Anatomical Data Augmentation For CNN based Pixel-wise Classification

TL;DR

This paper introduces an anatomical data augmentation method using adjacent CT slices to improve CNN-based pixel-wise classification of hepatic lesions, demonstrating significant accuracy improvements.

Contribution

The study presents a novel anatomical data augmentation technique leveraging adjacent CT slices to enhance CNN training for medical image segmentation.

Findings

3% improvement in success rate

5% increase in classification accuracy

4% higher Dice coefficient

Abstract

In this work we propose a method for anatomical data augmentation that is based on using slices of computed tomography (CT) examinations that are adjacent to labeled slices as another resource of labeled data for training the network. The extended labeled data is used to train a U-net network for a pixel-wise classification into different hepatic lesions and normal liver tissues. Our dataset contains CT examinations from 140 patients with 333 CT images annotated by an expert radiologist. We tested our approach and compared it to the conventional training process. Results indicate superiority of our method. Using the anatomical data augmentation we achieved an improvement of 3% in the success rate, 5% in the classification accuracy, and 4% in Dice.