of 27 cauca- sian individuals and can be segregated in diagnosed group and control group. Measurements are taken from the images obtained using the technique of study in two different moments, 15 minutes and 4 hours post-injection. This information is taken by three doctors, one of them realizes the same measurement 3 times. Once the variability study is finished, the results of the investigation are compared with those of an original study made years ago using a different methodology, therefore reporting different insights. Lastly, the negative effects of the applycation of a suboptimal methodology are exposed. To carry out the analysis, the statistical software ”R”and the packages related to modelling and visualization of data will be used.