Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma

Mehmet Gonen, Aydin Ulaş, Peter Schüffler, Umberto Castellani, Vittorio Murino

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

1 Citation (Scopus)

Abstract

In kernel-based machine learning algorithms, we can learn a combination of different kernel functions in order to obtain a similarity measure that better matches the underlying problem instead of using a single fixed kernel function. This approach is called multiple kernel learning (MKL). In this paper, we formulate a nonlinear MKL variant and apply it for nuclei classification in tissue microarray images of renal cell carcinoma (RCC). The proposed variant is tested on several feature representations extracted from the automatically segmented nuclei. We compare our results with single-kernel support vector machines trained on each feature representation separately and three linear MKL algorithms from the literature. We demonstrate that our variant obtains more accurate classifiers than competing algorithms for RCC detection by combining information from different feature representations nonlinearly.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages250-260
Number of pages11
Volume7005 LNCS
DOIs
StatePublished - 2011
Externally publishedYes
Event1st International Workshop on Similarity-Based Pattern Recognition, SIMBAD 2011 - Venice, Italy
Duration: Sep 28 2011Sep 30 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7005 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other1st International Workshop on Similarity-Based Pattern Recognition, SIMBAD 2011
CountryItaly
CityVenice
Period9/28/119/30/11

Fingerprint

Learning algorithms
Nucleus
Cells
kernel
Cell
Microarrays
Support vector machines
Learning systems
Kernel Function
Classifiers
Tissue
Learning Algorithm
Similarity Measure
Microarray
Support Vector Machine
Machine Learning
Classifier
Demonstrate
Learning

Keywords

  • multiple kernel learning
  • renal cell carcinoma
  • support vector machines

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Gonen, M., Ulaş, A., Schüffler, P., Castellani, U., & Murino, V. (2011). Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7005 LNCS, pp. 250-260). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7005 LNCS). https://doi.org/10.1007/978-3-642-24471-1_18

Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma. / Gonen, Mehmet; Ulaş, Aydin; Schüffler, Peter; Castellani, Umberto; Murino, Vittorio.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7005 LNCS 2011. p. 250-260 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7005 LNCS).

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

Gonen, M, Ulaş, A, Schüffler, P, Castellani, U & Murino, V 2011, Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7005 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7005 LNCS, pp. 250-260, 1st International Workshop on Similarity-Based Pattern Recognition, SIMBAD 2011, Venice, Italy, 9/28/11. https://doi.org/10.1007/978-3-642-24471-1_18
Gonen M, Ulaş A, Schüffler P, Castellani U, Murino V. Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7005 LNCS. 2011. p. 250-260. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-24471-1_18
Gonen, Mehmet ; Ulaş, Aydin ; Schüffler, Peter ; Castellani, Umberto ; Murino, Vittorio. / Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7005 LNCS 2011. pp. 250-260 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{a5fb8a8eee084609be9cbf4b2a4d9d9e,
title = "Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma",
abstract = "In kernel-based machine learning algorithms, we can learn a combination of different kernel functions in order to obtain a similarity measure that better matches the underlying problem instead of using a single fixed kernel function. This approach is called multiple kernel learning (MKL). In this paper, we formulate a nonlinear MKL variant and apply it for nuclei classification in tissue microarray images of renal cell carcinoma (RCC). The proposed variant is tested on several feature representations extracted from the automatically segmented nuclei. We compare our results with single-kernel support vector machines trained on each feature representation separately and three linear MKL algorithms from the literature. We demonstrate that our variant obtains more accurate classifiers than competing algorithms for RCC detection by combining information from different feature representations nonlinearly.",
keywords = "multiple kernel learning, renal cell carcinoma, support vector machines",
author = "Mehmet Gonen and Aydin Ulaş and Peter Sch{\"u}ffler and Umberto Castellani and Vittorio Murino",
year = "2011",
doi = "10.1007/978-3-642-24471-1_18",
language = "English (US)",
isbn = "9783642244704",
volume = "7005 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "250--260",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Combining data sources nonlinearly for cell nucleus classification of renal cell carcinoma

AU - Gonen, Mehmet

AU - Ulaş, Aydin

AU - Schüffler, Peter

AU - Castellani, Umberto

AU - Murino, Vittorio

PY - 2011

Y1 - 2011

N2 - In kernel-based machine learning algorithms, we can learn a combination of different kernel functions in order to obtain a similarity measure that better matches the underlying problem instead of using a single fixed kernel function. This approach is called multiple kernel learning (MKL). In this paper, we formulate a nonlinear MKL variant and apply it for nuclei classification in tissue microarray images of renal cell carcinoma (RCC). The proposed variant is tested on several feature representations extracted from the automatically segmented nuclei. We compare our results with single-kernel support vector machines trained on each feature representation separately and three linear MKL algorithms from the literature. We demonstrate that our variant obtains more accurate classifiers than competing algorithms for RCC detection by combining information from different feature representations nonlinearly.

AB - In kernel-based machine learning algorithms, we can learn a combination of different kernel functions in order to obtain a similarity measure that better matches the underlying problem instead of using a single fixed kernel function. This approach is called multiple kernel learning (MKL). In this paper, we formulate a nonlinear MKL variant and apply it for nuclei classification in tissue microarray images of renal cell carcinoma (RCC). The proposed variant is tested on several feature representations extracted from the automatically segmented nuclei. We compare our results with single-kernel support vector machines trained on each feature representation separately and three linear MKL algorithms from the literature. We demonstrate that our variant obtains more accurate classifiers than competing algorithms for RCC detection by combining information from different feature representations nonlinearly.

KW - multiple kernel learning

KW - renal cell carcinoma

KW - support vector machines

UR - http://www.scopus.com/inward/record.url?scp=80053367605&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053367605&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-24471-1_18

DO - 10.1007/978-3-642-24471-1_18

M3 - Conference contribution

AN - SCOPUS:80053367605

SN - 9783642244704

VL - 7005 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 250

EP - 260

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -