Unlocking Robust Segmentation Across All Age Groups via Continual Learning

TL;DR

This paper addresses the challenge of segmenting pediatric CT images by adapting adult-trained models using continual learning, achieving high accuracy across age groups.

Contribution

It introduces a continual learning framework to improve segmentation performance on pediatric data, a novel approach in medical imaging segmentation.

Findings

Continual learning improves segmentation accuracy on pediatric CT images.

The best model achieves Dice scores of 0.90 on adults and 0.84 on children.

Data augmentation combined with continual learning enhances robustness.

Abstract

Most deep learning models in medical imaging are trained on adult data with unclear performance on pediatric images. In this work, we aim to address this challenge in the context of automated anatomy segmentation in whole-body Computed Tomography (CT). We evaluate the performance of CT organ segmentation algorithms trained on adult data when applied to pediatric CT volumes and identify substantial age-dependent underperformance. We subsequently propose and evaluate strategies, including data augmentation and continual learning approaches, to achieve good segmentation accuracy across all age groups. Our best-performing model, trained using continual learning, achieves high segmentation accuracy on both adult and pediatric data (Dice scores of 0.90 and 0.84 respectively).