Main Host Institution: University of Granada (UGr) – Spain.

Host Lab: Computer Sciences and Artificial Intelligence Lab (DaSCI).

Project Title: NSCLC tumour identification via Machine Learning Techniques from Liquid Biopsy Components.

Project Background: The goal of my PhD thesis is to accurately detect Non-Small Cell Lung Cancer (NSCLC) samples amongst non-cancer ones and give a proof of the veracity of the workflow leading to the detection. Unlike deep learning, a fully transparent machine learning model must be used to learn pattern between the two conditions. Because in clinical settings there is the need to know which genes, for example, leads to classify a given sample to NSCLC. Samples could be any kind of biological information extracted from an individual by liquid biopsy. The latter technique aims to study biological components circulating in the blood and in contact with tumour: Tumour-Educated Platelets (TEPs), Circulating Tumour Cells (CTCs), circulating tumour DNA (ctDNA) and exosomes. From theses components, DNA and RNA can be retrieved. We can thus use DNA mutations, genes expressions, RNA editing, as input of the detection model.

Project Aim:

• Developing pioneer machine learning algorithms to accurately predict early stage NSCLC using non-invasively obtained biosources.