PhD defence: Ludvig Renbo Olsen
Cross-cohort pan-cancer detection from cell-free DNA
Info about event
Time
Location
The MOMA ground floor auditorium, Building D/E, Aarhus University
On Wednesday 19 March at 13:00, Ludvig Renbo Olsen defends his PhD dissertation entitled "Cross-dataset pan-cancer detection using machine learning on cell-free DNA fragmentation patterns".
What if a simple blood sample could detect cancer at an early stage, allowing treatment before it can spread to vital organs? A new PhD project from Aarhus University has developed two methods for tumor-agnostic cancer detection, using machine learning on the fragmentation patterns in cell-free DNA (cfDNA). The methods are specifically optimized to work across multiple cohorts - a fundamental challenge in the field, as biases in data generation often reduce the transferability of statistical models.
The first method, LIONHEART, correlates the amount of cfDNA across the genome with the chromatin accessibility of more than 400 cell types. These correlations enable cancer detection across nine different cohorts.
The second method separates the cohort-specific biases from cancer signals in the cfDNA fragment lengths using extended Non-Negative Matrix Factorization (NMF).
The summary is written by the PhD student.
The defence is public and takes place in the MOMA ground floor auditorium, Building D/E, Aarhus University. Please see the press release for more information.
Contact
PhD student Ludvig Renbo Olsen
Mail: ludvig@clin.au.dk
Phone: (+45) 71790209