TY  - CONF
AU  - Kügler, David
AU  - Uecker, Marc
AU  - Kuijper, Arjan
AU  - Mukhopadhyay, Anirban
A3  - Martel, Anne L.
A3  - Abolmaesumi, Purang
A3  - Stoyanov, Danail
A3  - Mateus, Diana
A3  - Zuluaga, Maria A.
A3  - Zhou, S. Kevin
A3  - Racoceanu, Daniel
A3  - Joskowicz, Leo
TI  - AutoSNAP: Automatically Learning Neural Architectures for Instrument Pose Estimation
VL  - 12263
CY  - Cham
PB  - Springer International Publishing
M1  - DZNE-2025-01010
SN  - 978-3-030-59715-3 (print)
T2  - Lecture Notes in Computer Science
SP  - 375 - 384
PY  - 2020
AB  - Despite recent successes, the advances in Deep Learning have not yet been fully translated to Computer Assisted Intervention (CAI) problems such as pose estimation of surgical instruments. Currently, neural architectures for classification and segmentation tasks are adopted ignoring significant discrepancies between CAI and these tasks. We propose an automatic framework (AutoSNAP) for instrument pose estimation problems, which discovers and learns architectures for neural networks. We introduce 1) an efficient testing environment for pose estimation, 2) a powerful architecture representation based on novel Symbolic Neural Architecture Patterns (SNAPs), and 3) an optimization of the architecture using an efficient search scheme. Using AutoSNAP, we discover an improved architecture (SNAPNet) which outperforms both the hand-engineered i3PosNet and the state-of-the-art architecture search method DARTS.
T2  - Medical Imaging Computing and Computer Assisted Intervention
CY  - 4 Oct 2020 - 8 Oct 2020, Lima (Peru)
Y2  - 4 Oct 2020 - 8 Oct 2020
M2  - Lima, Peru
LB  - PUB:(DE-HGF)8 ; PUB:(DE-HGF)7
DO  - DOI:10.1007/978-3-030-59716-0_36
UR  - https://pub.dzne.de/record/280927
ER  -