Radiomics for Distinguishing Benign and Malignant Lung Nodules

NCT07852000 | Not Yet Recruiting

• 1 other identifier: CT Radiomics for Lung Nodules
• Study Type: observational
• Target: 255
• Locations: 0 countries
• Sites: N/A

Brief Summary

Differentiating between benign (non-cancerous) and malignant (cancerous) pulmonary nodules is a critical step in determining appropriate patient management and cancer treatment planning. Conventional evaluation using chest computed tomography (CT) relies on visual inspection of features such as size, shape, and borders; however, benign and malignant nodules frequently exhibit overlapping characteristics, which often necessitates invasive biopsy procedures. Radiomics is an emerging analytical technique that extracts high-dimensional quantitative data from standard medical images, including tissue texture, density patterns, and complex spatial features, that cannot be detected by visual inspection alone. The purpose of this observational study is to evaluate the utility of CT-derived radiomics combined with machine learning algorithms to non-invasively differentiate between benign and malignant pulmonary nodules. Researchers will extract quantitative imaging features from chest CT scans of patients presenting with pulmonary nodules measuring less than 5 cm to train and validate predictive machine learning models, with the goal of improving non-invasive diagnostic accuracy and reducing unnecessary biopsy procedures.

Conditions

Solitary Pulmonary Nodule; Lung Neoplasms; Multiple Pulmonary Nodules

Keywords

Radiomics; Computed Tomography; Pulmonary Nodule

Outcome Measures

Primary Outcomes (1)

• Area Under the Receiver Operating Characteristic Curve (AUC) of the CT Radiomics Model

  The primary outcome is the discriminatory performance of the machine learning-based radiomics model in differentiating benign from malignant pulmonary nodules, quantified by the Area Under the Receiver Operating Characteristic Curve (AUC). AUC values range from 0.5 (no discrimination / performance equal to chance) to 1.0 (perfect diagnostic discrimination). Classification accuracy will be validated against histopathological confirmation or established clinical and radiological reference standards.

  Baseline

Study Arms (1)

Patients with Pulmonary Nodules

Patients presenting with pulmonary nodules measuring less than 5 cm on chest computed tomography (CT) images obtained within the preceding year. Participants undergo comprehensive clinical history taking, clinical examination, and CT image evaluation. Radiomic features, including nodule intensity, shape, texture analysis, and wavelet decompositions, are extracted from segmented nodule volumes and classified using machine learning algorithms to differentiate benign from malignant pulmonary lesions.

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Patients presenting to the Department of Diagnostic and Interventional Radiology at Assiut University Hospitals and the South Egypt Cancer Institute who have confirmed pulmonary nodules measuring less than 5 cm on chest computed tomography (CT) scans acquired within the preceding year.

You may qualify if:

• Patients presenting with pulmonary nodules measuring less than 5 cm in diameter.
• Chest computed tomography (CT) scan acquired within the preceding 1 year.

You may not qualify if:

• Patients with pulmonary nodules measuring greater than 5 cm in diameter.
• Chest computed tomography (CT) scan obtained more than 3 years prior.
• Low-quality or artifact-compromised CT images unsuitable for radiomic feature extraction.

Sponsors & Collaborators

• Assiut University: lead

MeSH Terms

Conditions

Solitary Pulmonary Nodule; Lung Neoplasms; Multiple Pulmonary Nodules

Condition Hierarchy (Ancestors)

Lung Diseases; Respiratory Tract Diseases; Respiratory Tract Neoplasms; Thoracic Neoplasms; Neoplasms by Site; Neoplasms

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: CROSS SECTIONAL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Resident

Study Record Dates

• First Submitted: September 25, 2026
• First Posted: October 1, 2026
• Study Start: October 1, 2026
• Primary Completion (Estimated): October 1, 2027
• Study Completion (Estimated): November 1, 2027
• Last Updated: October 1, 2026

Record last verified: 2026-09