A Segmentation-Assisted Model for Universal Lesion Detection with Partial Labels

Fei Lyu, Baoyao Yang, Andy J. Ma, Pong C. Yuen

Research output: Chapter in book/report/conference proceedingConference contributionpeer-review

Abstract

Developing a Universal Lesion Detector (ULD) that can detect various types of lesions from the whole body is of great importance for early diagnosis and timely treatment. Recently, deep neural networks have been applied for the ULD task, and existing methods assume that all the training samples are well-annotated. However, the partial label problem is unavoidable when curating large-scale datasets, where only a part of instances are annotated in each image. To address this issue, we propose a novel segmentation-assisted model, where an additional semantic segmentation branch with superpixel-guided selective loss is introduced to assist the conventional detection branch. The segmentation branch and the detection branch help each other to find unlabeled lesions with a mutual-mining strategy, and then the mined suspicious lesions are ignored for fine-tuning to reduce their negative impacts. Evaluation experiments on the DeepLesion dataset demonstrate that our proposed method allows the baseline detector to boost its average precision by 13%, outperforming the previous state-of-the-art methods.
Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2021
Subtitle of host publication24th International Conference, Strasbourg, France, September 27 – October 1, 2021, Proceedings, Part V
EditorsMarleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, Caroline Essert
PublisherSpringer Cham
Pages117–127
Number of pages11
Edition1st
ISBN (Electronic)9783030872403
ISBN (Print)9783030872397
DOIs
Publication statusPublished - 21 Sep 2021
Event24th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2021 - Strasbourg, France
Duration: 27 Sep 20211 Oct 2021

Publication series

NameLecture Notes in Computer Science
Volume12905
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameImage Processing, Computer Vision, Pattern Recognition, and Graphics
Volume12905

Conference

Conference24th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2021
Country/TerritoryFrance
CityStrasbourg
Period27/09/211/10/21

Fingerprint

Dive into the research topics of 'A Segmentation-Assisted Model for Universal Lesion Detection with Partial Labels'. Together they form a unique fingerprint.

Cite this