Nedim ŠIŠIĆ: Deep learning in whole brain MRI segmentation: A Review

V ponedeljek, 11. avgusta 2025, bo ob 16:00 uri izvedeno
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.

ČAS/PROSTOR: 11. avgust 2025 ob 16.00 prek Zoom-a (https://upr-si.zoom.us/j/297328207?pwd=S3Zpdk1VR3pjckNtWkQwKzlvcDR5UT09).

————————————
PREDAVATELJ: Nedim ŠIŠIĆ
————————————

Nedim Šišić is a PhD student at UP FAMNIT under the supervision of Assoc. Prof. Peter Rogelj. His research focuses on segmentation of T1-weighted human brain MRI using deep learning models. His broader interests include natural language processing and artificial general intelligence.

———————————————————————————————
NASLOV: Deep learning in whole brain MRI segmentation: A Review
———————————————————————————————

POVZETEK:

Whole brain MRI segmentation plays a crucial role in neuroimaging studies and clinical trials by enabling precise localization and quantification of brain tissues and structures. The advent of deep learning has transformed the field, offering accurate and fast tools for MRI segmentation. In this review, we provide a comprehensive analysis of developments in deep learning-based segmentation of whole brain MRI of adults into tissues, structures, or regions of interest. We explore key factors influencing segmentation performance, including architectural design and choice of input size and model dimensionality. We also address validation practices, which are particularly important given the scarcity of manual annotations, and identify limitations in current methodologies. We present an extensive compilation of existing segmentation works and highlight emerging trends and key results. Finally, we discuss the challenges and potential future directions in the field.

Seminar bo potekal v angleškem jeziku.

Seminar bo izveden online prek Zoom-a na sledeči povezavi:

https://upr-si.zoom.us/j/297328207?pwd=S3Zpdk1VR3pjckNtWkQwKzlvcDR5UT09

Meeting ID: 297 328 207
Passcode: 123456789

Vabljeni!