Learning Generalizable Features for Tibial Plateau Fracture Segmentation Using Masked Autoencoder and Limited Annotations

TL;DR

This paper introduces a masked autoencoder-based training strategy for tibial plateau fracture segmentation in CT scans, significantly reducing annotation needs while improving accuracy and generalizability across datasets.

Contribution

The study presents a novel MAE pretraining approach that captures global and detailed features from unlabeled data, enhancing fracture segmentation with limited annotations.

Findings

Achieved 95.81% Dice score with only 20 labeled cases.

Outperformed existing semi-supervised methods.

Demonstrated strong transferability to external datasets.

Abstract

Accurate automated segmentation of tibial plateau fractures (TPF) from computed tomography (CT) requires large amounts of annotated data to train deep learning models, but obtaining such annotations presents unique challenges. The process demands expert knowledge to identify diverse fracture patterns, assess severity, and account for individual anatomical variations, making the annotation process highly time-consuming and expensive. Although semi-supervised learning methods can utilize unlabeled data, existing approaches often struggle with the complexity and variability of fracture morphologies, as well as limited generalizability across datasets. To tackle these issues, we propose an effective training strategy based on masked autoencoder (MAE) for the accurate TPF segmentation in CT. Our method leverages MAE pretraining to capture global skeletal structures and fine-grained fracture details from unlabeled data, followed by fine-tuning with a small set of labeled data. This strategy reduces the dependence on extensive annotations while enhancing the model's ability to learn generalizable and transferable features. The proposed method is evaluated on an in-house dataset containing 180 CT scans with TPF. Experimental results demonstrate that our method consistently outperforms semi-supervised methods, achieving an average Dice similarity coefficient (DSC) of 95.81%, average symmetric surface distance (ASSD) of 1.91mm, and Hausdorff distance (95HD) of 9.42mm with only 20 annotated cases. Moreover, our method exhibits strong transferability when applying to another public pelvic CT dataset with hip fractures, highlighting its potential for broader applications in fracture segmentation tasks.