Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning

Bofan Song, Sumsum Sunny, Ross D. Uthoff, Sanjana Patrick, Amritha Suresh, Trupti Kolur, G. Keerthi, Afarin Anbarani, Petra Wilder-Smith, Moni Abraham Kuriakose, Praveen Birur, Jeffrey J Rodriguez, Rongguang Liang

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

With the goal to screen high-risk populations for oral cancer in low-and middle-income countries (LMICs), we have developed a low-cost, portable, easy to use smartphone-based intraoral dual-modality imaging platform. In this paper we present an image classification approach based on autofluorescence and white light images using deep learning methods. The information from the autofluorescence and white light image pair is extracted, calculated, and fused to feed the deep learning neural networks. We have investigated and compared the performance of different convolutional neural networks, transfer learning, and several regularization techniques for oral cancer classification. Our experimental results demonstrate the effectiveness of deep learning methods in classifying dual-modal images for oral cancer detection.

Original languageEnglish (US)
Article number#336298
Pages (from-to)5318-5329
Number of pages12
JournalBiomedical Optics Express
Volume9
Issue number11
DOIs
StatePublished - Nov 1 2018

Fingerprint

Mouth Neoplasms
learning
Learning
cancer
Light
Neoplasms
income
Developed Countries
image classification
classifying
Costs and Cost Analysis
platforms
Population
Smartphone

ASJC Scopus subject areas

  • Biotechnology
  • Atomic and Molecular Physics, and Optics

Cite this

Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning. / Song, Bofan; Sunny, Sumsum; Uthoff, Ross D.; Patrick, Sanjana; Suresh, Amritha; Kolur, Trupti; Keerthi, G.; Anbarani, Afarin; Wilder-Smith, Petra; Kuriakose, Moni Abraham; Birur, Praveen; Rodriguez, Jeffrey J; Liang, Rongguang.

In: Biomedical Optics Express, Vol. 9, No. 11, #336298, 01.11.2018, p. 5318-5329.

Research output: Contribution to journalArticle

Song, B, Sunny, S, Uthoff, RD, Patrick, S, Suresh, A, Kolur, T, Keerthi, G, Anbarani, A, Wilder-Smith, P, Kuriakose, MA, Birur, P, Rodriguez, JJ & Liang, R 2018, 'Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning', Biomedical Optics Express, vol. 9, no. 11, #336298, pp. 5318-5329. https://doi.org/10.1364/BOE.9.005318
Song, Bofan ; Sunny, Sumsum ; Uthoff, Ross D. ; Patrick, Sanjana ; Suresh, Amritha ; Kolur, Trupti ; Keerthi, G. ; Anbarani, Afarin ; Wilder-Smith, Petra ; Kuriakose, Moni Abraham ; Birur, Praveen ; Rodriguez, Jeffrey J ; Liang, Rongguang. / Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning. In: Biomedical Optics Express. 2018 ; Vol. 9, No. 11. pp. 5318-5329.
@article{9ee636257da94c8bba09e929258c9bdf,
title = "Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning",
abstract = "With the goal to screen high-risk populations for oral cancer in low-and middle-income countries (LMICs), we have developed a low-cost, portable, easy to use smartphone-based intraoral dual-modality imaging platform. In this paper we present an image classification approach based on autofluorescence and white light images using deep learning methods. The information from the autofluorescence and white light image pair is extracted, calculated, and fused to feed the deep learning neural networks. We have investigated and compared the performance of different convolutional neural networks, transfer learning, and several regularization techniques for oral cancer classification. Our experimental results demonstrate the effectiveness of deep learning methods in classifying dual-modal images for oral cancer detection.",
author = "Bofan Song and Sumsum Sunny and Uthoff, {Ross D.} and Sanjana Patrick and Amritha Suresh and Trupti Kolur and G. Keerthi and Afarin Anbarani and Petra Wilder-Smith and Kuriakose, {Moni Abraham} and Praveen Birur and Rodriguez, {Jeffrey J} and Rongguang Liang",
year = "2018",
month = "11",
day = "1",
doi = "10.1364/BOE.9.005318",
language = "English (US)",
volume = "9",
pages = "5318--5329",
journal = "Biomedical Optics Express",
issn = "2156-7085",
publisher = "The Optical Society",
number = "11",

}

TY - JOUR

T1 - Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning

AU - Song, Bofan

AU - Sunny, Sumsum

AU - Uthoff, Ross D.

AU - Patrick, Sanjana

AU - Suresh, Amritha

AU - Kolur, Trupti

AU - Keerthi, G.

AU - Anbarani, Afarin

AU - Wilder-Smith, Petra

AU - Kuriakose, Moni Abraham

AU - Birur, Praveen

AU - Rodriguez, Jeffrey J

AU - Liang, Rongguang

PY - 2018/11/1

Y1 - 2018/11/1

N2 - With the goal to screen high-risk populations for oral cancer in low-and middle-income countries (LMICs), we have developed a low-cost, portable, easy to use smartphone-based intraoral dual-modality imaging platform. In this paper we present an image classification approach based on autofluorescence and white light images using deep learning methods. The information from the autofluorescence and white light image pair is extracted, calculated, and fused to feed the deep learning neural networks. We have investigated and compared the performance of different convolutional neural networks, transfer learning, and several regularization techniques for oral cancer classification. Our experimental results demonstrate the effectiveness of deep learning methods in classifying dual-modal images for oral cancer detection.

AB - With the goal to screen high-risk populations for oral cancer in low-and middle-income countries (LMICs), we have developed a low-cost, portable, easy to use smartphone-based intraoral dual-modality imaging platform. In this paper we present an image classification approach based on autofluorescence and white light images using deep learning methods. The information from the autofluorescence and white light image pair is extracted, calculated, and fused to feed the deep learning neural networks. We have investigated and compared the performance of different convolutional neural networks, transfer learning, and several regularization techniques for oral cancer classification. Our experimental results demonstrate the effectiveness of deep learning methods in classifying dual-modal images for oral cancer detection.

UR - http://www.scopus.com/inward/record.url?scp=85056568347&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056568347&partnerID=8YFLogxK

U2 - 10.1364/BOE.9.005318

DO - 10.1364/BOE.9.005318

M3 - Article

VL - 9

SP - 5318

EP - 5329

JO - Biomedical Optics Express

JF - Biomedical Optics Express

SN - 2156-7085

IS - 11

M1 - #336298

ER -