A novel online boosting algorithm for automatic anatomy detection

Nima Tajbakhsh, Hong Wu, Wenzhe Xue, Michael Gotway, Jianming Liang

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper presents a novel online learning method for automatically detecting anatomic structures in medical images. Conventional off-line learning methods require collecting a complete set of representative samples prior to training a detector. Once the detector is trained, its performance is fixed. To improve the performance, the detector must be completely retrained, demanding the maintenance of historical training samples. Our proposed online approach eliminates the need for storing historical training samples and is capable of continually improving performance with new samples. We evaluate our approach with three distinct thoracic structures, demonstrating that our approach yields performance competitive with the off-line approach. Furthermore, we investigate the properties of our proposed method in comparison with an online learning method suggested by Grabner and Bischof (IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2006, vol. 1, pp. 260-267, 2006), which is the state of the art, indicating that our proposed method runs faster, offers more stability, improves handling of "catastrophic forgetting", and simultaneously achieves a satisfactory level of adaptability. The enhanced performance is attributed to our novel online learning structure coupled with more accurate weaker learners based on histograms.

Original languageEnglish (US)
Pages (from-to)1359-1370
Number of pages12
JournalMachine Vision and Applications
Volume24
Issue number7
DOIs
StatePublished - 2013

Fingerprint

Detectors
Computer vision
Pattern recognition

Keywords

  • Anatomy detection
  • Histogram
  • Kalman filter
  • Online boosting

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Software
  • Computer Science Applications

Cite this

A novel online boosting algorithm for automatic anatomy detection. / Tajbakhsh, Nima; Wu, Hong; Xue, Wenzhe; Gotway, Michael; Liang, Jianming.

In: Machine Vision and Applications, Vol. 24, No. 7, 2013, p. 1359-1370.

Research output: Contribution to journalArticle

Tajbakhsh, Nima ; Wu, Hong ; Xue, Wenzhe ; Gotway, Michael ; Liang, Jianming. / A novel online boosting algorithm for automatic anatomy detection. In: Machine Vision and Applications. 2013 ; Vol. 24, No. 7. pp. 1359-1370.
@article{87076bca1b35483e85455dc9cddd5832,
title = "A novel online boosting algorithm for automatic anatomy detection",
abstract = "This paper presents a novel online learning method for automatically detecting anatomic structures in medical images. Conventional off-line learning methods require collecting a complete set of representative samples prior to training a detector. Once the detector is trained, its performance is fixed. To improve the performance, the detector must be completely retrained, demanding the maintenance of historical training samples. Our proposed online approach eliminates the need for storing historical training samples and is capable of continually improving performance with new samples. We evaluate our approach with three distinct thoracic structures, demonstrating that our approach yields performance competitive with the off-line approach. Furthermore, we investigate the properties of our proposed method in comparison with an online learning method suggested by Grabner and Bischof (IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2006, vol. 1, pp. 260-267, 2006), which is the state of the art, indicating that our proposed method runs faster, offers more stability, improves handling of {"}catastrophic forgetting{"}, and simultaneously achieves a satisfactory level of adaptability. The enhanced performance is attributed to our novel online learning structure coupled with more accurate weaker learners based on histograms.",
keywords = "Anatomy detection, Histogram, Kalman filter, Online boosting",
author = "Nima Tajbakhsh and Hong Wu and Wenzhe Xue and Michael Gotway and Jianming Liang",
year = "2013",
doi = "10.1007/s00138-012-0455-z",
language = "English (US)",
volume = "24",
pages = "1359--1370",
journal = "Machine Vision and Applications",
issn = "0932-8092",
publisher = "Springer Verlag",
number = "7",

}

TY - JOUR

T1 - A novel online boosting algorithm for automatic anatomy detection

AU - Tajbakhsh, Nima

AU - Wu, Hong

AU - Xue, Wenzhe

AU - Gotway, Michael

AU - Liang, Jianming

PY - 2013

Y1 - 2013

N2 - This paper presents a novel online learning method for automatically detecting anatomic structures in medical images. Conventional off-line learning methods require collecting a complete set of representative samples prior to training a detector. Once the detector is trained, its performance is fixed. To improve the performance, the detector must be completely retrained, demanding the maintenance of historical training samples. Our proposed online approach eliminates the need for storing historical training samples and is capable of continually improving performance with new samples. We evaluate our approach with three distinct thoracic structures, demonstrating that our approach yields performance competitive with the off-line approach. Furthermore, we investigate the properties of our proposed method in comparison with an online learning method suggested by Grabner and Bischof (IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2006, vol. 1, pp. 260-267, 2006), which is the state of the art, indicating that our proposed method runs faster, offers more stability, improves handling of "catastrophic forgetting", and simultaneously achieves a satisfactory level of adaptability. The enhanced performance is attributed to our novel online learning structure coupled with more accurate weaker learners based on histograms.

AB - This paper presents a novel online learning method for automatically detecting anatomic structures in medical images. Conventional off-line learning methods require collecting a complete set of representative samples prior to training a detector. Once the detector is trained, its performance is fixed. To improve the performance, the detector must be completely retrained, demanding the maintenance of historical training samples. Our proposed online approach eliminates the need for storing historical training samples and is capable of continually improving performance with new samples. We evaluate our approach with three distinct thoracic structures, demonstrating that our approach yields performance competitive with the off-line approach. Furthermore, we investigate the properties of our proposed method in comparison with an online learning method suggested by Grabner and Bischof (IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2006, vol. 1, pp. 260-267, 2006), which is the state of the art, indicating that our proposed method runs faster, offers more stability, improves handling of "catastrophic forgetting", and simultaneously achieves a satisfactory level of adaptability. The enhanced performance is attributed to our novel online learning structure coupled with more accurate weaker learners based on histograms.

KW - Anatomy detection

KW - Histogram

KW - Kalman filter

KW - Online boosting

UR - http://www.scopus.com/inward/record.url?scp=84885306958&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885306958&partnerID=8YFLogxK

U2 - 10.1007/s00138-012-0455-z

DO - 10.1007/s00138-012-0455-z

M3 - Article

AN - SCOPUS:84885306958

VL - 24

SP - 1359

EP - 1370

JO - Machine Vision and Applications

JF - Machine Vision and Applications

SN - 0932-8092

IS - 7

ER -