Hyperspectral imaging of head and neck squamous cell carcinoma for cancer margin detection in surgical specimens from 102 patients using deep learning

Martin Halicek, James D. Dormer, James V. Little, Amy Y. Chen, Larry Myers, Baran D. Sumer, Baowei Fei

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Surgical resection of head and neck (H and N) squamous cell carcinoma (SCC) may yield inadequate surgical cancer margins in 10 to 20% of cases. This study investigates the performance of label-free, reflectance-based hyperspectral imaging (HSI) and autofluorescence imaging for SCC detection at the cancer margin in excised tissue specimens from 102 patients and uses fluorescent dyes for comparison. Fresh surgical specimens (n = 293) were collected during H and N SCC resections (n = 102). The tissue specimens were imaged with reflectance-based HSI and autofluorescence imaging and afterwards with two fluorescent dyes for comparison. A histopathological ground truth was made. Deep learning tools were developed to detect SCC with new patient samples (inter-patient) and machine learning for intra-patient tissue samples. Area under the curve (AUC) of the receiver-operator characteristic was used as the main evaluation metric. Additionally, the performance was estimated in mm increments circumferentially from the tumor-normal margin. In intra-patient experiments, HSI classified conventional SCC with an AUC of 0.82 up to 3 mm from the cancer margin, which was more accurate than proflavin dye and autofluorescence (both p < 0.05). Intra-patient autofluorescence imaging detected human papilloma virus positive (HPV+) SCC with an AUC of 0.99 at 3 mm and greater accuracy than proflavin dye (p < 0.05). The inter-patient results showed that reflectance-based HSI and autofluorescence imaging outperformed proflavin dye and standard red, green, and blue (RGB) images (p < 0.05). In new patients, HSI detected conventional SCC in the larynx, oropharynx, and nasal cavity with 0.85–0.95 AUC score, and autofluorescence imaging detected HPV+ SCC in tonsillar tissue with 0.91 AUC score. This study demonstrates that label-free, reflectance-based HSI and autofluorescence imaging methods can accurately detect the cancer margin in ex-vivo specimens within minutes. This non-ionizing optical imaging modality could aid surgeons and reduce inadequate surgical margins during SCC resections.

Original languageEnglish (US)
Article number1367
JournalCancers
Volume11
Issue number9
DOIs
StatePublished - Sep 2019

Fingerprint

Squamous Cell Neoplasms
Head and Neck Neoplasms
Optical Imaging
Squamous Cell Carcinoma
Learning
Proflavine
Area Under Curve
Coloring Agents
Neoplasms
Fluorescent Dyes
Papillomaviridae
Oropharynx
Carcinoma, squamous cell of head and neck
Nasal Cavity
Larynx

Keywords

  • Convolutional neural network
  • Deep learning
  • Head and neck cancer
  • Hyperspectral imaging
  • Squamous cell carcinoma

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this

Hyperspectral imaging of head and neck squamous cell carcinoma for cancer margin detection in surgical specimens from 102 patients using deep learning. / Halicek, Martin; Dormer, James D.; Little, James V.; Chen, Amy Y.; Myers, Larry; Sumer, Baran D.; Fei, Baowei.

In: Cancers, Vol. 11, No. 9, 1367, 09.2019.

Research output: Contribution to journalArticle

@article{69b71e8c87714ef2a446011fee79bcc0,
title = "Hyperspectral imaging of head and neck squamous cell carcinoma for cancer margin detection in surgical specimens from 102 patients using deep learning",
abstract = "Surgical resection of head and neck (H and N) squamous cell carcinoma (SCC) may yield inadequate surgical cancer margins in 10 to 20{\%} of cases. This study investigates the performance of label-free, reflectance-based hyperspectral imaging (HSI) and autofluorescence imaging for SCC detection at the cancer margin in excised tissue specimens from 102 patients and uses fluorescent dyes for comparison. Fresh surgical specimens (n = 293) were collected during H and N SCC resections (n = 102). The tissue specimens were imaged with reflectance-based HSI and autofluorescence imaging and afterwards with two fluorescent dyes for comparison. A histopathological ground truth was made. Deep learning tools were developed to detect SCC with new patient samples (inter-patient) and machine learning for intra-patient tissue samples. Area under the curve (AUC) of the receiver-operator characteristic was used as the main evaluation metric. Additionally, the performance was estimated in mm increments circumferentially from the tumor-normal margin. In intra-patient experiments, HSI classified conventional SCC with an AUC of 0.82 up to 3 mm from the cancer margin, which was more accurate than proflavin dye and autofluorescence (both p < 0.05). Intra-patient autofluorescence imaging detected human papilloma virus positive (HPV+) SCC with an AUC of 0.99 at 3 mm and greater accuracy than proflavin dye (p < 0.05). The inter-patient results showed that reflectance-based HSI and autofluorescence imaging outperformed proflavin dye and standard red, green, and blue (RGB) images (p < 0.05). In new patients, HSI detected conventional SCC in the larynx, oropharynx, and nasal cavity with 0.85–0.95 AUC score, and autofluorescence imaging detected HPV+ SCC in tonsillar tissue with 0.91 AUC score. This study demonstrates that label-free, reflectance-based HSI and autofluorescence imaging methods can accurately detect the cancer margin in ex-vivo specimens within minutes. This non-ionizing optical imaging modality could aid surgeons and reduce inadequate surgical margins during SCC resections.",
keywords = "Convolutional neural network, Deep learning, Head and neck cancer, Hyperspectral imaging, Squamous cell carcinoma",
author = "Martin Halicek and Dormer, {James D.} and Little, {James V.} and Chen, {Amy Y.} and Larry Myers and Sumer, {Baran D.} and Baowei Fei",
year = "2019",
month = "9",
doi = "10.3390/cancers11091367",
language = "English (US)",
volume = "11",
journal = "Cancers",
issn = "2072-6694",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "9",

}

TY - JOUR

T1 - Hyperspectral imaging of head and neck squamous cell carcinoma for cancer margin detection in surgical specimens from 102 patients using deep learning

AU - Halicek, Martin

AU - Dormer, James D.

AU - Little, James V.

AU - Chen, Amy Y.

AU - Myers, Larry

AU - Sumer, Baran D.

AU - Fei, Baowei

PY - 2019/9

Y1 - 2019/9

N2 - Surgical resection of head and neck (H and N) squamous cell carcinoma (SCC) may yield inadequate surgical cancer margins in 10 to 20% of cases. This study investigates the performance of label-free, reflectance-based hyperspectral imaging (HSI) and autofluorescence imaging for SCC detection at the cancer margin in excised tissue specimens from 102 patients and uses fluorescent dyes for comparison. Fresh surgical specimens (n = 293) were collected during H and N SCC resections (n = 102). The tissue specimens were imaged with reflectance-based HSI and autofluorescence imaging and afterwards with two fluorescent dyes for comparison. A histopathological ground truth was made. Deep learning tools were developed to detect SCC with new patient samples (inter-patient) and machine learning for intra-patient tissue samples. Area under the curve (AUC) of the receiver-operator characteristic was used as the main evaluation metric. Additionally, the performance was estimated in mm increments circumferentially from the tumor-normal margin. In intra-patient experiments, HSI classified conventional SCC with an AUC of 0.82 up to 3 mm from the cancer margin, which was more accurate than proflavin dye and autofluorescence (both p < 0.05). Intra-patient autofluorescence imaging detected human papilloma virus positive (HPV+) SCC with an AUC of 0.99 at 3 mm and greater accuracy than proflavin dye (p < 0.05). The inter-patient results showed that reflectance-based HSI and autofluorescence imaging outperformed proflavin dye and standard red, green, and blue (RGB) images (p < 0.05). In new patients, HSI detected conventional SCC in the larynx, oropharynx, and nasal cavity with 0.85–0.95 AUC score, and autofluorescence imaging detected HPV+ SCC in tonsillar tissue with 0.91 AUC score. This study demonstrates that label-free, reflectance-based HSI and autofluorescence imaging methods can accurately detect the cancer margin in ex-vivo specimens within minutes. This non-ionizing optical imaging modality could aid surgeons and reduce inadequate surgical margins during SCC resections.

AB - Surgical resection of head and neck (H and N) squamous cell carcinoma (SCC) may yield inadequate surgical cancer margins in 10 to 20% of cases. This study investigates the performance of label-free, reflectance-based hyperspectral imaging (HSI) and autofluorescence imaging for SCC detection at the cancer margin in excised tissue specimens from 102 patients and uses fluorescent dyes for comparison. Fresh surgical specimens (n = 293) were collected during H and N SCC resections (n = 102). The tissue specimens were imaged with reflectance-based HSI and autofluorescence imaging and afterwards with two fluorescent dyes for comparison. A histopathological ground truth was made. Deep learning tools were developed to detect SCC with new patient samples (inter-patient) and machine learning for intra-patient tissue samples. Area under the curve (AUC) of the receiver-operator characteristic was used as the main evaluation metric. Additionally, the performance was estimated in mm increments circumferentially from the tumor-normal margin. In intra-patient experiments, HSI classified conventional SCC with an AUC of 0.82 up to 3 mm from the cancer margin, which was more accurate than proflavin dye and autofluorescence (both p < 0.05). Intra-patient autofluorescence imaging detected human papilloma virus positive (HPV+) SCC with an AUC of 0.99 at 3 mm and greater accuracy than proflavin dye (p < 0.05). The inter-patient results showed that reflectance-based HSI and autofluorescence imaging outperformed proflavin dye and standard red, green, and blue (RGB) images (p < 0.05). In new patients, HSI detected conventional SCC in the larynx, oropharynx, and nasal cavity with 0.85–0.95 AUC score, and autofluorescence imaging detected HPV+ SCC in tonsillar tissue with 0.91 AUC score. This study demonstrates that label-free, reflectance-based HSI and autofluorescence imaging methods can accurately detect the cancer margin in ex-vivo specimens within minutes. This non-ionizing optical imaging modality could aid surgeons and reduce inadequate surgical margins during SCC resections.

KW - Convolutional neural network

KW - Deep learning

KW - Head and neck cancer

KW - Hyperspectral imaging

KW - Squamous cell carcinoma

UR - http://www.scopus.com/inward/record.url?scp=85073361745&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073361745&partnerID=8YFLogxK

U2 - 10.3390/cancers11091367

DO - 10.3390/cancers11091367

M3 - Article

C2 - 31540063

AN - SCOPUS:85073361745

VL - 11

JO - Cancers

JF - Cancers

SN - 2072-6694

IS - 9

M1 - 1367

ER -