A Visualization System for Interactive Diagnosis and Intelligent Statistical Analysis of Oral Diseases based on Panoramic Radiography: Method Study (Preprint)

Abstract

Panoramic radiography is an essential auxiliary diagnostic tool for oral diseases. It is a difficult and time-consuming task to generate electronic medical records and explore oral diseases based on collective analysis after conducting extensive panoramic radiography interpretation.

The objective of our work is to develop a visualization system for the interactive diagnosis and intelligent statistical analysis of oral diseases based on panoramic radiographs.

Firstly, we employ a deep learning model, Mask Region-based Convolutional Neural Network (Mask-RCNN), to conduct tooth segmentation in the panoramic radiographs. Then, we provide a rich set of user interfaces for doctors to revise the tooth segmentations and identify the lesion location for the analysis of oral diseases. After that, we provide a rule-based Natural Language Processing (NLP) method to generate electronic medical records. We further develop statistical correlation analysis to visually evaluate and interactively explore the statistical data from oral health surveys.

We conduct tooth segmentation and manual diagnosis, obtain their electronic medical records and do collective analysis based on a group of real-world panoramic radiographs. The results are reported from a comprehensive case study showing that our system is capable of improving the efficiency in disease detection and data mining, which also can fuel research studies in the field of public oral health.

This paper provides an efficient, automatic, visualizing, and user-friendly system, which helps dentists easily conduct personal oral diagnoses and access collective oral health status. Moreover, extensive population-level oral health surveys and visual analyses provide robust guidance and support for oral healthcare strategies.