Pulmonary nodules (PNs) are detected in approximately 1.6 million adults in the United States each year, most of which are benign. Lung cancer is the leading cause of cancer death, making accurate differentiation between malignant and benign PNs essential to avoid unnecessary invasive procedures and delays in diagnosis. Radiomics-based computer-aided diagnosis (CAD) tools utilizing artificial intelligence analyze CT features to estimate malignancy risk and may enhance clinical decision-making. The Assessment of a Radiomics‑based Computer‑Aided Diagnosis Tool for Pulmonary Nodules (ARCADES) pragmatic randomized controlled trial evaluates whether integrating a CAD‑generated malignancy risk score into routine clinical assessment improves the appropriateness of initial management decisions for newly detected PNs. Eligible patients are aged 35–89 years with an 8–30 mm solid or part‑solid PN, identified in University of Pennsylvania Health System pulmonary nodule clinics, and randomized 1:1 to clinician assessment alone or clinician assessment plus CAD‑based risk stratification. The primary outcome is appropriate management at the index visit, with follow‑up at 12 and 24 months to confirm diagnosis. To date, 101 patients have been enrolled (49 in the control group and 50 in the intervention group for baseline analysis), with baseline characteristics largely comparable between the groups. Notable differences include a higher proportion of current smokers and greater spiculation rates in the intervention arm. Results will determine whether CAD integration can meaningfully improve PN management decisions and better align initial treatment strategies with actual nodule outcomes.