Related Experiment Video
Updated: Aug 6, 2026
10:26
A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
First steps in clinical implementation of a lung nodule classification method by artificial intelligence
Abderrazzak Ajertil1,2, Zineb Farahat2, Abla Bouallou2,3
1Department of Radiology, Cheikh Zaid International University Hospital, Rabat, Morocco.
Frontiers in Radiology
|August 5, 2026
Summary
A new artificial intelligence (AI) model using deep learning accurately classifies pulmonary nodules from CT scans. This AI tool shows promise for improving early lung cancer detection and aiding radiologists in clinical workflows.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Lung cancer is a leading cause of cancer mortality globally, making early detection critical for improving patient survival rates.
- Computed tomography (CT) is the standard imaging technique for lung cancer screening.
- Artificial Intelligence (AI), especially Deep Learning (DL), has advanced automated medical image analysis and diagnostic capabilities.
Purpose of the Study:
- To develop and evaluate a Convolutional Neural Network (CNN)-based approach for classifying pulmonary nodules.
- To assess the model's performance using a diverse dataset including public and real-world CT scans.
- To explore the clinical potential of AI in assisting radiologists with nodule classification.
Main Methods:
- A CNN model was developed for pulmonary nodule classification.
- The model was trained on a merged dataset comprising public CT image databases (IQ-OTH/NCCD, SPIE-AAPM) and real-world scans from Cheikh Zaid Hospital.
- The dataset included 1,103 malignant, 508 benign, and 427 normal pulmonary nodule images.
- Contrast Limited Adaptive Histogram Equalization (CLAHE) was used for image contrast enhancement prior to training.
Main Results:
- The proposed CNN model achieved high performance metrics for a three-class classification task.
- Achieved accuracy of 99.84%, precision of 99.97%, sensitivity of 99.84%, and specificity of 99.82%.
- Demonstrated the robustness and clinical potential of the AI model.
Conclusions:
- The AI-based model shows significant promise for assisting radiologists in classifying indeterminate pulmonary nodules.
- This study represents a crucial step towards integrating AI tools into routine radiology workflows at Cheikh Zaid Hospital.
- The findings highlight the feasibility of AI-driven lung nodule classification in real-world clinical settings.
