Stochastic tissue window normalization of deep learning on computed tomography

TL;DR

This study evaluates tissue window normalization in CT deep learning models, introduces a stochastic normalization method to enhance generalizability across diverse datasets, and demonstrates its effectiveness through multi-organ segmentation experiments.

Contribution

The paper proposes a novel stochastic tissue window normalization (SWN) method that improves model generalizability across different CT cohorts compared to traditional normalization techniques.

Findings

SWN outperforms traditional methods on diverse datasets

Traditional normalization performs best on same-scanner data

SWN provides better generalization across different CT contrasts and pathologies

Abstract

Tissue window filtering has been widely used in deep learning for computed tomography (CT) image analyses to improve training performance (e.g., soft tissue windows for abdominal CT). However, the effectiveness of tissue window normalization is questionable since the generalizability of the trained model might be further harmed, especially when such models are applied to new cohorts with different CT reconstruction kernels, contrast mechanisms, dynamic variations in the acquisition, and physiological changes. We evaluate the effectiveness of both with and without using soft tissue window normalization on multisite CT cohorts. Moreover, we propose a stochastic tissue window normalization (SWN) method to improve the generalizability of tissue window normalization. Different from the random sampling, the SWN method centers the randomization around the soft tissue window to maintain the specificity for abdominal organs. To evaluate the performance of different strategies, 80 training and 453 validation and testing scans from six datasets are employed to perform multi-organ segmentation using standard 2D U-Net. The six datasets cover the scenarios, where the training and testing scans are from (1) same scanner and same population, (2) same CT contrast but different pathology, and (3) different CT contrast and pathology. The traditional soft tissue window and nonwindowed approaches achieved better performance on (1). The proposed SWN achieved general superior performance on (2) and (3) with statistical analyses, which offers better generalizability for a trained model.