CTARR: A fast and robust method for identifying anatomical regions on CT images via atlas registration

TL;DR

CTARR is a fast, atlas registration-based method that accurately identifies and crops anatomical regions in CT images, enhancing deep learning tasks by reducing computational load and improving focus on relevant areas.

Contribution

The paper introduces CTARR, a novel, robust, and fast atlas registration method for automatic anatomical region recognition in CT images, applicable across multiple body regions.

Findings

Achieves 97.45-100% accuracy in preserving regions of interest

Operates in 0.1-0.21 seconds per scan

Reduces segmentation runtime by 2.0-12.7 times

Abstract

Medical image analysis tasks often focus on regions or structures located in a particular location within the patient's body. Often large parts of the image may not be of interest for the image analysis task. When using deep-learning based approaches, this causes an unnecessary increases the computational burden during inference and raises the chance of errors. In this paper, we introduce CTARR, a novel generic method for CT Anatomical Region Recognition. The method serves as a pre-processing step for any deep learning-based CT image analysis pipeline by automatically identifying the pre-defined anatomical region that is relevant for the follow-up task and removing the rest. It can be used in (i) image segmentation to prevent false positives in anatomically implausible regions and speeding up the inference, (ii) image classification to produce image crops that are consistent in their anatomical context, and (iii) image registration by serving as a fast pre-registration step. Our proposed method is based on atlas registration and provides a fast and robust way to crop any anatomical region encoded as one or multiple bounding box(es) from any unlabeled CT scan of the brain, chest, abdomen and/or pelvis. We demonstrate the utility and robustness of the proposed method in the context of medical image segmentation by evaluating it on six datasets of public segmentation challenges. The foreground voxels in the regions of interest are preserved in the vast majority of cases and tasks (97.45-100%) while taking only fractions of a seconds to compute (0.1-0.21s) on a deep learning workstation and greatly reducing the segmentation runtime (2.0-12.7x). Our code is available at https://github.com/ThomasBudd/ctarr.