A computer-aided diagnosis (CAD) system based on convolutional neural networks for lung cancer diagnosis from 2D [¹⁸F]- PET/CT images.

ABSTRACT:

Objective

This study aims to automatically classify lung conditions into normal, non-small cell lung cancer (NSCLC), and small cell lung cancer (SCLC) using [¹⁸F] FDG PET/CT images and deep learning.

Methods

PET/CT scans from 146 patients (1974 scans) were retrospectively analyzed using two strategies: (1) transfer learning with pre-trained CNNs, and (2) a custom CNN (Res-SE Net) incorporating residual and squeeze-and-excitation (SE) modules. A patient-based data splitting approach was used to avoid data leakage. Models were trained and validated at the scan level and evaluated at the patient level using majority voting. Grad-CAM was employed to generate lesion-localization heatmaps.

Results

Among the seven evaluated CNN models, the proposed Res-SE Net demo

SUBMITTER: Karimpour M 

PROVIDER: S-EPMC12509238 | biostudies-literature | 2025 Oct

REPOSITORIES: biostudies-literature