Transductive local fisher discriminant analysis for gene expression profile-based cancer classification

Danping Li, Lei Wang, Jiajun Wang, Zhong Xue, Stephen T.C. Wong

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

4 Scopus citations

Abstract

Gene expression profiles provide hidden biological knowledge and key information that can be used to distinguish different types of cancer. Due to their high dimensionality and redundancy, gene expression data are often preprocessed by dimensionality reduction (DR) methods. Conventional supervised DR methods use only labeled samples to train the model, leading to a limited performance due to small number of labeled samples in the real world. This paper proposes a transductive local Fisher discriminant analysis (TLFDA) method that uses the available unlabeled data in the learning process. On the one hand, the label information is utilized to maximize the inter-class distance in the embedding space. On the other hand, the local structural information of all data samples is taken into consideration to maintain the smoothness property. In this way, the TLFDA provides more discriminative power than state-of-the-art supervised or semi-supervised DR methods, even when the number of labeled samples is very limited. Our experimental results on benchmark GCM and Acute Leukemia datasets show its promising performance on gene expression profile-based cancer classification.

Original languageEnglish (US)
Title of host publication2017 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages49-52
Number of pages4
ISBN (Electronic)9781509041794
DOIs
StatePublished - Apr 11 2017
Event4th IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2017 - Orlando, United States
Duration: Feb 16 2017Feb 19 2017

Publication series

Name2017 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2017

Conference

Conference4th IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2017
CountryUnited States
CityOrlando
Period2/16/172/19/17

ASJC Scopus subject areas

  • Health Informatics
  • Computer Science Applications
  • Biomedical Engineering

Fingerprint Dive into the research topics of 'Transductive local fisher discriminant analysis for gene expression profile-based cancer classification'. Together they form a unique fingerprint.

Cite this