Dental workers, particularly less experienced examiners, are unable to accurately distinguish dysplastic or cancerous lesions of the oral mucosa needing treatment from benign oral lesions. Furthermore, variable stages of dysplasia to carcinoma occur within a single oral lesion, and dental workers often do not biopsy the site which contains the worst pathologic diagnosis to inform treatment as is ideal. The goal of this project is to develop an automated, optical imaging-based platform to promote early detection of cancer in community dental clinics. Its central hypothesis is that optical properties of dysplastic or cancerous oral epithelium can be measured and exploited to augment the ability of dental workers to diagnose these lesions. In particular, two imaging modalities used: 1) widefield autofluorescence imaging (AFI), which has high sensitivity but low specificity for dysplasia and cancer, and 2) high-resolution microendoscopy (HRME), which has a higher specificity but can only interrogate a <1 mm diameter of tissue at once. To achieve the objective, two specific aims will be completed. First, an automated, multimodal imaging early- detection platform that identifies high-risk sites within oral mucosal lesions using widefield images, then distinguishes dysplastic or cancerous lesions from benign lesions by combining HRME and widefield AFI features will be developed. To achieve this aim, an automated image analysis algorithm capable of identifying high-risk sites within oral lesions will be developed; this algorithm will then be combined with existing HRME software to create a single, integrated user interface for the platform. Second, the performance and clinical utility of the early-detection platform will be evaluated by performing a clinical study in 75 patients with oral mucosal lesions with low-prevalence of oral neoplasia (and high-prevalence of benign lesions) similar to community dental clinics. The ability of the early-detection platform to identify high-risk sites within a lesion and distinguish dysplastic or cancerous sites from benign sites will be compared to an expert clinician and dental workers. The clinical utility of the platform will be evaluated based on its ability to provide diagnostic information that improves the ability of dental workers to distinguish dysplastic or cancerous sites from benign sites. This research is significant because it will provide dental workers with an easy to use platform that promotes the early diagnosis of oral neoplasia, a critical step to reducing the global oral cancer burden, and may also reduce unnecessary diagnostic workup of low-risk lesions.

Public Health Relevance

A major barrier to reducing the global oral cancer burden is the high frequency of late diagnosis, resulting in higher mortality, morbidity, and healthcare costs. The proposed research will result in a portable, handheld, automated imaging system useful to community dental workers for the identification of oral precancer and early- stage cancer, which could significantly decrease the global oral cancer burden through earlier diagnosis.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Individual Predoctoral NRSA for M.D./Ph.D. Fellowships (ADAMHA) (F30)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Korczak, Jeannette F
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Rice University
Biomedical Engineering
Biomed Engr/Col Engr/Engr Sta
United States
Zip Code
Yang, Eric C; Schwarz, Richard A; Lang, Alexander K et al. (2018) In Vivo Multimodal Optical Imaging: Improved Detection of Oral Dysplasia in Low-Risk Oral Mucosal Lesions. Cancer Prev Res (Phila) 11:465-476
Yang, Eric C; Tan, Melody T; Schwarz, Richard A et al. (2018) Noninvasive diagnostic adjuncts for the evaluation of potentially premalignant oral epithelial lesions: current limitations and future directions. Oral Surg Oral Med Oral Pathol Oral Radiol 125:670-681