Lejeune, Laurent Georges Pascal; Christoudias, Mario; Sznitman, Raphael (16 July 2017). Expected exponential loss for gaze-based video and volume ground truth annotation. In: International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), LABELS - Workshop. 10.1007/978-3-319-67534-3_12
Text
1707.04905.pdf - Published Version Restricted to registered users only Available under License Publisher holds Copyright. Download (4MB) |
Many recent machine learning approaches used in medical imaging are highly reliant on
large amounts of image and groundtruth data. In the context of object segmentation, pixelwise annotations are extremely expensive to collect, especially in video and 3D volumes. To reduce this annotation burden, we propose a novel framework to allow annotators to simply observe the object to segment and record where they have looked at with a $200 eye gaze tracker. Our method then estimates pixel-wise probabilities for the presence of the object throughout the sequence from which we train a classifier in semi-supervised setting using a novel Expected Exponential loss function. We show that our framework provides superior performances on a wide range of medical image settings compared to existing strategies and that our method can be combined with current crowd-sourcing paradigms as well.
Item Type: |
Conference or Workshop Item (Paper) |
---|---|
Division/Institute: |
10 Strategic Research Centers > ARTORG Center for Biomedical Engineering Research > ARTORG Center - AI in Medical Imaging Laboratory |
Graduate School: |
Graduate School for Cellular and Biomedical Sciences (GCB) |
UniBE Contributor: |
Lejeune, Laurent Georges Pascal, Sznitman, Raphael |
Subjects: |
600 Technology > 610 Medicine & health 600 Technology > 620 Engineering |
Series: |
Lecture Notes in Computer Science |
Language: |
English |
Submitter: |
Raphael Sznitman |
Date Deposited: |
01 May 2018 09:58 |
Last Modified: |
05 Dec 2022 15:09 |
Publisher DOI: |
10.1007/978-3-319-67534-3_12 |
ArXiv ID: |
1707.04905 |
BORIS DOI: |
10.7892/boris.108438 |
URI: |
https://boris.unibe.ch/id/eprint/108438 |