Thesis topic proposal
Ágnes Fogarassyné Vathy
Supporting data analysis of retrospective health studies by data mining methods


Institute: University of Pannonia
computer sciences
Doctoral School of Information Science and Technology

Thesis supervisor: Ágnes Fogarassyné Vathy
Location of studies (in Hungarian): University of Pannonia, Faculty of Information Technology, Department of Computer Science and Systems Technology
Abbreviation of location of studies: PE

Description of the research topic:

The escalating growth of electronic information in the field of healthcare puts a spotlight on the retrospective clinical trials. Processing of large amount data with traditional statistical methods is not always effective, or it is sometimes limited to implementation constraints. Data mining, including its specialized fields such as network analysis and process mining, is one of the most important tools for discovering patterns from large data sets. However, due to the unique nature of the health care and the complexity of the human biological system, for efficient analysis in the healthcare domain, data mining methods can only be used after area specific extensions. The improved and healthcare-adapted data mining algorithms can effectively contribute to the analysis of retrospective clinical trials, and they can provide a basis to explore the influencing factors affecting the human biology system. This new knowledge helps physicians achieve individualized medicine. The aim of the research is to develop such new healthcare-adapted data mining methods and algorithms which can effectively contribute to the exploration of information from large
healthcare datasets. The effectiveness of the suggested methods should be presented in real case studies. The research includes the following topics: development of new control group selection methods for case-control trials; analyzing the effect of control group selection methods on the analysis outcome; exploration of complex connection systems of independent
and confounding variables and the analysis of their effects on healing processes. Suggested methods should be evaluated by medical experts.

Preliminary results can be found in the following publications:
[1] Peter C Austin, Nathaniel Jembere, Maria Chiu. Propensity score matching and complex surveys. Statistical Methods in Medical Research (2016) Online First
[2] Austin, Peter C. The Performance of Different Propensity Score Methods for Estimating Marginal Hazard Ratios. Statistics in Medicine 32.16 (2013): 2837-2849.
[3] Hosmer Jr, David W., Stanley Lemeshow, and Rodney X. Sturdivant. Applied logistic regression. Vol. 398. John Wiley & Sons, 2013.
[4] Harrell, Frank E., Kerry L. Lee, and Daniel B. Mark. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Statistics in Medicine 15.4 (1996): 361-387.

Number of students who can be accepted: 1

Deadline for application: 2018-06-22

2019. I. 10.
ODT ülés
Az ODT következő ülésére 2019. február 22-én 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).

All rights reserved © 2007, Hungarian Doctoral Council. Doctoral Council registration number at commissioner for data protection: 02003/0001. Program version: 2.2358 ( 2017. X. 31. )