Deformable registration of histological cancer margins to gross hyperspectral images using demons

Martin Halicek, James V. Little, Xu Wang, Zhuo Georgia Chen, Mihir Patel, Christopher C. Griffith, Mark W. El-Deiry, Nabil F. Saba, Amy Y. Chen, Baowei Fei

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

4 Citations (Scopus)

Abstract

Hyperspectral imaging (HSI), a non-contact optical imaging technique, has been recently used along with machine learning technique to provide diagnostic information about ex-vivo surgical specimens for optical biopsy. The computer-Aided diagnostic approach requires accurate ground truths for both training and validation. This study details a processing pipeline for registering the cancer-normal margin from a digitized histological image to the gross-level HSI of a tissue specimen. Our work incorporates an initial affine and control-point registration followed by a deformable Demons-based registration of the moving mask obtained from the histological image to the fixed mask made from the HS image. To assess registration quality, Dice similarity coefficient (DSC) measures the image overlap, visual inspection is used to evaluate the margin, and average target registration error (TRE) of needle-bored holes measures the registration error between the histologic and HSI images. Excised tissue samples from seventeen patients, 11 head and neck squamous cell carcinoma (HNSCCa) and 6 thyroid carcinoma, were registered according to the proposed method. Three registered specimens are illustrated in this paper, which demonstrate the efficacy of the registration workflow. Further work is required to apply the technique to more patient data and investigate the ability of this procedure to produce suitable gold standards for machine learning validation.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2018
Subtitle of host publicationDigital Pathology
EditorsMetin N. Gurcan, John E. Tomaszewski
PublisherSPIE
ISBN (Electronic)9781510616516
DOIs
StatePublished - Jan 1 2018
Externally publishedYes
EventMedical Imaging 2018: Digital Pathology - Houston, United States
Duration: Feb 11 2018Feb 12 2018

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10581
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2018: Digital Pathology
CountryUnited States
CityHouston
Period2/11/182/12/18

Fingerprint

Masks
margins
cancer
Learning systems
Workflow
Optical Imaging
Thyroid Neoplasms
machine learning
Tissue
Needles
Neoplasms
Biopsy
masks
Pipelines
Inspection
ground truth
Imaging techniques
imaging techniques
needles
inspection

Keywords

  • Demons registration
  • Digital histology
  • Head and neck cancer surgery
  • Hyperspectral imaging

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Cite this

Halicek, M., Little, J. V., Wang, X., Chen, Z. G., Patel, M., Griffith, C. C., ... Fei, B. (2018). Deformable registration of histological cancer margins to gross hyperspectral images using demons. In M. N. Gurcan, & J. E. Tomaszewski (Eds.), Medical Imaging 2018: Digital Pathology [105810N] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10581). SPIE. https://doi.org/10.1117/12.2293165

Deformable registration of histological cancer margins to gross hyperspectral images using demons. / Halicek, Martin; Little, James V.; Wang, Xu; Chen, Zhuo Georgia; Patel, Mihir; Griffith, Christopher C.; El-Deiry, Mark W.; Saba, Nabil F.; Chen, Amy Y.; Fei, Baowei.

Medical Imaging 2018: Digital Pathology. ed. / Metin N. Gurcan; John E. Tomaszewski. SPIE, 2018. 105810N (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10581).

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

Halicek, M, Little, JV, Wang, X, Chen, ZG, Patel, M, Griffith, CC, El-Deiry, MW, Saba, NF, Chen, AY & Fei, B 2018, Deformable registration of histological cancer margins to gross hyperspectral images using demons. in MN Gurcan & JE Tomaszewski (eds), Medical Imaging 2018: Digital Pathology., 105810N, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 10581, SPIE, Medical Imaging 2018: Digital Pathology, Houston, United States, 2/11/18. https://doi.org/10.1117/12.2293165
Halicek M, Little JV, Wang X, Chen ZG, Patel M, Griffith CC et al. Deformable registration of histological cancer margins to gross hyperspectral images using demons. In Gurcan MN, Tomaszewski JE, editors, Medical Imaging 2018: Digital Pathology. SPIE. 2018. 105810N. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.2293165
Halicek, Martin ; Little, James V. ; Wang, Xu ; Chen, Zhuo Georgia ; Patel, Mihir ; Griffith, Christopher C. ; El-Deiry, Mark W. ; Saba, Nabil F. ; Chen, Amy Y. ; Fei, Baowei. / Deformable registration of histological cancer margins to gross hyperspectral images using demons. Medical Imaging 2018: Digital Pathology. editor / Metin N. Gurcan ; John E. Tomaszewski. SPIE, 2018. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{2dcd2631375b47099876e8bbc9f6c835,
title = "Deformable registration of histological cancer margins to gross hyperspectral images using demons",
abstract = "Hyperspectral imaging (HSI), a non-contact optical imaging technique, has been recently used along with machine learning technique to provide diagnostic information about ex-vivo surgical specimens for optical biopsy. The computer-Aided diagnostic approach requires accurate ground truths for both training and validation. This study details a processing pipeline for registering the cancer-normal margin from a digitized histological image to the gross-level HSI of a tissue specimen. Our work incorporates an initial affine and control-point registration followed by a deformable Demons-based registration of the moving mask obtained from the histological image to the fixed mask made from the HS image. To assess registration quality, Dice similarity coefficient (DSC) measures the image overlap, visual inspection is used to evaluate the margin, and average target registration error (TRE) of needle-bored holes measures the registration error between the histologic and HSI images. Excised tissue samples from seventeen patients, 11 head and neck squamous cell carcinoma (HNSCCa) and 6 thyroid carcinoma, were registered according to the proposed method. Three registered specimens are illustrated in this paper, which demonstrate the efficacy of the registration workflow. Further work is required to apply the technique to more patient data and investigate the ability of this procedure to produce suitable gold standards for machine learning validation.",
keywords = "Demons registration, Digital histology, Head and neck cancer surgery, Hyperspectral imaging",
author = "Martin Halicek and Little, {James V.} and Xu Wang and Chen, {Zhuo Georgia} and Mihir Patel and Griffith, {Christopher C.} and El-Deiry, {Mark W.} and Saba, {Nabil F.} and Chen, {Amy Y.} and Baowei Fei",
year = "2018",
month = "1",
day = "1",
doi = "10.1117/12.2293165",
language = "English (US)",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",
editor = "Gurcan, {Metin N.} and Tomaszewski, {John E.}",
booktitle = "Medical Imaging 2018",

}

TY - GEN

T1 - Deformable registration of histological cancer margins to gross hyperspectral images using demons

AU - Halicek, Martin

AU - Little, James V.

AU - Wang, Xu

AU - Chen, Zhuo Georgia

AU - Patel, Mihir

AU - Griffith, Christopher C.

AU - El-Deiry, Mark W.

AU - Saba, Nabil F.

AU - Chen, Amy Y.

AU - Fei, Baowei

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Hyperspectral imaging (HSI), a non-contact optical imaging technique, has been recently used along with machine learning technique to provide diagnostic information about ex-vivo surgical specimens for optical biopsy. The computer-Aided diagnostic approach requires accurate ground truths for both training and validation. This study details a processing pipeline for registering the cancer-normal margin from a digitized histological image to the gross-level HSI of a tissue specimen. Our work incorporates an initial affine and control-point registration followed by a deformable Demons-based registration of the moving mask obtained from the histological image to the fixed mask made from the HS image. To assess registration quality, Dice similarity coefficient (DSC) measures the image overlap, visual inspection is used to evaluate the margin, and average target registration error (TRE) of needle-bored holes measures the registration error between the histologic and HSI images. Excised tissue samples from seventeen patients, 11 head and neck squamous cell carcinoma (HNSCCa) and 6 thyroid carcinoma, were registered according to the proposed method. Three registered specimens are illustrated in this paper, which demonstrate the efficacy of the registration workflow. Further work is required to apply the technique to more patient data and investigate the ability of this procedure to produce suitable gold standards for machine learning validation.

AB - Hyperspectral imaging (HSI), a non-contact optical imaging technique, has been recently used along with machine learning technique to provide diagnostic information about ex-vivo surgical specimens for optical biopsy. The computer-Aided diagnostic approach requires accurate ground truths for both training and validation. This study details a processing pipeline for registering the cancer-normal margin from a digitized histological image to the gross-level HSI of a tissue specimen. Our work incorporates an initial affine and control-point registration followed by a deformable Demons-based registration of the moving mask obtained from the histological image to the fixed mask made from the HS image. To assess registration quality, Dice similarity coefficient (DSC) measures the image overlap, visual inspection is used to evaluate the margin, and average target registration error (TRE) of needle-bored holes measures the registration error between the histologic and HSI images. Excised tissue samples from seventeen patients, 11 head and neck squamous cell carcinoma (HNSCCa) and 6 thyroid carcinoma, were registered according to the proposed method. Three registered specimens are illustrated in this paper, which demonstrate the efficacy of the registration workflow. Further work is required to apply the technique to more patient data and investigate the ability of this procedure to produce suitable gold standards for machine learning validation.

KW - Demons registration

KW - Digital histology

KW - Head and neck cancer surgery

KW - Hyperspectral imaging

UR - http://www.scopus.com/inward/record.url?scp=85049165453&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049165453&partnerID=8YFLogxK

U2 - 10.1117/12.2293165

DO - 10.1117/12.2293165

M3 - Conference contribution

C2 - 30220773

AN - SCOPUS:85049165453

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2018

A2 - Gurcan, Metin N.

A2 - Tomaszewski, John E.

PB - SPIE

ER -