Interpretation of Optical Coherence Tomography Images for Breast Tissue Assessment

Kiran S. Yemul, Adam M. Zysk, Andrea Richardson, Krishnarao V. Tangella, Lisa Jacobs

Research output: Contribution to journalArticle

Abstract

Purpose. Initial studies have shown that optical coherence tomography (OCT) is an effective margin-evaluation tool for breast-conserving surgery, but methods for the interpretation of breast OCT images have not been directly studied. In this work, breast pathologies were assessed with a handheld OCT probe. OCT images and corresponding histology were used to develop guidelines for the identification of breast tissue features in OCT images. Methods. Mastectomy and breast-conserving surgery specimens from 26 women were imaged with a handheld OCT probe. During standard pathology specimen dissection, representative 1-cm × 1-cm tissue regions were grossly identified, assessed with OCT, inked for orientation and image-matching purposes, and processed. Histology slides corresponding to the OCT image region were digitally photographed. OCT and histology images from the same region were paired by selecting the best structural matches. Results. In total, 2880 OCT images were acquired from 26 breast specimens (from 26 patients) and 48 matching OCT-histology image pairs were identified. These matched image pairs illustrate tissue types including adipose tissue, dense fibrosis, fibroadipose tissue, blood vessels, regular and hyperplastic ducts and lobules, cysts, cyst, fibroadenoma, invasive ductal carcinoma, invasive lobular carcinoma, ductal carcinoma in situ, calcifications, and biopsy cavities. Differentiation between pathologies was achieved by considering feature boundaries, interior appearance, posterior shadowing or enhancement, and overall morphologic patterns. Conclusions. This is the first work to systematically catalog the critical features of breast OCT images. The results indicate that OCT can be used to identify and distinguish between benign and malignant features in human breast tissue.

Original languageEnglish (US)
Pages (from-to)50-56
Number of pages7
JournalSurgical Innovation
Volume26
Issue number1
DOIs
StatePublished - Feb 1 2019

Fingerprint

Optical Coherence Tomography
Breast
Histology
Segmental Mastectomy
Pathology
Cysts
Lobular Carcinoma
Fibroadenoma
Ductal Carcinoma
Carcinoma, Intraductal, Noninfiltrating
Mastectomy
Blood Vessels
Adipose Tissue
Dissection
Fibrosis

Keywords

  • breast surgery
  • image-guided surgery
  • surgical oncology

ASJC Scopus subject areas

  • Surgery

Cite this

Interpretation of Optical Coherence Tomography Images for Breast Tissue Assessment. / Yemul, Kiran S.; Zysk, Adam M.; Richardson, Andrea; Tangella, Krishnarao V.; Jacobs, Lisa.

In: Surgical Innovation, Vol. 26, No. 1, 01.02.2019, p. 50-56.

Research output: Contribution to journalArticle

Yemul, Kiran S. ; Zysk, Adam M. ; Richardson, Andrea ; Tangella, Krishnarao V. ; Jacobs, Lisa. / Interpretation of Optical Coherence Tomography Images for Breast Tissue Assessment. In: Surgical Innovation. 2019 ; Vol. 26, No. 1. pp. 50-56.
@article{816049b2071c409ab109a0ac5950edc8,
title = "Interpretation of Optical Coherence Tomography Images for Breast Tissue Assessment",
abstract = "Purpose. Initial studies have shown that optical coherence tomography (OCT) is an effective margin-evaluation tool for breast-conserving surgery, but methods for the interpretation of breast OCT images have not been directly studied. In this work, breast pathologies were assessed with a handheld OCT probe. OCT images and corresponding histology were used to develop guidelines for the identification of breast tissue features in OCT images. Methods. Mastectomy and breast-conserving surgery specimens from 26 women were imaged with a handheld OCT probe. During standard pathology specimen dissection, representative 1-cm × 1-cm tissue regions were grossly identified, assessed with OCT, inked for orientation and image-matching purposes, and processed. Histology slides corresponding to the OCT image region were digitally photographed. OCT and histology images from the same region were paired by selecting the best structural matches. Results. In total, 2880 OCT images were acquired from 26 breast specimens (from 26 patients) and 48 matching OCT-histology image pairs were identified. These matched image pairs illustrate tissue types including adipose tissue, dense fibrosis, fibroadipose tissue, blood vessels, regular and hyperplastic ducts and lobules, cysts, cyst, fibroadenoma, invasive ductal carcinoma, invasive lobular carcinoma, ductal carcinoma in situ, calcifications, and biopsy cavities. Differentiation between pathologies was achieved by considering feature boundaries, interior appearance, posterior shadowing or enhancement, and overall morphologic patterns. Conclusions. This is the first work to systematically catalog the critical features of breast OCT images. The results indicate that OCT can be used to identify and distinguish between benign and malignant features in human breast tissue.",
keywords = "breast surgery, image-guided surgery, surgical oncology",
author = "Yemul, {Kiran S.} and Zysk, {Adam M.} and Andrea Richardson and Tangella, {Krishnarao V.} and Lisa Jacobs",
year = "2019",
month = "2",
day = "1",
doi = "10.1177/1553350618803245",
language = "English (US)",
volume = "26",
pages = "50--56",
journal = "Surgical Innovation",
issn = "1553-3506",
publisher = "SAGE Publications Inc.",
number = "1",

}

TY - JOUR

T1 - Interpretation of Optical Coherence Tomography Images for Breast Tissue Assessment

AU - Yemul, Kiran S.

AU - Zysk, Adam M.

AU - Richardson, Andrea

AU - Tangella, Krishnarao V.

AU - Jacobs, Lisa

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Purpose. Initial studies have shown that optical coherence tomography (OCT) is an effective margin-evaluation tool for breast-conserving surgery, but methods for the interpretation of breast OCT images have not been directly studied. In this work, breast pathologies were assessed with a handheld OCT probe. OCT images and corresponding histology were used to develop guidelines for the identification of breast tissue features in OCT images. Methods. Mastectomy and breast-conserving surgery specimens from 26 women were imaged with a handheld OCT probe. During standard pathology specimen dissection, representative 1-cm × 1-cm tissue regions were grossly identified, assessed with OCT, inked for orientation and image-matching purposes, and processed. Histology slides corresponding to the OCT image region were digitally photographed. OCT and histology images from the same region were paired by selecting the best structural matches. Results. In total, 2880 OCT images were acquired from 26 breast specimens (from 26 patients) and 48 matching OCT-histology image pairs were identified. These matched image pairs illustrate tissue types including adipose tissue, dense fibrosis, fibroadipose tissue, blood vessels, regular and hyperplastic ducts and lobules, cysts, cyst, fibroadenoma, invasive ductal carcinoma, invasive lobular carcinoma, ductal carcinoma in situ, calcifications, and biopsy cavities. Differentiation between pathologies was achieved by considering feature boundaries, interior appearance, posterior shadowing or enhancement, and overall morphologic patterns. Conclusions. This is the first work to systematically catalog the critical features of breast OCT images. The results indicate that OCT can be used to identify and distinguish between benign and malignant features in human breast tissue.

AB - Purpose. Initial studies have shown that optical coherence tomography (OCT) is an effective margin-evaluation tool for breast-conserving surgery, but methods for the interpretation of breast OCT images have not been directly studied. In this work, breast pathologies were assessed with a handheld OCT probe. OCT images and corresponding histology were used to develop guidelines for the identification of breast tissue features in OCT images. Methods. Mastectomy and breast-conserving surgery specimens from 26 women were imaged with a handheld OCT probe. During standard pathology specimen dissection, representative 1-cm × 1-cm tissue regions were grossly identified, assessed with OCT, inked for orientation and image-matching purposes, and processed. Histology slides corresponding to the OCT image region were digitally photographed. OCT and histology images from the same region were paired by selecting the best structural matches. Results. In total, 2880 OCT images were acquired from 26 breast specimens (from 26 patients) and 48 matching OCT-histology image pairs were identified. These matched image pairs illustrate tissue types including adipose tissue, dense fibrosis, fibroadipose tissue, blood vessels, regular and hyperplastic ducts and lobules, cysts, cyst, fibroadenoma, invasive ductal carcinoma, invasive lobular carcinoma, ductal carcinoma in situ, calcifications, and biopsy cavities. Differentiation between pathologies was achieved by considering feature boundaries, interior appearance, posterior shadowing or enhancement, and overall morphologic patterns. Conclusions. This is the first work to systematically catalog the critical features of breast OCT images. The results indicate that OCT can be used to identify and distinguish between benign and malignant features in human breast tissue.

KW - breast surgery

KW - image-guided surgery

KW - surgical oncology

UR - http://www.scopus.com/inward/record.url?scp=85059300997&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059300997&partnerID=8YFLogxK

U2 - 10.1177/1553350618803245

DO - 10.1177/1553350618803245

M3 - Article

C2 - 30295149

AN - SCOPUS:85059300997

VL - 26

SP - 50

EP - 56

JO - Surgical Innovation

JF - Surgical Innovation

SN - 1553-3506

IS - 1

ER -