Related Experiment Video
Updated: Apr 11, 2026
07:53
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
2.3K
Multimodal AI for pneumonia and lung cancer classification using x-ray and HRCT
Chauhan Pradip1, Chauhan Girish2, Chauhan Bhoomika3
1Department of Anatomy, All India Institute of Medical Sciences, Rajkot, Gujarat, India.
Bioinformation
|April 10, 2026
Summary
DeepScan, an AI model using both Chest X-rays and HRCT scans, significantly improves diagnosing lung cancer and pneumonia. This multimodal approach enhances accuracy and aids earlier patient intervention.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Chest X-rays and High-Resolution Computed Tomography (HRCT) are vital for diagnosing lung conditions like pneumonia and lung cancer.
- Current diagnostic accuracy limitations necessitate advanced methods.
Purpose of the Study:
- To develop and evaluate DeepScan, a multimodal artificial intelligence (AI) model for improved diagnosis of pneumonia and lung cancer.
- To assess the efficacy of combining Chest X-ray and HRCT data using AI.
Main Methods:
- Developed DeepScan, a multimodal AI model integrating Convolutional Neural Networks (CNNs).
- Utilized ResNet-50 for X-ray analysis and DenseNet-121 for HRCT data.
- Employed a late-fusion network architecture trained on public datasets.
- Validated the model on a test set of 2,000 patients.
Main Results:
- DeepScan achieved 94.6% accuracy, 95.2% sensitivity, and 93.9% specificity.
- The model demonstrated a high Area Under the Curve (AUC) of 0.97.
- Outperformed single-modality AI models in diagnostic performance.
- Reduced false negatives for early-stage lung cancer detection.
Conclusions:
- Multimodal AI integration significantly enhances diagnostic accuracy for lung cancer and pneumonia compared to single imaging types.
- DeepScan shows potential for earlier disease intervention and improved clinical workflow efficiency.
- AI-driven analysis of combined imaging modalities offers a promising advancement in thoracic diagnostics.
