Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
  • Applied Spectroscopy
  • Vol. 52,
  • Issue 6,
  • pp. 833-839
  • (1998)

Identification of Colonic Dysplasia and Neoplasia by Diffuse Reflectance Spectroscopy and Pattern Recognition Techniques

Not Accessible

Your library or personal account may give you access


Diffuse reflectance spectroscopy of colonic tissue was employed to determine whether the spectra can be used to distinguish between neoplastic and non-neoplastic tissue in vivo. A total of 224 spectra were obtained in the wavelength range of 350-800 nm from 107 non-neoplastic tissue samples (84 normal mucosa, 23 hyperplastic polyps) and 53 neoplastic tissue samples (44 adenomatous polyps, 9 adenocarcinomas). Pattern recognition algorithms including multiple linear regression (MLR), linear discriminant analysis (LDA), and backpropagating neural network (BNN) were used to distinguish between the two tissue classes. The spectra were randomly separated into training and prediction sets for data analyses. The mean predictive accuracies of distinguishing neoplastic tissue from non-neoplastic tissue with MLR, LDA, and BNN were 85, 82, and 85%, respectively. In a similar fashion, the more clinically relevant problem of distinguishing adenomatous polyps from hyperplastic polyps was assessed. The mean predictive accuracies of distinguishing adenomatous polyps from hyperplastic polyps with MLR, LDA, and BNN were 85, 81, and 82%, respectively. The major spectral differences between tissues were attributed to changes in blood volume, oxygen saturation of hemoglobin, mean vessel depth within tissue, and tissue scattering.

PDF Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
Login to access Optica Member Subscription

Select as filters

Select Topics Cancel
© Copyright 2022 | Optica Publishing Group. All Rights Reserved