AI-Based Assessment of Chest Diseases on Chest X-ray Imaging Using DenseNet-121 and a CheXpert-Trained Model
AI-Based Assessment of Chest Diseases on Chest X-ray Imaging Using DenseNet-121 and a CheXpert-Trained Model: A Comparative Diagnostic Accuracy Study
Chest X-rays are widely used to detect thoracic and lung conditions, but reviewing high volumes of radiographs can lead to workload strain and variation between interpreters. Artificial intelligence (AI), particularly deep learning neural networks like DenseNet-121, has shown strong potential to assist clinicians with automated image interpretation. However, AI models trained on large international datasets, such as CheXpert, may perform differently across distinct patient populations due to variations in imaging technique, patient demographics, and disease presentation.
The primary purpose of this study is to compare the diagnostic accuracy of a DenseNet-121 model trained or fine-tuned on local data against a DenseNet-121 model pretrained on the CheXpert dataset for identifying thoracic pathologies. Both models will evaluate de-identified frontal chest radiographs from adult patients. Model predictions will be compared against a reference standard established by expert radiologist consensus, with discordant findings resolved using chest computed tomography (CT). Findings will evaluate whether local model adaptation improves diagnostic precision and workflow efficiency in clinical settings.
Study Overview
Status
Status
Conditions
Conditions
Study Type
Study Type
Enrollment (Estimated)
Enrollment
Participation Criteria
Eligibility Criteria
Eligibility Criteria
Ages Eligible for Study
Accepts Healthy Volunteers
Genders Eligible for Study
Sampling Method
Study Population
The study population comprises adult patients aged 18 years and older who undergo frontal chest radiography within institutional radiology departments. Eligible subjects have de-identified, diagnostic-quality frontal chest X-ray images with available reference diagnostic labels verified through expert radiologist consensus and chest CT adjudication. Pediatric cases, non-diagnostic images, incomplete metadata, and duplicate scans are excluded.
Description
Inclusion Criteria:
- Adult patients aged 18 years or older.
- Undergoing frontal chest radiography.
- Diagnostic-quality frontal chest radiographs.
- Availability of reference diagnostic labels (verified by expert radiologist consensus or chest CT).
Exclusion Criteria:
- Pediatric patients (under 18 years of age).
- Non-diagnostic or poor-quality chest radiographs.
- Incomplete clinical or imaging metadata.
- Duplicate radiographic images or repeat patient entries.
Study Plan
How is the study designed?
Design Details
- Observational Models: COHORT
- Time Perspectives: CROSS_SECTIONAL
Number of groups / cohorts
Cohorts and Interventions
Group / CohortGroup / Cohort |
|---|
|
Adult Frontal Chest Radiography Cohort
This cohort includes adult patients aged 18 years and older who undergo diagnostic-quality frontal chest radiography with available reference diagnostic labels. De-identified chest radiographs from this group are evaluated by two deep learning architectures: a study-trained/fine-tuned DenseNet-121 convolutional neural network and a CheXpert-pretrained DenseNet-121 model. Model predictions are evaluated against reference standards derived from expert radiologist consensus, with discordant cases adjudicated via chest computed tomography (CT). |
What is the study measuring?
Primary Outcome Measures
Primary Outcome Measures
Outcome Measure |
Measure Description |
Time Frame |
|---|---|---|
|
Area Under the Receiver Operating Characteristic Curve (AUROC)
Time Frame: Baseline
|
AUROC will be calculated to assess and compare the diagnostic performance of the study-trained DenseNet-121 model versus the CheXpert-pretrained DenseNet-121 model in detecting thoracic pathologies on frontal chest radiographs. AI predictions will be compared against the reference standard of expert radiologist consensus, with discordant findings adjudicated by chest CT. AUROC values range from 0.5 (no discrimination) to 1.0 (perfect discrimination). |
Baseline
|
Collaborators and Investigators
Sponsor
Sponsor
Study record dates
Study Major Dates
Study Start (Estimated)
Study Start
Primary Completion (Estimated)
Primary Completion
Study Completion (Estimated)
Study Completion
Study Registration Dates
First Submitted
First Submitted
First Submitted That Met QC Criteria
First Submitted That Met QC Criteria
First Posted (Actual)
First Posted
Study Record Updates
Last Update Posted (Actual)
Last Update Posted
Last Update Submitted That Met QC Criteria
Last Update Submitted That Met QC Criteria
Last Verified
Last Verified
More Information
Terms related to this study
Keywords
Additional Relevant MeSH Terms
Other Study ID Numbers
Other Study ID Numbers
- AI Chest X-Ray Assessment
This information was retrieved directly from the website clinicaltrials.gov without any changes. If you have any requests to change, remove or update your study details, please contact register@clinicaltrials.gov. As soon as a change is implemented on clinicaltrials.gov, this will be updated automatically on our website as well.