HEp-2 Staining Pattern Classification

Petter Strandmark; Johannes Ulén; Fredrik Kahl

HEp-2 Staining Pattern Classification

Petter Strandmark, Johannes Ulén, Fredrik Kahl

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

174 Downloads (Pure)

Abstract

Classifying images of HEp-2 cells from indirect immunofluorescence has important clinical applications. We have developed an automatic method based on random forests that classifies an HEp-2 cell image into one of six classes. The method is applied to the data set of the ICPR 2012 contest. The previously obtained best accuracy is 79.3% for this data set, whereas we obtain an accuracy of 97.4%. The key to our result is due to carefully designed feature descriptors for multiple level sets of the image intensity. These features characterize both the appearance and the shape of the cell image in a robust manner.

Original language	English
Title of host publication	Pattern Recognition (ICPR), 2012 21st International Conference on
Publisher	IEEE - Institute of Electrical and Electronics Engineers Inc.
Number of pages	4
ISBN (Print)	978-1-4673-2216-4
Publication status	Published - 2012
Event	21st International Conference on Pattern Recognition (ICPR 2012) - Tsukuba, Japan Duration: 2012 Nov 11 → 2012 Nov 15

Conference

Conference	21st International Conference on Pattern Recognition (ICPR 2012)
Country/Territory	Japan
City	Tsukuba
Period	2012/11/11 → 2012/11/15

Subject classification (UKÄ)

Computer graphics and computer vision
Mathematical Sciences

Access to Document

hep2

Cite this

@inproceedings{15a4fd9feadc4dd6951205dd59efc9fa,

title = "HEp-2 Staining Pattern Classification",

abstract = "Classifying images of HEp-2 cells from indirect immunofluorescence has important clinical applications. We have developed an automatic method based on random forests that classifies an HEp-2 cell image into one of six classes. The method is applied to the data set of the ICPR 2012 contest. The previously obtained best accuracy is 79.3\% for this data set, whereas we obtain an accuracy of 97.4\%. The key to our result is due to carefully designed feature descriptors for multiple level sets of the image intensity. These features characterize both the appearance and the shape of the cell image in a robust manner.",

author = "Petter Strandmark and Johannes Ul{\'e}n and Fredrik Kahl",

year = "2012",

language = "English",

isbn = "978-1-4673-2216-4",

booktitle = "Pattern Recognition (ICPR), 2012 21st International Conference on",

publisher = "IEEE - Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "21st International Conference on Pattern Recognition (ICPR 2012) ; Conference date: 11-11-2012 Through 15-11-2012",

}

Strandmark, P, Ulén, J & Kahl, F 2012, HEp-2 Staining Pattern Classification. in Pattern Recognition (ICPR), 2012 21st International Conference on. IEEE - Institute of Electrical and Electronics Engineers Inc., 21st International Conference on Pattern Recognition (ICPR 2012), Tsukuba, Japan, 2012/11/11. <http://www.maths.lth.se/vision/publdb/reports/pdf/strandmark-ulen-etal-icpr.pdf>

TY - GEN

T1 - HEp-2 Staining Pattern Classification

AU - Strandmark, Petter

AU - Ulén, Johannes

AU - Kahl, Fredrik

PY - 2012

Y1 - 2012

N2 - Classifying images of HEp-2 cells from indirect immunofluorescence has important clinical applications. We have developed an automatic method based on random forests that classifies an HEp-2 cell image into one of six classes. The method is applied to the data set of the ICPR 2012 contest. The previously obtained best accuracy is 79.3% for this data set, whereas we obtain an accuracy of 97.4%. The key to our result is due to carefully designed feature descriptors for multiple level sets of the image intensity. These features characterize both the appearance and the shape of the cell image in a robust manner.

AB - Classifying images of HEp-2 cells from indirect immunofluorescence has important clinical applications. We have developed an automatic method based on random forests that classifies an HEp-2 cell image into one of six classes. The method is applied to the data set of the ICPR 2012 contest. The previously obtained best accuracy is 79.3% for this data set, whereas we obtain an accuracy of 97.4%. The key to our result is due to carefully designed feature descriptors for multiple level sets of the image intensity. These features characterize both the appearance and the shape of the cell image in a robust manner.

UR - https://www.scopus.com/pages/publications/84874579172

M3 - Paper in conference proceeding

SN - 978-1-4673-2216-4

BT - Pattern Recognition (ICPR), 2012 21st International Conference on

PB - IEEE - Institute of Electrical and Electronics Engineers Inc.

T2 - 21st International Conference on Pattern Recognition (ICPR 2012)

Y2 - 11 November 2012 through 15 November 2012

ER -

HEp-2 Staining Pattern Classification

Abstract

Conference

Subject classification (UKÄ)

Access to Document

Other files and links

Fingerprint

Mathematical Imaging Group

Cite this

HEp-2 Staining Pattern Classification

Abstract

Conference

Subject classification (UKÄ)

Access to Document

Other files and links

Fingerprint

Projects

Mathematical Imaging Group

Cite this