Liver Tumor Screening and Diagnosis in CT with Pixel-Lesion-Patient Network

TL;DR

This paper introduces the Pixel-Lesion-Patient Network (PLAN), a novel deep learning framework that improves liver tumor screening and diagnosis in CT scans by jointly segmenting and classifying lesions, achieving high accuracy and clinical relevance.

Contribution

The paper proposes a new transformer-based framework with enhanced sampling and global classification for liver tumor detection and diagnosis in CT, outperforming existing CNN and transformer models.

Findings

Achieves 95-96% sensitivity and specificity in tumor screening.

Outperforms CNN and transformer models in lesion detection and classification.

On par with senior radiologists in clinical reader study.

Abstract

Liver tumor segmentation and classification are important tasks in computer aided diagnosis. We aim to address three problems: liver tumor screening and preliminary diagnosis in non-contrast computed tomography (CT), and differential diagnosis in dynamic contrast-enhanced CT. A novel framework named Pixel-Lesion-pAtient Network (PLAN) is proposed. It uses a mask transformer to jointly segment and classify each lesion with improved anchor queries and a foreground-enhanced sampling loss. It also has an image-wise classifier to effectively aggregate global information and predict patient-level diagnosis. A large-scale multi-phase dataset is collected containing 939 tumor patients and 810 normal subjects. 4010 tumor instances of eight types are extensively annotated. On the non-contrast tumor screening task, PLAN achieves 95% and 96% in patient-level sensitivity and specificity. On contrast-enhanced CT, our lesion-level detection precision, recall, and classification accuracy are 92%, 89%, and 86%, outperforming widely used CNN and transformers for lesion segmentation. We also conduct a reader study on a holdout set of 250 cases. PLAN is on par with a senior human radiologist, showing the clinical significance of our results.