Personalized reference interval determination using quantile mixed model

In recent years, the concept of personalized medicine has gained increasing importance. In this context, the unique molecular, physiological, and behavioral characteristics of individuals vary, influenced by environmental factors. However, the success of personalized interventions is dependent on accurate diagnosis and patient monitoring, and for these purposes, it is necessary to interpret patients' test results in a "personalized" manner. This can be done by comparing patients' test results with their own personal reference intervals (prRI) and reducing the use of common population-based reference intervals (popRI). This thesis aims to increase the accuracy and applicability of personalized reference intervals by incorporating demographic variables such as age and gender into quantile mixed-effects models. The performance of these models will be evaluated using bootstrap methods. The model will include demographic factors such as age and gender and will treat biological variations as random effects. These additions are expected to contribute to a more precise and personalized interpretation of clinical test results. Especially considering that individuals' demographic characteristics can create significant differences in disease detection and monitoring, it is necessary to expand this model. Future studies will focus on evaluating the effectiveness of these expansions. As a result, it is anticipated that personalized medicine practices can be more widely and effectively utilized in clinical laboratories.