Deep Learning for Dental Hyperspectral Image Analysis

Abstract

The aim of this work is automatic and efficient detection of medically-relevant features from oral and dental hyperspectral images by applying up-to-date deep learning convolutional neural network techniques. This will help dentists to identify and classify unhealthy areas automatically and to prevent the progression of diseases. Hyperspectral imaging approach allows one to do so without exposing the patient to ionizing X-ray radiation. Spectral imaging provides information in the visible and near-infrared wavelength ranges. The dataset used in this paper contains 116 hyperspectral images from 18 patients taken from different viewing angles. Image annotation (ground truth) includes 38 classes in six different sub-groups assessed by dental experts. Mask region-based convolutional neural network (Mask R-CNN) is used as a deep learning model, for instance segmentation of areas. Preliminary results show high potential and accuracy for classification and segmentation of different classes.

The Optimization of the Light-Source Spectrum Utilizing Neural Networks for Detecting Oral Lesions

“…This problem is exacerbated in low- and middle-income countries where expensive radiographic instruments are not widely available. Automatic diagnostic methods for periodontal disease via machine learning have been implemented in specific research using X-ray and hyperspectral images as the inputs [ 27 , 28 , 29 , 30 ]. However, both of these methods are not practical compared to diagnosis by a medical practitioner using X-rays because of the high cost of imaging and the protracted processes involved.…”

Oral and Dental Spectral Image Database—ODSI-DB

“…In the simplest cases, spectral images can be used to design systems that enhance the visibility of a targeted feature, for example, by applying different weights on the bands while computing a color image representation [7], or by optimizing contrast to produce optimal optical filters [8,9]. The selected spectra could also be used in training targeted diagnostic imaging systems, like spectral filter array cameras [10], or in automatic diagnostic image segmentation based on deep learning [11].…”

“…The annotations were made by dental experts, who used an annotation tool built for the project. We have successfully used the database to develop a prototype optical imaging system based on partially negative filters derived from principal component analysis of healthy and diseased tissue [9], and in automatic image segmentation and classification based on a convolutional neural network [11].…”