Aristotle University Of Thessaloniki (AUTH)
Utilizing the results of multi-omics data analysis and machine learning algorithms towards the investigation of clinical predictions in acute myeloid leukemia
Abstract
dc:descriptionThis thesis investigates the integration of network analysis with machine learning for predictive modeling in Acute Myeloid Leukemia, leveraging multi-omics data. A scoping review identifies a gap in the application of network-based approaches to outcome prediction in cancer research. Building on this, the thesis evaluates network-based feature selection against traditional methods using gene expression data, demonstrating the superior predictive power of incorporating network context. A configurable pipeline is then developed to apply this methodology for AML outcome prediction using both single and dual-layer omics data, showing improved accuracy over standard approaches. This is extended to a multi-layer framework incorporating additional omics types such as SNPs and CNAs, further enhancing the proposed methodology. To bridge model predictions with biological relevance, an interpretability framework is proposed, combining feature importance with biological network visualizations. Furthermore, a clinician-facing user interface is developed to evaluate the usability and decision-support potential of the models in simulated clinical settings. The thesis concludes with a discussion on the implications of multi-omics integration for AML prognosis, the challenges of clinical translation, and directions for future research in personalized oncology.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kosvyra, Alexandra
- Κοσβύρα, Αλεξάνδρα
Subjects
dc:subject × 13- Μηχανική μάθηση
- Πολυομικά Δεδομένα
- Σύστημα Υποστήριξη Λήψης Απόφασης
- Συνδυαστική Ανάλυση Δικτύων
- Machine learning
- Multi-omics
- Clinical decision support systems
- Integrated network analysis
- Βιοϊατρική Μηχανική
- Επιστήμες Μηχανικού και Τεχνολογία
- Medical Engineering
- Engineering and Technology
- Biomedical Engineering
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/60056