ABSTRACT
This paper aims to simplify the application of optical coherence tomography (OCT) for the examination of subsurface morphology in the oral cavity and reduce barriers towards the adoption of OCT as a biopsy guidance device. The aim of this work was to develop automated software tools for the simplified analysis of the large volume of data collected during OCT. Imaging and corresponding histopathology were acquired in-clinic using a wide-field endoscopic OCT system. An annotated dataset (n = 294 images) from 60 patients (34 male and 26 female) was assembled to train four unique neural networks. A deep learning pipeline was built using convolutional and modified u-net models to detect the imaging field of view (network 1), detect artifacts (network 2), identify the tissue surface (network 3), and identify the presence and location of the epithelial-stromal boundary (network 4). The area under the curve of the image and artifact detection networks was 1.00 and 0.94, respectively. The Dice similarity score for the surface and epithelial-stromal boundary segmentation networks was 0.98 and 0.83, respectively. Deep learning (DL) techniques can identify the location and variations in the epithelial surface and epithelial-stromal boundary in OCT images of the oral mucosa. Segmentation results can be synthesized into accessible en face maps to allow easier visualization of changes.
ABSTRACT
BACKGROUND/PURPOSE: Individuals with low income bear a number of health challenges to healthcare services. Vancouver's Downtown Eastside (DTES) is known to be a low-income community in a metropolitan city. Because it is difficult to reach, the oral health (OH) status of these residents is unknown. The objectives of this study are (1) to design a tool and strategy to collect OH information in a low-income community, (2) to characterize the OH status and related factors among low-income adults, and (3) to identify the explanatory factors for their OH status. MATERIALS AND METHODS: Mobile screening clinics were established in the gathering centers of the DTES, and those of 19 years of age or older were recruited. Data were collected through survey interviews and clinical examinations. Potential explanatory factors were investigated by regression analysis. RESULTS: The 356 screened participants were mostly males, middle-aged, less educated, and living with low income (≤CAD$20,000/y). About 80% had dental coverage, mostly from public programs (94%). Many (86%) perceived a dental need. Among dentate participants (n = 306), on average, 3.8 decayed, 8.6 missing, 4.9 filled teeth, and a care index of 41.5% were observed. Social factors (barriers to care and length of DTES residence), dental hygiene (brushing/flossing), and personal (hepatitis C virus infection/methadone usage) factors contributed to their care index level. CONCLUSION: This is the first time that comprehensive information regarding OH status has been collected from a low-income, inner-city community in Canada. Further investigations in the challenges and needs in accessing dental care may develop solutions for better OH in similar communities.
