Mobile-based oral cancer classification for point-of-care screening

Abstract. Significance: Oral cancer is among the most common cancers globally, especially in low- and middle-income countries. Early detection is the most effective way to reduce the mortality rate. Deep learning-based cancer image classification models usually need to be hosted on a computing server. However, internet connection is unreliable for screening in low-resource settings. Aim: To develop a mobile-based dual-mode image classification method and customized Android application for point-of-care oral cancer detection. Approach: The dataset used in our study was captured among 5025 patients with our customized dual-modality mobile oral screening devices. We trained an efficient network MobileNet with focal loss and converted the model into TensorFlow Lite format. The finalized lite format model is ∼16.3  MB and ideal for smartphone platform operation. We have developed an Android smartphone application in an easy-to-use format that implements the mobile-based dual-modality image classification approach to distinguish oral potentially malignant and malignant images from normal/benign images. Results: We investigated the accuracy and running speed on a cost-effective smartphone computing platform. It takes ∼300  ms to process one image pair with the Moto G5 Android smartphone. We tested the proposed method on a standalone dataset and achieved 81% accuracy for distinguishing normal/benign lesions from clinically suspicious lesions, using a gold standard of clinical impression based on the review of images by oral specialists. Conclusions: Our study demonstrates the effectiveness of a mobile-based approach for oral cancer screening in low-resource settings.

[1]  S Silverman,et al.  Early diagnosis of oral cancer , 1988, Cancer.

[2]  Sameer Antani,et al.  An Observational Study of Deep Learning and Automated Evaluation of Cervical Images for Cancer Screening. , 2019, Journal of the National Cancer Institute.

[3]  Guolan Lu,et al.  Deep convolutional neural networks for classifying head and neck cancer using hyperspectral imaging , 2017, Journal of biomedical optics.

[4]  Jacques Ferlay,et al.  The global incidence of lip, oral cavity, and pharyngeal cancers by subsite in 2012 , 2017, CA: a cancer journal for clinicians.

[5]  A. Jemal,et al.  Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries , 2018, CA: a cancer journal for clinicians.

[6]  A. Jemal,et al.  Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries , 2021, CA: a cancer journal for clinicians.

[7]  Bofan Song,et al.  Small form factor, flexible, dual-modality handheld probe for smartphone-based, point-of-care oral and oropharyngeal cancer screening , 2019, Journal of biomedical optics.

[8]  M. Plummer,et al.  International agency for research on cancer. , 2020, Archives of pathology.

[9]  Bo Chen,et al.  MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications , 2017, ArXiv.

[10]  Ken Russell Coelho.  Challenges of the Oral Cancer Burden in India , 2012, Journal of cancer epidemiology.

[11]  Bofan Song,et al.  Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning. , 2018, Biomedical optics express.

[12]  Prevalence of Oral Cancer in India , 2015 .

[13]  Sumsum P. Sunny,et al.  Role of community health worker in a mobile health program for early detection of oral cancer. , 2019, Indian journal of cancer.

[14]  Nils-Claudius Gellrich,et al.  Clinical evaluation of an autofluorescence diagnostic device for oral cancer detection: a prospective randomized diagnostic study , 2012, European journal of cancer prevention : the official journal of the European Cancer Prevention Organisation.

[15]  J. Califano,et al.  Global Oral Health Inequalities in Incidence and Outcomes for Oral Cancer , 2011, Advances in dental research.

[16]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[17]  G. Keerthi,et al.  Point-of-care, smartphone-based, dual-modality, dual-view, oral cancer screening device with neural network classification for low-resource communities , 2018, PloS one.

[18]  Geoffrey E. Hinton.  Deep Learning-A Technology With the Potential to Transform Health Care. , 2018, JAMA.

[19]  Camile S Farah,et al.  Efficacy of tissue autofluorescence imaging (velscope) in the visualization of oral mucosal lesions , 2012, Head & neck.

[20]  N. Johnson,et al.  Oral cancer: Indian pandemic , 2017, BDJ.

[21]  Joachim Denzler,et al.  Automatic Classification of Cancerous Tissue in Laserendomicroscopy Images of the Oral Cavity using Deep Learning , 2017, Scientific Reports.

[22]  Rebecca Richards-Kortum,et al.  Understanding the Biological Basis of Autofluorescence Imaging for Oral Cancer Detection: High-Resolution Fluorescence Microscopy in Viable Tissue , 2008, Clinical Cancer Research.

[23]  Stéphane Mallat,et al.  Rigid-Motion Scattering for Texture Classification , 2014, ArXiv.

[24]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[25]  Pankaj Shukla,et al.  A cross-sectional study evaluating chemiluminescence and autofluorescence in the detection of clinically innocuous precancerous and cancerous oral lesions. , 2010, Journal of the American Dental Association.

[26]  Poul Erik Petersen,et al.  Oral Cancer Prevention and Control – the Approach of the World Health Organization , 2022 .

[27]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[28]  Fei Wang,et al.  Deep learning for healthcare: review, opportunities and challenges , 2018, Briefings Bioinform..

[29]  Yi Yin,et al.  Automatic classification of cervical cancer from cytological images by using convolutional neural network , 2018, Bioscience reports.