MRC Biostatistics Unit

I’m a PhD student at the University of Cambridge, currenly supervised by Oscar Rueda at the MRC Biostatistics Unit.

I’m interested in many areas across statistics, machine learning, and their interactions. My research focuses on tailored model development, that is targeted model building for a specific task of interest, mostly prediction. Under this scenario it is desired to use a metric which reflects the loss function to be used for the prediction problem thus “making” the model perform well for the particular task. Our approach is based on general Bayesian learning by incorporating loss functions into Bayesian inference. We will explore the use of such a framework for predictive and prognostic model building, whereby loss functions could be used to target metrics of real world clinical utility tailored to a particular setting.

Before starting my PhD, I studied for a MSc in Statistics at KU Leuven. I focused on the Biometrics track and wrote my dissertation under the supervision of professor Geert Verbeke. I worked as Research Scientist at the MRC Biostatistics Unit during the 2016-2017 academic year supervised by Paul Newcombe and Chris Jackson.

- Bayesian Methods
- Biomedical Research
- Statistical Computing
- Statistical Modelling
- Longitudinal Data Analysis
- Statistical Learning
- Machine Learning
- Medical Decision Making
- Predictive Modelling and Model Validation
- Survival Analysis

—> *Quickly discover relevant content by filtering publications.*

False Positive Multiparametric Magnetic Resonance Imaging Phenotypes in the Biopsy-naïve Prostate: Are They Distinct from Significant Cancer-associated Lesions? Lessons from PROMIS.
In *EUROPEAN UROLOGY*.

(2021).
Development and External Validation of Prediction Models for 10-Year Survival of Invasive Breast Cancer. Comparison with PREDICT and CancerMath.
In *Clinical Cancer Research*.

(2018).
Ventilatory limitation and dynamic hyperinflation during exercise testing in Cystic Fibrosis.
In *Pediatric pulmonology*, 52(1), 29-33.

(2017).
Isocapnic hyperpnea with a portable device in Cystic Fibrosis: an agreement study between two different set-up modalities.
In *Journal of Clinical Monitoring and Computing*, 29(5), 569-572.

(2015).
Cystic fibrosis patients’ performance on Modified Shuttle Walk Test.
In *Journal of Cystic Fibrosis*, 13(2), S91.

(2014).
I have supervised Statistics IB, Lent 2019 (2nd year undergraduate course from the Department of Pure Mathematics and Mathematical Statistics, University of Cambridge).

This is the handbook for the course I teach as a Brilliant club tutor. The Brilliant Club is a charity aiming to increase the number of pupils from under-represented backgrounds progressing to highly selective universities. They do this by mobilising PhD researchers to share their academic expertise with state schools.

The material is designed for Key stage 4 pupils but can be easily adapted to Key stage 5.

About the course: Virtually every decision is made in the face of uncertainty. In this course, I quantify uncertainty using probability theory. I then introduce the expected utility framework as a model of choice behaviour under uncertainty.

Cite this work as: Solon Karapanagiotis. Which bicycle lock should I buy? A journey to decision making under uncertainty, 2019.

Statistics · R · Random

Attempt to solve #115 “A random robot” puzzle from the New Scientist. This week (No 3336 - 29 May 2021) the New Scientist published the following puzzle:
Roman the test robot is being given one final roam before being consigned to the scrapheap where he can rust in peace.

It is usually taught in statistics classes that Binomial probabilities can be approximated by Poisson probabilities, which are generally easier to calculate. This approximation is valid “when \(n\) is large and \(np\) is small,” and rules of thumb are sometimes given.

I present a solution to a modification of the “hardest logic puzzle ever” using probability theory.
Background “The hardest logic puzzle” was originally presented by Boolos (1996) and since then it has been amended several times in order to make it harder (see Rabern and Rabern 2008; Novozhilov 2012).