Unlocking Robust Segmentation Across All Age Groups via Continual Learning

Document detail

ID

oai:arXiv.org:2404.13185

Topic

Electrical Engineering and Systems... Computer Science - Computer Vision...

Author

Liu, Chih-Ying Valanarasu, Jeya Maria Jose Gonzalez, Camila Langlotz, Curtis Ng, Andrew Gatidis, Sergios

Year

2024

listing date

4/24/2024

Keywords

trained ct pediatric

Metrics

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).

Liu, Chih-Ying,Valanarasu, Jeya Maria Jose,Gonzalez, Camila,Langlotz, Curtis,Ng, Andrew,Gatidis, Sergios, 2024, Unlocking Robust Segmentation Across All Age Groups via Continual Learning