Naslov
3D U-Net based method for fast segmentation of whole heart from CT images

Autori
Novoselnik, Filip ; Leventic, Hrvoje ; Galic, Irena ; Babin, Danilo

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
2022 International Symposium ELMAR / - Zagreb : Institute of Electrical and Electronics Engineers (IEEE), 2022, 159-164

ISBN
978-1-6654-7003-2

Skup
64th International Symposium ELMAR-2022

Mjesto i datum
Zadar, Hrvatska, 12.09.2022. - 14.09.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Whole heart segmentation ; deep learning ; deep neural networks ; NoNewNet

Sažetak
There is an increasing number of clinical applications where deep learning plays an important role. Heart chamber segmentation enables delineation of anatomical structures of heart and it is a prerequisite for a wide range of clinical applications. 3D U-Net architecture consistently achieves the highest scores in various medical imaging challenges. NoNewNet architecture is a modification of the 3D U-Net architecture which was shown to outperform the original 3D U-Net and was recently implemented inside the NiftyNet package. In this paper we demonstrate that with the properly trained NoNewNet network and NiftyNet we can outperform the current state-of-the-art networks. The evaluation of the trained network was performed on 20 3D CT images from MICCAI 2017 Multi-Modality Whole Heart Segmentation Challenge dataset using five-fold cross-validation. We experimentally prove that border size can significantly reduce inference time without affecting segmentation accuracy. Additionally, we provide the discussion of the effects of some of the NiftyNet configuration parameters on the performance of the network.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo, Interdisciplinarne biotehničke znanosti, Biotehnologija u biomedicini (prirodno područje, biomedicina i zdravstvo, biotehničko područje)

POVEZANOST RADA