Robust End-to-End Focal Liver Lesion Detection using Unregistered Multiphase Computed Tomography Images

TL;DR

This paper introduces a fully automated, end-to-end deep learning framework for detecting focal liver lesions in multiphase CT images that is robust to misalignments, improving clinical applicability and outperforming previous methods.

Contribution

The study proposes an attention-guided multiphase alignment in feature space, enabling robust FLL detection without reliance on image registration quality.

Findings

Outperforms previous state-of-the-art methods on a large dataset.

Significantly reduces performance degradation with misaligned images.

Enhances clinical adoption of deep-learning-based detection systems.

Abstract

The computer-aided diagnosis of focal liver lesions (FLLs) can help improve workflow and enable correct diagnoses; FLL detection is the first step in such a computer-aided diagnosis. Despite the recent success of deep-learning-based approaches in detecting FLLs, current methods are not sufficiently robust for assessing misaligned multiphase data. By introducing an attention-guided multiphase alignment in feature space, this study presents a fully automated, end-to-end learning framework for detecting FLLs from multiphase computed tomography (CT) images. Our method is robust to misaligned multiphase images owing to its complete learning-based approach, which reduces the sensitivity of the model's performance to the quality of registration and enables a standalone deployment of the model in clinical practice. Evaluation on a large-scale dataset with 280 patients confirmed that our method outperformed previous state-of-the-art methods and significantly reduced the performance degradation for detecting FLLs using misaligned multiphase CT images. The robustness of the proposed method can enhance the clinical adoption of the deep-learning-based computer-aided detection system.