Label-free reflectance hyperspectral imaging for tumor margin assessment: a pilot study on surgical specimens of cancer patients

Baowei Fei, Guolan Lu, Xu Wang, Hongzheng Zhang, James V. Little, Mihir R. Patel, Christopher C. Griffith, Mark W. El-Diery, Amy Y. Chen

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

A label-free, hyperspectral imaging (HSI) approach has been proposed for tumor margin assessment. HSI data, i.e., hypercube (x,y,λ), consist of a series of high-resolution images of the same field of view that are acquired at different wavelengths. Every pixel on an HSI image has an optical spectrum. In this pilot clinical study, a pipeline of a machine-learning-based quantification method for HSI data was implemented and evaluated in patient specimens. Spectral features from HSI data were used for the classification of cancer and normal tissue. Surgical tissue specimens were collected from 16 human patients who underwent head and neck (H&N) cancer surgery. HSI, autofluorescence images, and fluorescence images with 2-deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]-D-glucose (2-NBDG) and proflavine were acquired from each specimen. Digitized histologic slides were examined by an H&N pathologist. The HSI and classification method were able to distinguish between cancer and normal tissue from the oral cavity with an average accuracy of 90%±8%, sensitivity of 89%±9%, and specificity of 91%±6%. For tissue specimens from the thyroid, the method achieved an average accuracy of 94%±6%, sensitivity of 94%±6%, and specificity of 95%±6%. HSI outperformed autofluorescence imaging or fluorescence imaging with vital dye (2-NBDG or proflavine). This study demonstrated the feasibility of label-free, HSI for tumor margin assessment in surgical tissue specimens of H&N cancer patients. Further development of the HSI technology is warranted for its application in image-guided surgery.

Original languageEnglish (US)
Pages (from-to)1-7
Number of pages7
JournalJournal of biomedical optics
Volume22
Issue number8
DOIs
StatePublished - Aug 1 2017
Externally publishedYes

Fingerprint

Labels
Tumors
margins
tumors
cancer
reflectance
surgery
Tissue
Proflavine
fluorescence
machine learning
sensitivity
chutes
glucose
Surgery
field of view
optical spectrum
dyes
Fluorescence
pixels

Keywords

  • cancer detection
  • head and neck cancer
  • hyperspectral imaging
  • image classification
  • image quantification
  • image-guided surgery
  • label-free
  • tumor margin assessment

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Biomedical Engineering

Cite this

Label-free reflectance hyperspectral imaging for tumor margin assessment : a pilot study on surgical specimens of cancer patients. / Fei, Baowei; Lu, Guolan; Wang, Xu; Zhang, Hongzheng; Little, James V.; Patel, Mihir R.; Griffith, Christopher C.; El-Diery, Mark W.; Chen, Amy Y.

In: Journal of biomedical optics, Vol. 22, No. 8, 01.08.2017, p. 1-7.

Research output: Contribution to journalArticle

Fei, Baowei ; Lu, Guolan ; Wang, Xu ; Zhang, Hongzheng ; Little, James V. ; Patel, Mihir R. ; Griffith, Christopher C. ; El-Diery, Mark W. ; Chen, Amy Y. / Label-free reflectance hyperspectral imaging for tumor margin assessment : a pilot study on surgical specimens of cancer patients. In: Journal of biomedical optics. 2017 ; Vol. 22, No. 8. pp. 1-7.
@article{98cdeef323794273b11046b1d2678cfe,
title = "Label-free reflectance hyperspectral imaging for tumor margin assessment: a pilot study on surgical specimens of cancer patients",
abstract = "A label-free, hyperspectral imaging (HSI) approach has been proposed for tumor margin assessment. HSI data, i.e., hypercube (x,y,λ), consist of a series of high-resolution images of the same field of view that are acquired at different wavelengths. Every pixel on an HSI image has an optical spectrum. In this pilot clinical study, a pipeline of a machine-learning-based quantification method for HSI data was implemented and evaluated in patient specimens. Spectral features from HSI data were used for the classification of cancer and normal tissue. Surgical tissue specimens were collected from 16 human patients who underwent head and neck (H&N) cancer surgery. HSI, autofluorescence images, and fluorescence images with 2-deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]-D-glucose (2-NBDG) and proflavine were acquired from each specimen. Digitized histologic slides were examined by an H&N pathologist. The HSI and classification method were able to distinguish between cancer and normal tissue from the oral cavity with an average accuracy of 90{\%}±8{\%}, sensitivity of 89{\%}±9{\%}, and specificity of 91{\%}±6{\%}. For tissue specimens from the thyroid, the method achieved an average accuracy of 94{\%}±6{\%}, sensitivity of 94{\%}±6{\%}, and specificity of 95{\%}±6{\%}. HSI outperformed autofluorescence imaging or fluorescence imaging with vital dye (2-NBDG or proflavine). This study demonstrated the feasibility of label-free, HSI for tumor margin assessment in surgical tissue specimens of H&N cancer patients. Further development of the HSI technology is warranted for its application in image-guided surgery.",
keywords = "cancer detection, head and neck cancer, hyperspectral imaging, image classification, image quantification, image-guided surgery, label-free, tumor margin assessment",
author = "Baowei Fei and Guolan Lu and Xu Wang and Hongzheng Zhang and Little, {James V.} and Patel, {Mihir R.} and Griffith, {Christopher C.} and El-Diery, {Mark W.} and Chen, {Amy Y.}",
year = "2017",
month = "8",
day = "1",
doi = "10.1117/1.JBO.22.8.086009",
language = "English (US)",
volume = "22",
pages = "1--7",
journal = "Journal of Biomedical Optics",
issn = "1083-3668",
publisher = "SPIE",
number = "8",

}

TY - JOUR

T1 - Label-free reflectance hyperspectral imaging for tumor margin assessment

T2 - a pilot study on surgical specimens of cancer patients

AU - Fei, Baowei

AU - Lu, Guolan

AU - Wang, Xu

AU - Zhang, Hongzheng

AU - Little, James V.

AU - Patel, Mihir R.

AU - Griffith, Christopher C.

AU - El-Diery, Mark W.

AU - Chen, Amy Y.

PY - 2017/8/1

Y1 - 2017/8/1

N2 - A label-free, hyperspectral imaging (HSI) approach has been proposed for tumor margin assessment. HSI data, i.e., hypercube (x,y,λ), consist of a series of high-resolution images of the same field of view that are acquired at different wavelengths. Every pixel on an HSI image has an optical spectrum. In this pilot clinical study, a pipeline of a machine-learning-based quantification method for HSI data was implemented and evaluated in patient specimens. Spectral features from HSI data were used for the classification of cancer and normal tissue. Surgical tissue specimens were collected from 16 human patients who underwent head and neck (H&N) cancer surgery. HSI, autofluorescence images, and fluorescence images with 2-deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]-D-glucose (2-NBDG) and proflavine were acquired from each specimen. Digitized histologic slides were examined by an H&N pathologist. The HSI and classification method were able to distinguish between cancer and normal tissue from the oral cavity with an average accuracy of 90%±8%, sensitivity of 89%±9%, and specificity of 91%±6%. For tissue specimens from the thyroid, the method achieved an average accuracy of 94%±6%, sensitivity of 94%±6%, and specificity of 95%±6%. HSI outperformed autofluorescence imaging or fluorescence imaging with vital dye (2-NBDG or proflavine). This study demonstrated the feasibility of label-free, HSI for tumor margin assessment in surgical tissue specimens of H&N cancer patients. Further development of the HSI technology is warranted for its application in image-guided surgery.

AB - A label-free, hyperspectral imaging (HSI) approach has been proposed for tumor margin assessment. HSI data, i.e., hypercube (x,y,λ), consist of a series of high-resolution images of the same field of view that are acquired at different wavelengths. Every pixel on an HSI image has an optical spectrum. In this pilot clinical study, a pipeline of a machine-learning-based quantification method for HSI data was implemented and evaluated in patient specimens. Spectral features from HSI data were used for the classification of cancer and normal tissue. Surgical tissue specimens were collected from 16 human patients who underwent head and neck (H&N) cancer surgery. HSI, autofluorescence images, and fluorescence images with 2-deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]-D-glucose (2-NBDG) and proflavine were acquired from each specimen. Digitized histologic slides were examined by an H&N pathologist. The HSI and classification method were able to distinguish between cancer and normal tissue from the oral cavity with an average accuracy of 90%±8%, sensitivity of 89%±9%, and specificity of 91%±6%. For tissue specimens from the thyroid, the method achieved an average accuracy of 94%±6%, sensitivity of 94%±6%, and specificity of 95%±6%. HSI outperformed autofluorescence imaging or fluorescence imaging with vital dye (2-NBDG or proflavine). This study demonstrated the feasibility of label-free, HSI for tumor margin assessment in surgical tissue specimens of H&N cancer patients. Further development of the HSI technology is warranted for its application in image-guided surgery.

KW - cancer detection

KW - head and neck cancer

KW - hyperspectral imaging

KW - image classification

KW - image quantification

KW - image-guided surgery

KW - label-free

KW - tumor margin assessment

UR - http://www.scopus.com/inward/record.url?scp=85044746228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044746228&partnerID=8YFLogxK

U2 - 10.1117/1.JBO.22.8.086009

DO - 10.1117/1.JBO.22.8.086009

M3 - Article

C2 - 28849631

AN - SCOPUS:85044746228

VL - 22

SP - 1

EP - 7

JO - Journal of Biomedical Optics

JF - Journal of Biomedical Optics

SN - 1083-3668

IS - 8

ER -