Evaluating Utility of Memory Efficient Medical Image Generation: A Study on Lung Nodule Segmentation

TL;DR

This paper introduces a memory-efficient diffusion model for generating synthetic lung CT images with nodules, which improves segmentation performance and addresses data scarcity in medical imaging.

Contribution

It presents a novel memory-efficient patch-wise DDPM for high-utility synthetic medical image generation, enabling effective data augmentation and training with limited real data.

Findings

Synthetic images achieve comparable segmentation accuracy to real data.

Augmentation with synthetic images improves segmentation performance.

The method efficiently manages memory constraints during image generation.

Abstract

The scarcity of publicly available medical imaging data limits the development of effective AI models. This work proposes a memory-efficient patch-wise denoising diffusion probabilistic model (DDPM) for generating synthetic medical images, focusing on CT scans with lung nodules. Our approach generates high-utility synthetic images with nodule segmentation while efficiently managing memory constraints, enabling the creation of training datasets. We evaluate the method in two scenarios: training a segmentation model exclusively on synthetic data, and augmenting real-world training data with synthetic images. In the first case, models trained solely on synthetic data achieve Dice scores comparable to those trained on real-world data benchmarks. In the second case, augmenting real-world data with synthetic images significantly improves segmentation performance. The generated images demonstrate their potential to enhance medical image datasets in scenarios with limited real-world data.