Invasiveness Prediction of Pulmonary Adenocarcinomas Using Deep Feature Fusion Networks

TL;DR

This paper presents a deep feature fusion network that combines radiomics and deep-learning features to improve the prediction of pulmonary adenocarcinoma invasiveness from CT scans, potentially aiding early diagnosis.

Contribution

The study introduces a novel deep feature fusion approach that leverages the complementarity of radiomics and deep features for better invasiveness prediction.

Findings

Fusion network outperforms individual feature-based models

Effective on a private dataset of 676 patients

Improves early diagnosis of pulmonary adenocarcinoma

Abstract

Early diagnosis of pathological invasiveness of pulmonary adenocarcinomas using computed tomography (CT) imaging would alter the course of treatment of adenocarcinomas and subsequently improve the prognosis. Most of the existing systems use either conventional radiomics features or deep-learning features alone to predict the invasiveness. In this study, we explore the fusion of the two kinds of features and claim that radiomics features can be complementary to deep-learning features. An effective deep feature fusion network is proposed to exploit the complementarity between the two kinds of features, which improves the invasiveness prediction results. We collected a private dataset that contains lung CT scans of 676 patients categorized into four invasiveness types from a collaborating hospital. Evaluations on this dataset demonstrate the effectiveness of our proposal.