Your browser does not support JavaScript!

Home    Collections    Type of Work    Post-graduate theses  

Post-graduate theses

Current Record: 29 of 4837

Back to Results Previous page
Next page
Add to Basket
[Add to Basket]
Identifier 000428258
Title Towards a universal molecular classifier
Alternative Title Προς την ανάπτυξη ενός καθολικού μοριακού ταξινομητή
Author Φαφαλιός, Στέφανος Μ.
Thesis advisor Τσαμαρδινός, Ιωάννης
Reviewer Τόλλης, Ιωάννης
Conesa, Ana
Abstract Despite the plethora of biological datasets, it is difficult to gather a satisfactory amount of as homogeneous as possible biological data and derive safe conclusions regarding disease diagnosis. On top of that, a challenging problem of disease diagnosis and their treatment is to identify their complex characteristics that span to tens of thousands. We mathematically shape these characteristics using publicly available datasets, from disease studies, via a low dimensional space representation. In essence, using sophisticated statistical methods, suitable for very high dimensional data, we are able to provide a list of the most probable diseases for a given patient, but also include don’t-know-predictions (cases with uncertain disease diagnosis). The benefit of this diagnostic framework is that it can assist medical doctors by directing them towards the most probable diseases for a given patient and allows for timely disease diagnosis and treatment. The analysis is time-efficient and can be conducted in real time, with very little computational power and low ram and disk memory requirements.
Language English
Subject Gene-expression
High-dimensional
Statistical analysis
Issue date 2020-03-27
Collection   Faculty/Department--Faculty of Sciences and Engineering--Department of Computer Science--Post-graduate theses
  Type of Work--Post-graduate theses
Permanent Link https://elocus.lib.uoc.gr//dlib/9/f/c/metadata-dlib-1581667818-84711-3641.tkl Bookmark and Share
Views 221

Digital Documents
No preview available

View document
Views : 8