File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Cephalometric Landmark Detection by Attentive Feature Pyramid Fusion and Regression-Voting

TitleCephalometric Landmark Detection by Attentive Feature Pyramid Fusion and Regression-Voting
Authors
KeywordsCephalometric landmarks
Deep learning
Self-attention
Fusion feature
Regression-voting
Issue Date2019
PublisherSpringer
Citation
The 22nd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2019), Shenzhen, China, 13-17 October 2019. In Shen, D. et al. (eds), Proceedings, pt 3, p. 873-881 How to Cite?
AbstractMarking anatomical landmarks in cephalometric radiography is a critical operation in cephalometric analysis. Automatically and accurately locating these landmarks is a challenging issue because different landmarks require different levels of resolutions and semantics. Based on this observation, we propose a novel attentive feature pyramid fusion module (AFPF) to explicitly shape high-resolution and semantically enhanced fusion features to achieve significantly higher accuracy than existing deep learning-based methods. We also combine heat maps and offset maps to perform pixel-wise regression-voting to improve detection accuracy. By incorporating the AFPF and regression-voting, we develop an end-to-end deep learning framework that improves detection accuracy by 7%–11% for all the evaluation metrics over the state-of-the-art method. We present ablation studies to give more insights into different components of our method and demonstrate its generalization capability and stability for unseen data from diverse devices.
Persistent Identifierhttp://hdl.handle.net/10722/293821
ISBN
ISI Accession Number ID
Series/Report no.Lecture Notes in Computer Science (LNCS); v. 11766

 

DC FieldValueLanguage
dc.contributor.authorChen, R-
dc.contributor.authorMa, Y-
dc.contributor.authorChen, N-
dc.contributor.authorLee, D-
dc.contributor.authorWang, WP-
dc.date.accessioned2020-11-23T08:22:16Z-
dc.date.available2020-11-23T08:22:16Z-
dc.date.issued2019-
dc.identifier.citationThe 22nd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2019), Shenzhen, China, 13-17 October 2019. In Shen, D. et al. (eds), Proceedings, pt 3, p. 873-881-
dc.identifier.isbn9783030322472-
dc.identifier.urihttp://hdl.handle.net/10722/293821-
dc.description.abstractMarking anatomical landmarks in cephalometric radiography is a critical operation in cephalometric analysis. Automatically and accurately locating these landmarks is a challenging issue because different landmarks require different levels of resolutions and semantics. Based on this observation, we propose a novel attentive feature pyramid fusion module (AFPF) to explicitly shape high-resolution and semantically enhanced fusion features to achieve significantly higher accuracy than existing deep learning-based methods. We also combine heat maps and offset maps to perform pixel-wise regression-voting to improve detection accuracy. By incorporating the AFPF and regression-voting, we develop an end-to-end deep learning framework that improves detection accuracy by 7%–11% for all the evaluation metrics over the state-of-the-art method. We present ablation studies to give more insights into different components of our method and demonstrate its generalization capability and stability for unseen data from diverse devices.-
dc.languageeng-
dc.publisherSpringer-
dc.relation.ispartofInternational Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2019 Proceedings, Part III-
dc.relation.ispartofseriesLecture Notes in Computer Science (LNCS); v. 11766-
dc.subjectCephalometric landmarks-
dc.subjectDeep learning-
dc.subjectSelf-attention-
dc.subjectFusion feature-
dc.subjectRegression-voting-
dc.titleCephalometric Landmark Detection by Attentive Feature Pyramid Fusion and Regression-Voting-
dc.typeConference_Paper-
dc.identifier.emailWang, WP: wenping@cs.hku.hk-
dc.identifier.authorityWang, WP=rp00186-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-030-32248-9_97-
dc.identifier.hkuros319240-
dc.identifier.volumept 3-
dc.identifier.spage873-
dc.identifier.epage881-
dc.identifier.isiWOS:000548733600097-
dc.publisher.placeCham-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats