Abstract:Rationale and objectives: Evaluating the capability of CT nomograms and CT-based radiomics nomograms to differentiate between Bronchiolar Adenoma (BA) and Early-stage Lung Adenocarcinoma (LUAD). Materials and methods: In this retrospective study; we analyzed data from 226 patients who were treated at our institution and pathologically confirmed to have either BA or Early-stage LUAD. Patients were randomly divided into a training cohort (n=158) and a testing cohort (n=68). All CT images were independently analyzed and measured by two radiologists using conventional computed tomography. Clinical predictive factors were identified using logistic regression. Multivariable logistic regression analysis was used to construct differential diagnostic models for BA and early-stage LUAD, including traditional CT and radiomics models. The performance of the models was determined based on the area under the receiver operating characteristic curve, discrimination ability, and decision curve analysis (DCA). Results: Lesion shape, tumor-lung interface, and pleural retraction signs were identified as independent clinical predictors. The areas under the curve for the CT nomogram, radiomic features, and radiomics nomogram were 0.854, 0.769, and 0.901, respectively. Both the CT nomogram and the radiomics nomogram demonstrated good generalizability in distinguishing between the two entities. DCA indicated that the nomograms achieved a higher net benefit compared to the use of radiomic features alone. Conclusion: The two preoperative nomograms hold significant value in differentiating between patients with BA and those with Early-stage LUAD, and they contribute to informed clinical treatment decision-making.

Preoperative CT and Radiomics Nomograms for Distinguishing Bronchiolar Adenoma and Early-Stage Lung Adenocarcinoma