Bano, Sophia;
Shepherd, Luke M;
Manjaly, Joseph G;
Lindenroth, Lukas;
Stoyanov, Danail;
(2022)
Intra-Domain Adaptation for Robust Visual Guidance in Intratympanic Injections.
Presented at: 11th International Conference on New Technologies for Computer and Robot Assisted Surgery, Naples, Italy.
Preview |
Text
CRAS_Luke.pdf - Accepted Version Download (3MB) | Preview |
Abstract
Intratympanic steroid injections are commonly used for the treatment of ear diseases. During this treatment, an expert Ear, Nose & Throat (ENT) clinician delivers the drug by viewing through a large microscope that provides a close-up view of the anatomical landmarks on the middle ear. A steady hand and swift response to any patient movement are required to avoid improper placement of the needle. To assist the clinician during this treatment, a fluidic soft robot is proposed in \cite{lindenroth2021fluidic} that can steer inside a lumen for providing steady guidance for drug delivery. For robust visual guidance, stable anatomical landmarks (tympanic membrane, malleus, umbo) segmentation is required. In this work, we perform intra-domain adaptation to learn a generalized model that provides stable and consistent segmentation on unseen patients and phantom ear data.
Type: | Conference item (Presentation) |
---|---|
Title: | Intra-Domain Adaptation for Robust Visual Guidance in Intratympanic Injections |
Event: | 11th International Conference on New Technologies for Computer and Robot Assisted Surgery |
Location: | Naples, Italy |
Dates: | 25 - 27 Apr 2022 |
Open access status: | An open access version is available from UCL Discovery |
Publisher version: | https://cras-eu.org/cras-2022/ |
Language: | English |
UCL classification: | UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science UCL > Provost and Vice Provost Offices > UCL BEAMS UCL |
URI: | https://discovery.ucl.ac.uk/id/eprint/10157677 |




Archive Staff Only
![]() |
View Item |