Publication: Anonimización de bases de datos médicas
Loading...
Official URL
Full text at PDC
Publication Date
2016
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
La publicación de los resultados de programas de screening (exámenes médicos orientados a un grupo genérico de la población) es de sumo interés para la comunidad científica. A pesar de eliminar datos personales como DNI, nombre, etc. el resto de la información, los llamados cuasiidentificadores (código postal, género, edad, profesión, o información de la cita médica incluyendo el centro y la hora) pueden ser utilizados para desvelar la identidad de los participantes en el programa. En particular, la información de la cita médica puede resultar comprometedora si, ya sea intencionadamente o por casualidad, se descubre que una persona determinada ha acudido a un determinado centro en una fecha concreta. Bastaría entonces con consultar los resultados del test para saber si el individuo cuyos cuasi-identificadores conocemos padece la enfermedad. El objetivo de este trabajo es programar la asignación de citas, de manera que se aumente el nivel de anonimato de las bases de datos finales. Para ello, se pretende que personas con características comunes (edad, etc.) acudan a la misma cita (hora y centro), dificultando la identificación de los resultados médicos de un individuo aunque se conozcan sus datos personales, ya que se encontrarán con varias personas con los que comparte los mismos cuasi-identificadores.
The publication of screening programmes’ results (examination or testing of a group of individuals to separate those who are well from those who have an undiagnosed disease or defect or who are at high risk [11]) is an important factor for the scientific community. In spite of deleting personal data such as ID, name, etc. the rest of the information, which is called quasi-identifiers (P.O. Box, gender, age, profession or the information of the appointment including the medical centre and the time) could be used to reveal the identity of the participants of the screening programme. Particularly, the information of the medical appointment may be incriminating, either intentionally or by accident, that disclose a specific person attends a particular centre at a known date. Then, it will be enough with consulting the tests results to know if the individual with the quasi-identifiers, which we know, suffers from illness. The objective of this project is to programme the assignment of the appointments, in order to increase the anonymity level of the final database. Then, the plan is that people with the same characteristics (age, etc) attend the same appointment (time and centre), hinder the identification of the medical results of a specific individual even if his or her personal data are known, because, the information of the same quasiidentifiers of more than one person will be found.
The publication of screening programmes’ results (examination or testing of a group of individuals to separate those who are well from those who have an undiagnosed disease or defect or who are at high risk [11]) is an important factor for the scientific community. In spite of deleting personal data such as ID, name, etc. the rest of the information, which is called quasi-identifiers (P.O. Box, gender, age, profession or the information of the appointment including the medical centre and the time) could be used to reveal the identity of the participants of the screening programme. Particularly, the information of the medical appointment may be incriminating, either intentionally or by accident, that disclose a specific person attends a particular centre at a known date. Then, it will be enough with consulting the tests results to know if the individual with the quasi-identifiers, which we know, suffers from illness. The objective of this project is to programme the assignment of the appointments, in order to increase the anonymity level of the final database. Then, the plan is that people with the same characteristics (age, etc) attend the same appointment (time and centre), hinder the identification of the medical results of a specific individual even if his or her personal data are known, because, the information of the same quasiidentifiers of more than one person will be found.
Description
Trabajo de Fin de Grado en Ingeniería Informática (Universidad Complutense, Facultad de Informática, curso 2015/2016)