DeepOPG: Improving Orthopantomogram Finding Summarization with Weak Supervision

TL;DR

DeepOPG introduces a weakly supervised method for summarizing dental findings from orthopantomograms, combining functional segmentation, tooth localization, and a novel coherence module to improve detection accuracy.

Contribution

The paper presents a novel deep learning framework that leverages weak supervision and a dental coherence module for effective summarization of dental radiographs.

Findings

Achieved 88.2% AUC in finding detection.

Weak supervision improved detection AP by 0.4%.

Dental coherence module increased AP by 5.9%.

Abstract

Clinical finding summaries from an orthopantomogram, or a dental panoramic radiograph, have significant potential to improve patient communication and speed up clinical judgments. While orthopantomogram is a first-line tool for dental examinations, no existing work has explored the summarization of findings from it. A finding summary has to find teeth in the imaging study and label the teeth with several types of past treatments. To tackle the problem, we developDeepOPG that breaks the summarization process into functional segmentation and tooth localization, the latter of which is further refined by a novel dental coherence module. We also leverage weak supervision labels to improve detection results in a reinforcement learning scenario. Experiments show high efficacy of DeepOPG on finding summarization, achieving an overall AUC of 88.2% in detecting six types of findings. The proposed dental coherence and weak supervision both are shown to improve DeepOPG by adding 5.9% and 0.4% to AP@IoU=0.5.