Practical Window Setting Optimization for Medical Image Deep Learning

TL;DR

This paper introduces a trainable window setting optimization module for deep learning models to improve clinical CT image interpretation by automatically finding optimal display settings, enhancing abnormality detection.

Contribution

The proposed WSO module is fully trainable and improves CT analysis performance by optimizing window settings, a previously overlooked aspect in deep learning models for medical imaging.

Findings

WSO outperforms models with fixed window settings.

WSO enhances detection of intracranial hemorrhage and urinary stones.

The method is adaptable to other medical imaging modalities.

Abstract

The recent advancements in deep learning have allowed for numerous applications in computed tomography (CT), with potential to improve diagnostic accuracy, speed of interpretation, and clinical efficiency. However, the deep learning community has to date neglected window display settings - a key feature of clinical CT interpretation and opportunity for additional optimization. Here we propose a window setting optimization (WSO) module that is fully trainable with convolutional neural networks (CNNs) to find optimal window settings for clinical performance. Our approach was inspired by the method commonly used by practicing radiologists to interpret CT images by adjusting window settings to increase the visualization of certain pathologies. Our approach provides optimal window ranges to enhance the conspicuity of abnormalities, and was used to enable performance enhancement for intracranial hemorrhage and urinary stone detection. On each task, the WSO model outperformed models trained over the full range of Hounsfield unit values in CT images, as well as images windowed with pre-defined settings. The WSO module can be readily applied to any analysis of CT images, and can be further generalized to tasks on other medical imaging modalities.