Utilization of Deep Learning Architectures for the Automated Detection of Pulmonary tumor and Tuberculosis in Thoracic Radiographic Imaging
Principal Investigator
Name
Mutlu Avci
Degrees
Ph.D.,
Institution
Çukurova University
Position Title
Professor and Head of Biomedical Engineering Department
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1380
Initial CDAS Request Approval
Jan 10, 2025
Title
Utilization of Deep Learning Architectures for the Automated Detection of Pulmonary tumor and Tuberculosis in Thoracic Radiographic Imaging
Summary
This study analyzes chest radiographs using sophisticated deep learning (DL) algorithms to identify tuberculosis and lung cancer. The goal of the study is to create and validate a multimodal deep-learning model that can distinguish between tuberculosis and lung cancer in a screening population. The project intends to improve diagnosis accuracy and assist clinical decision-making by utilizing state-of-the-art image processing and artificial intelligence techniques.
Aims
1. Create a deep learning-based AI model for lung cancer and tuberculosis detection outside of chest radiography.
2. Improve diagnostic accuracy by enhancing sensitivity, specificity, and illness detection confidence.
3. Assist clinical decision-making with AI-driven insights, enabling scalable and reliable dual-disease screening.
4. Analyze lung cancer nodules and TB patterns, examining ambiguity levels and diagnostic outcomes.
Collaborators
Amna Khan M.Sc Cukurova University Biomedical Engineering Department
Zkeia Abdalla Abdrhman Jazam Cukurova University Computer Engineering Department