From Volumes to Slices: Computationally Efficient Contrastive Learning for Sequential Abdominal CT Analysis

TL;DR

This paper introduces 2D-VoCo, an efficient slice-level contrastive learning method for 3D abdominal CT analysis that reduces computational costs and improves injury classification performance using unlabeled data.

Contribution

The authors adapt volume contrastive learning to a slice-based approach, significantly decreasing resource requirements while enhancing model accuracy in medical imaging tasks.

Findings

2D-VoCo improves performance metrics over training from scratch.

Pre-training with 2D-VoCo enhances multi-organ injury classification.

The method reduces computational costs compared to 3D approaches.

Abstract

The requirement for expert annotations limits the effectiveness of deep learning for medical image analysis. Although 3D self-supervised methods like volume contrast learning (VoCo) are powerful and partially address the labeling scarcity issue, their high computational cost and memory consumption are barriers. We propose 2D-VoCo, an efficient adaptation of the VoCo framework for slice-level self-supervised pre-training that learns spatial-semantic features from unlabeled 2D CT slices via contrastive learning. The pre-trained CNN backbone is then integrated into a CNN-LSTM architecture to classify multi-organ injuries. In the RSNA 2023 Abdominal Trauma dataset, 2D-VoCo pre-training significantly improves mAP, precision, recall, and RSNA score over training from scratch. Our framework provides a practical method to reduce the dependency on labeled data and enhance model performance in clinical CT analysis. We release the code for reproducibility. https://github.com/tkz05/2D-VoCo-CT-Classifier