Inferior Alveolar Nerve Segmentation in CBCT images using Connectivity-Based Selective Re-training

TL;DR

This paper introduces a semi-supervised, connectivity-based re-training method for improving inferior alveolar nerve segmentation in CBCT images, achieving state-of-the-art results and winning a challenge.

Contribution

It presents a novel selective re-training framework based on IAN connectivity to enhance segmentation accuracy under sparse labeling conditions.

Findings

Achieved DSC of 0.7956 on verification cases

Attained HD95 of 4.4905, outperforming previous methods

Won the ToothFairy2023 Challenge

Abstract

Inferior Alveolar Nerve (IAN) canal detection in CBCT is an important step in many dental and maxillofacial surgery applications to prevent irreversible damage to the nerve during the procedure.The ToothFairy2023 Challenge aims to establish a 3D maxillofacial dataset consisting of all sparse labels and partial dense labels, and improve the ability of automatic IAN segmentation. In this work, in order to avoid the negative impact brought by sparse labeling, we transform the mixed supervised problem into a semi-supervised problem. Inspired by self-training via pseudo labeling, we propose a selective re-training framework based on IAN connectivity. Our method is quantitatively evaluated on the ToothFairy verification cases, achieving the dice similarity coefficient (DSC) of 0.7956, and 95\% hausdorff distance (HD95) of 4.4905, and wining the champion in the competition. Code is available at https://github.com/GaryNico517/SSL-IAN-Retraining.