What question did this study set out to answer?

To develop and validate a multi-modal fusion model that predicts invasiveness in lung adenocarcinoma subtypes using CT images and clinical variables.

May 14, 2026Open Access

Deep residual network fusing CT images and clinical variables to predict lung adenocarcinoma aggressiveness

JPJia PengSun Yat-sen University WZWenqiang ZhongSun Yat-sen University KLK W LiSun Yat-sen University

Key Points

To develop and validate a multi-modal fusion model that predicts invasiveness in lung adenocarcinoma subtypes using CT images and clinical variables.
Retrospective study including 431 patients with ground-glass nodules confirmed by pathology.
Constructed a multi-modal deep learning model based on ResNet50, integrating clinical characteristics and laboratory indicators.
Evaluated model performance with metrics such as accuracy, AUC, precision, recall, and F1-score through five-fold cross-validation.
The model achieved overall accuracy of 72.2%, precision of 95.6%, and an AUC of 0.87.
Per-class precision was 84.6% for AIS, 35.7% for MIA, and 84.4% for IAC with corresponding recalls of 57.9%, 29.4%, and 81.8%.
The fusion model outperformed both CT-only model (AUC 0.79) and human radiologists (senior 0.67, junior 0.57).

Abstract

Lung adenocarcinoma presenting as ground-glass nodules (GGNs) comprises three invasive subtypes (adenocarcinoma in situ AIS, minimally invasive adenocarcinoma MIA, invasive adenocarcinoma IAC) with distinct prognoses and management strategies. Preoperative discrimination of these subtypes remains challenging for radiologists, and existing deep learning models rarely integrate multi-modal data for reliable prediction. This study aimed to develop and internally validate a multi-modal fusion framework based on the standard ResNet50 architecture, integrating CT images, clinical variables, and tumor markers, to improve the preoperative prediction of ground-glass nodule invasiveness. A retrospective study was conducted including 431 patients with pathologically confirmed ground-glass nodules. All patients underwent standard chest computed tomography before surgery. A multi-modal deep learning model was constructed based on the ResNet50 network, combined with clinical characteristics and laboratory indicators. Model performance was evaluated using accuracy, area under the receiver operating characteristic curve, precision, recall, and F1-score with five-fold cross-validation. The proposed multi-modal model achieved an overall accuracy of 72.2%, precision of 95.6%, negative predictive value of 96.0%, weighted F1-score of 40.0%, and multiclass Matthews correlation coefficient of 73.1% in the three-class classification of AIS, MIA, and IAC. Per-class analysis showed precision of 84.6%, 35.7%, and 84.4% and recall of 57.9%, 29.4%, and 81.8% for AIS, MIA, and IAC, respectively. The fusion model yielded a macro-average AUC of 0.87, which was higher than the CT-only model (0.79) and both the senior (0.67) and junior radiologists (0.57). The model demonstrated superior diagnostic performance compared to human readers, particularly for the challenging MIA subtype. This multi-modal deep learning model combining CT images, clinical variables, and serum tumor markers enables accurate and robust three-class classification of AIS, MIA, and IAC in ground-glass nodules. The proposed model outperforms both human radiologists and the imaging-only model, suggesting its potential as a reliable auxiliary tool to improve preoperative prediction of lung adenocarcinoma invasiveness and assist clinical decision-making.

Ask AI

Helpful

Bookmark

View Full Paper