LIST OF ON-GOING AND RECENTLY COMPLETED THESES
[Collaboration between Greece-Italy-Germany, GIG.]
Aim: The study of images PET, CT and MRI images from patients with cancers which present metastases, such as melanomas and sarcomas. The scope is to use the methods of fractal and multifractal analyses to develop indices related to the stage of the disease as well as parameters to evaluate the therapeutic outcome after medical treatments. Relevant quantities are the fractal dimensions of the image, the mass dimension, the autocorrelation function, the information dimension and the multifractal spectrum. The proposed MSc (or BA) thesis is a continuation of an earlier work by the same collaboration group (see ref. ).
For each case the following steps will be followed:
- A number of high resolution 2D Dicom images are given for each subject covering all human body (see typical 2D picture).
- As a first step the superposition of the 2D images into a 3D structure is required.
- A 3D digital representation of the human body is performed.
- Fractal dimensions, Correlations, Multifractal Spectra and the other biomarkers are computed.
- Comparisons of biomarkers between patients and controls The method will be applied to 25 patients, at different stages of the melanoma/sarcoma disease, before and after medical treatments and will be compared with those of a group of healthy subjects (control group).
 Breki CM, Dimitrakopoulou-Strauss A, Hassel J, Theoxaris T, Sachpekidis C, Pan L, Provata A, “Fractal and multifractal analysis of PET/CT images of metastatic melanoma before and after treatment with ipilimumab”, EJNMMI Research 6(1):61, 2016.
 Medical PET Group - Biological Imaging, “Quantitative approaches of dynamic FDG-PET and PET/CT studies (dPET/CT) for the evaluation of oncological patients”, Cancer Imaging, 10/2012; 12(1):283-289.  Dimitrakopoulou-Strauss A, Strauss LG, Burger C, Mikolajczyk K, Lehnert T, Bernd L, Ewerbeck V, “On the fractal nature of dynamic positron emission tomography (PET) studies”. W J Nucl Med, 2:306-313, 2003.
The proposed project aims to study the spatial correlations in systems of leaky integrate-and-fire neurons arranged on two rings of mutually coupled elements. The question is to find the parameter values for which the multiplex network synchonises or forms chimera states and to use the correlation function as a measure of synchronization.
The proposed project aims to highlight the influence of the spatial distribution of a population in the evolution of an SIRS epidemic process. The model is composed by a number of susceptible (S), infected (I) and recovered (R) individuals (agents) interacting locally on a 2D lattice. The agents, apart from interacting and infecting their nearest neighbours, they have the chance to travel long or short distances on the lattice and thus influence distant regions. We investigate how the existence of spatial distribution of the agents and motion (diffusion) on the lattice modifies the evolution of the epidemics compared to the results of the SIRS mean field model.