An Attention-Guided Deep Regression Model for Landmark Detection in Cephalograms

TL;DR

This paper introduces an attention-guided deep learning framework using a 2-stage U-Net for accurate, robust landmark detection in cephalometric X-ray images, enhancing orthodontic diagnosis with minimal manual tuning.

Contribution

It presents a novel 2-stage U-Net architecture with integrated attention mechanisms and an exploration strategy for improved landmark detection in cephalograms.

Findings

Achieves state-of-the-art accuracy on public cephalometric datasets.

Reduces manual tuning and computational complexity.

Demonstrates robustness with expanded search scope.

Abstract

Cephalometric tracing method is usually used in orthodontic diagnosis and treatment planning. In this paper, we propose a deep learning based framework to automatically detect anatomical landmarks in cephalometric X-ray images. We train the deep encoder-decoder for landmark detection, and combine global landmark configuration with local high-resolution feature responses. The proposed frame-work is based on 2-stage u-net, regressing the multi-channel heatmaps for land-mark detection. In this framework, we embed attention mechanism with global stage heatmaps, guiding the local stage inferring, to regress the local heatmap patches in a high resolution. Besides, the Expansive Exploration strategy improves robustness while inferring, expanding the searching scope without increasing model complexity. We have evaluated our framework in the most widely-used public dataset of landmark detection in cephalometric X-ray images. With less computation and manually tuning, our framework achieves state-of-the-art results.