TU Darmstadt / ULB / TUbiblio

OSS-Net: Memory Efficient High Resolution Semantic Segmentation of 3D Medical Data

Reich, C. ; Prangemeier, T. ; Özdemir, C. ; Koeppl, H. (2021)
OSS-Net: Memory Efficient High Resolution Semantic Segmentation of 3D Medical Data.
32nd British Machine Vision Conference. virtual Conference (22.-25.11.2021)
Conference or Workshop Item, Bibliographie

Abstract

Convolutional neural networks (CNNs) are the current state-of-the-art meta-algorithm for volumetric segmentation of medical data, for example, to localize COVID-19 infected tissue on computer tomography scans or the detection of tumour volumes in magnetic resonance imaging. A key limitation of 3D CNNs on voxelised data is that the memory consumption grows cubically with the training data resolution. Occupancy networks (O-Nets) are an alternative for which the data is represented continuously in a function space and 3D shapes are learned as a continuous decision boundary. While O-Nets are significantly more memory efficient than 3D CNNs, they are limited to simple shapes, are relatively slow at inference, and have not yet been adapted for 3D semantic segmentation of medical data. Here, we propose Occupancy Networks for Semantic Segmentation (OSS-Nets) to accurately and memory-efficiently segment 3D medical data. We build upon the original O-Net with modifications for increased expressiveness leading to improved segmentation performance comparable to 3D CNNs, as well as modifications for faster inference. We leverage local observations to represent complex shapes and prior encoder predictions to expedite inference. We showcase OSS-Net's performance on 3D brain tumour and liver segmentation against a function space baseline (O-Net), a performance baseline (3D residual U-Net), and an efficiency baseline (2D residual U-Net). OSS-Net yields segmentation results similar to the performance baseline and superior to the function space and efficiency baselines. In terms of memory efficiency, OSS-Net consumes comparable amounts of memory as the function space baseline, somewhat more memory than the efficiency baseline and significantly less than the performance baseline. As such, OSS-Net enables memory-efficient and accurate 3D semantic segmentation that can scale to high resolutions.

Item Type: Conference or Workshop Item
Erschienen: 2021
Creators: Reich, C. ; Prangemeier, T. ; Özdemir, C. ; Koeppl, H.
Type of entry: Bibliographie
Title: OSS-Net: Memory Efficient High Resolution Semantic Segmentation of 3D Medical Data
Language: English
Date: 2021
Event Title: 32nd British Machine Vision Conference
Event Location: virtual Conference
Event Dates: 22.-25.11.2021
URL / URN: https://arxiv.org/abs/2110.10640
Abstract:

Convolutional neural networks (CNNs) are the current state-of-the-art meta-algorithm for volumetric segmentation of medical data, for example, to localize COVID-19 infected tissue on computer tomography scans or the detection of tumour volumes in magnetic resonance imaging. A key limitation of 3D CNNs on voxelised data is that the memory consumption grows cubically with the training data resolution. Occupancy networks (O-Nets) are an alternative for which the data is represented continuously in a function space and 3D shapes are learned as a continuous decision boundary. While O-Nets are significantly more memory efficient than 3D CNNs, they are limited to simple shapes, are relatively slow at inference, and have not yet been adapted for 3D semantic segmentation of medical data. Here, we propose Occupancy Networks for Semantic Segmentation (OSS-Nets) to accurately and memory-efficiently segment 3D medical data. We build upon the original O-Net with modifications for increased expressiveness leading to improved segmentation performance comparable to 3D CNNs, as well as modifications for faster inference. We leverage local observations to represent complex shapes and prior encoder predictions to expedite inference. We showcase OSS-Net's performance on 3D brain tumour and liver segmentation against a function space baseline (O-Net), a performance baseline (3D residual U-Net), and an efficiency baseline (2D residual U-Net). OSS-Net yields segmentation results similar to the performance baseline and superior to the function space and efficiency baselines. In terms of memory efficiency, OSS-Net consumes comparable amounts of memory as the function space baseline, somewhat more memory than the efficiency baseline and significantly less than the performance baseline. As such, OSS-Net enables memory-efficient and accurate 3D semantic segmentation that can scale to high resolutions.

Alternative keywords:
Alternative keywordsLanguage
Image and Video Processing, Computer Vision and Pattern Recognition, Machine LearningEnglish
Divisions: 18 Department of Electrical Engineering and Information Technology
18 Department of Electrical Engineering and Information Technology > Institute for Telecommunications > Bioinspired Communication Systems
18 Department of Electrical Engineering and Information Technology > Institute for Telecommunications
Interdisziplinäre Forschungsprojekte
Interdisziplinäre Forschungsprojekte > Centre for Synthetic Biology
Date Deposited: 29 Oct 2021 06:31
Last Modified: 29 Oct 2021 06:31
PPN:
Alternative keywords:
Alternative keywordsLanguage
Image and Video Processing, Computer Vision and Pattern Recognition, Machine LearningEnglish
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details