dc.rights.license | CC-BY-NC-ND | |
dc.contributor.advisor | Ophelders, Tim | |
dc.contributor.author | Kemp, Matthijs | |
dc.date.accessioned | 2024-02-05T00:00:47Z | |
dc.date.available | 2024-02-05T00:00:47Z | |
dc.date.issued | 2024 | |
dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/45895 | |
dc.description.abstract | Recently it was discovered that brain tumor cells exhibit tumor microtubes, long membranous tubes that provide a physical connection between neighboring tumor cells which can be harnessed to communicate or facilitate invasion. Cell segmentation and tracking is essential to better study the role of tumor microtubes in the invasiveness of certain brain malignancies. This research aims to develop a deep learning model to segment tumor microtubes and track their nuclei. A supervised training method was used to train several models. The performance of these models is compared to state-of-the-art models using widely-used metrics. The method developed in this work enables higher throughput research into cell-to-cell interactions and vastly speeds up behavioral studies on tumor microtubes. | |
dc.description.sponsorship | Utrecht University | |
dc.language.iso | EN | |
dc.subject | Deep learning-based segmentation and tracking of astrocytes exhibiting tumor microtubes due to Diffuse Midline Glioma. | |
dc.title | Deep learning-based segmentation of cell protrusions | |
dc.type.content | Master Thesis | |
dc.rights.accessrights | Open Access | |
dc.subject.keywords | segmentation; tracking; cancer; DMG; DIPG; deep learning | |
dc.subject.courseuu | Artificial Intelligence | |
dc.thesis.id | 14342 | |