Unsupervised deep novelty detection: Application to muscle ultrasound and myositis screening

P. Burlina, N. Joshi, S. Billings, I. J. Wang, J. Albayda

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This study investigates unsupervised novelty detection (ND) for screening of rare myopathies and specifically myositis. To support this study we developed from the ground up a novel and fully annotated dataset consisting of 3586 images taken of eighty nine individuals obtained under informed consent during 2016-2017. We developed and compared performance for several ND methods leveraging deep feature embeddings, utilizing generative as well as discriminative deep learning approaches for embeddings, and using various novelty scores. We carried out several performance comparisons including with a clinician, supervised binary classification approaches, and a generative method, demonstrating that our best performing approach is competitive with human performance and other best of breed algorithms.

Original languageEnglish (US)
Title of host publicationISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
PublisherIEEE Computer Society
Pages1910-1914
Number of pages5
ISBN (Electronic)9781538636411
DOIs
StatePublished - Apr 2019
Event16th IEEE International Symposium on Biomedical Imaging, ISBI 2019 - Venice, Italy
Duration: Apr 8 2019Apr 11 2019

Publication series

NameProceedings - International Symposium on Biomedical Imaging
Volume2019-April
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Conference

Conference16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
CountryItaly
CityVenice
Period4/8/194/11/19

Fingerprint

Myositis
Muscle
Screening
Ultrasonics
Muscles
Muscular Diseases
Informed Consent
Learning
Deep learning
Datasets

Keywords

  • Deep embeddings
  • Myopathy
  • Myositis
  • Novelty detection
  • Unsupervised learning

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Burlina, P., Joshi, N., Billings, S., Wang, I. J., & Albayda, J. (2019). Unsupervised deep novelty detection: Application to muscle ultrasound and myositis screening. In ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging (pp. 1910-1914). [8759565] (Proceedings - International Symposium on Biomedical Imaging; Vol. 2019-April). IEEE Computer Society. https://doi.org/10.1109/ISBI.2019.8759565

Unsupervised deep novelty detection : Application to muscle ultrasound and myositis screening. / Burlina, P.; Joshi, N.; Billings, S.; Wang, I. J.; Albayda, J.

ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2019. p. 1910-1914 8759565 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2019-April).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Burlina, P, Joshi, N, Billings, S, Wang, IJ & Albayda, J 2019, Unsupervised deep novelty detection: Application to muscle ultrasound and myositis screening. in ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging., 8759565, Proceedings - International Symposium on Biomedical Imaging, vol. 2019-April, IEEE Computer Society, pp. 1910-1914, 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019, Venice, Italy, 4/8/19. https://doi.org/10.1109/ISBI.2019.8759565
Burlina P, Joshi N, Billings S, Wang IJ, Albayda J. Unsupervised deep novelty detection: Application to muscle ultrasound and myositis screening. In ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society. 2019. p. 1910-1914. 8759565. (Proceedings - International Symposium on Biomedical Imaging). https://doi.org/10.1109/ISBI.2019.8759565
Burlina, P. ; Joshi, N. ; Billings, S. ; Wang, I. J. ; Albayda, J. / Unsupervised deep novelty detection : Application to muscle ultrasound and myositis screening. ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2019. pp. 1910-1914 (Proceedings - International Symposium on Biomedical Imaging).
@inproceedings{627fc7d3f0a143cf91543d833d5c63de,
title = "Unsupervised deep novelty detection: Application to muscle ultrasound and myositis screening",
abstract = "This study investigates unsupervised novelty detection (ND) for screening of rare myopathies and specifically myositis. To support this study we developed from the ground up a novel and fully annotated dataset consisting of 3586 images taken of eighty nine individuals obtained under informed consent during 2016-2017. We developed and compared performance for several ND methods leveraging deep feature embeddings, utilizing generative as well as discriminative deep learning approaches for embeddings, and using various novelty scores. We carried out several performance comparisons including with a clinician, supervised binary classification approaches, and a generative method, demonstrating that our best performing approach is competitive with human performance and other best of breed algorithms.",
keywords = "Deep embeddings, Myopathy, Myositis, Novelty detection, Unsupervised learning",
author = "P. Burlina and N. Joshi and S. Billings and Wang, {I. J.} and J. Albayda",
year = "2019",
month = "4",
doi = "10.1109/ISBI.2019.8759565",
language = "English (US)",
series = "Proceedings - International Symposium on Biomedical Imaging",
publisher = "IEEE Computer Society",
pages = "1910--1914",
booktitle = "ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Unsupervised deep novelty detection

T2 - Application to muscle ultrasound and myositis screening

AU - Burlina, P.

AU - Joshi, N.

AU - Billings, S.

AU - Wang, I. J.

AU - Albayda, J.

PY - 2019/4

Y1 - 2019/4

N2 - This study investigates unsupervised novelty detection (ND) for screening of rare myopathies and specifically myositis. To support this study we developed from the ground up a novel and fully annotated dataset consisting of 3586 images taken of eighty nine individuals obtained under informed consent during 2016-2017. We developed and compared performance for several ND methods leveraging deep feature embeddings, utilizing generative as well as discriminative deep learning approaches for embeddings, and using various novelty scores. We carried out several performance comparisons including with a clinician, supervised binary classification approaches, and a generative method, demonstrating that our best performing approach is competitive with human performance and other best of breed algorithms.

AB - This study investigates unsupervised novelty detection (ND) for screening of rare myopathies and specifically myositis. To support this study we developed from the ground up a novel and fully annotated dataset consisting of 3586 images taken of eighty nine individuals obtained under informed consent during 2016-2017. We developed and compared performance for several ND methods leveraging deep feature embeddings, utilizing generative as well as discriminative deep learning approaches for embeddings, and using various novelty scores. We carried out several performance comparisons including with a clinician, supervised binary classification approaches, and a generative method, demonstrating that our best performing approach is competitive with human performance and other best of breed algorithms.

KW - Deep embeddings

KW - Myopathy

KW - Myositis

KW - Novelty detection

KW - Unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85073906901&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073906901&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2019.8759565

DO - 10.1109/ISBI.2019.8759565

M3 - Conference contribution

AN - SCOPUS:85073906901

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 1910

EP - 1914

BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

ER -