Near-infrared Raman spectroscopy for in-vivo diagnosis of cervical dysplasia - A probability-based multi-class diagnostic algorithm

Shovan K. Majumder, Elizabeth Kanter, Amy Robichaux Viehoever, Howard Jones, Anita Mahadevan-Jansen

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

6 Scopus citations

Abstract

We report the development of a probability-based multi-class diagnostic algorithm to simultaneously distinguish high-grade dysplasia from low-grade dysplasia, squamous metaplasia as well as normal human cervical tissues using near-infrared Raman spectra acquired in-vivo from the cervix of patients at the Vanderbilt University Medical Center. Extraction of diagnostic features from the Raman spectra uses the recently formulated theory of nonlinear Maximum Representation and Discrimination Feature (MRDF), and classification into respective tissue categories is based on the theory of Sparse Multinomial Logistic Regression (SMLR), a recent Bayesian machine-learning framework of statistical pattern recognition. The algorithm based on MRDF and SMLR was found to provide very good diagnostic performance with a predictive accuracy of ∼90% based on leave-one-out cross validation in classifying the tissue Raman spectra into the four different classes, using histology as the "gold standard". The inherently multi-class nature of the algorithm facilitates a rapid and simultaneous classification of tissue spectra into various tissue categories without the need to train and heuristically combine multiple binary classifiers. Further, the probabilistic framework of the algorithm makes it possible to predict the posterior probability of class membership in discriminating the different tissue types.

Original languageEnglish
Title of host publicationAdvanced Biomedical and Clinical Diagnostic Systems V
DOIs
StatePublished - Apr 30 2007
Externally publishedYes
EventAdvanced Biomedical and Clinical Diagnostic Systems V - San Jose, CA, United States
Duration: Jan 21 2007Jan 27 2007

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume6430
ISSN (Print)1605-7422

Conference

ConferenceAdvanced Biomedical and Clinical Diagnostic Systems V
CountryUnited States
CitySan Jose, CA
Period01/21/0701/27/07

Keywords

  • Cervical dysplasia
  • Maximum Representation and Discrimination Feature (MRDF)
  • Multi-class diagnostic algorithm
  • Posterior probability
  • Raman spectroscopy
  • Sparse Multinomial Logistic Regression (SMLR)

Fingerprint Dive into the research topics of 'Near-infrared Raman spectroscopy for in-vivo diagnosis of cervical dysplasia - A probability-based multi-class diagnostic algorithm'. Together they form a unique fingerprint.

  • Cite this

    Majumder, S. K., Kanter, E., Viehoever, A. R., Jones, H., & Mahadevan-Jansen, A. (2007). Near-infrared Raman spectroscopy for in-vivo diagnosis of cervical dysplasia - A probability-based multi-class diagnostic algorithm. In Advanced Biomedical and Clinical Diagnostic Systems V [64300Q] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 6430). https://doi.org/10.1117/12.724873