V
Velizar Efremov
Researcher at University of Zurich
Publications - 4
Citations - 383
Velizar Efremov is an academic researcher from University of Zurich. The author has contributed to research in topics: Deep learning & Overfitting. The author has an hindex of 4, co-authored 4 publications receiving 206 citations. Previous affiliations of Velizar Efremov include Technische Universität München.
Papers
More filters
Journal ArticleDOI
Machine learning analysis of whole mouse brain vasculature.
Mihail Ivilinov Todorov,Johannes C. Paetzold,Oliver Schoppe,Giles Tetteh,Suprosanna Shit,Velizar Efremov,Velizar Efremov,Katalin Todorov-Völgyi,Marco Düring,Martin Dichgans,Martin Dichgans,Marie Piraud,Bjoern H. Menze,Ali Ertürk +13 more
TL;DR: A deep learning-based framework to quantify and analyze brain vasculature, named Vessel Segmentation & Analysis Pipeline (VesSAP), which uses a convolutional neural network with a transfer learning approach for segmentation and achieves human-level accuracy.
Journal ArticleDOI
DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes.
Giles Tetteh,Velizar Efremov,Velizar Efremov,Nils D. Forkert,Matthias Schneider,Jan S. Kirschke,Bruno Weber,Claus Zimmer,Marie Piraud,Bjoern H. Menze,Bjoern H. Menze +10 more
TL;DR: The DeepVesselNet architecture does not use any form of sub-sampling layer and works well for vessel segmentation, centerline prediction, and bifurcation detection, and the results show that cross-hair filters achieve over 23% improvement in speed, lower memory footprint, lower network complexity which prevents overfitting and comparable accuracy that does not differ from full 3-D filters.
Posted Content
DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes
Giles Tetteh,Velizar Efremov,Velizar Efremov,Nils D. Forkert,Matthias Schneider,Jan S. Kirschke,Bruno Weber,Claus Zimmer,Marie Piraud,Bjoern H. Menze,Bjoern H. Menze +10 more
TL;DR: DeepVesselNet as discussed by the authors uses 2-D orthogonal cross-hair filters to extract vessel networks or trees and corresponding features in 3-D angiographic volumes using deep learning.
Transfer learning from synthetic data reduces need for labels to segment brain vasculature and neural pathways in 3D
Johannes C. Paetzold,Oliver Schoppe,Rami Al-Maskari,Giles Tetteh,Velizar Efremov,Mihail Ivilinov Todorov,Ruiyao Cai,Hongcheng Mai,Zhouyi Rong,Ali Ertuerk,Bjoern H. Menze +10 more
TL;DR: This work shows how transfer learning and synthetic data generation can be used to train deep neural networks to segment complex anatomical structures successfully in the absence of or with very limited training data.