Related Experiment Video
Updated: Aug 22, 2025
Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
Development and Validation of a Risk Assessment Model for Pulmonary Nodules Using Plasma Proteins and Clinical
Anil Vachani1, Stephen Lam2, Pierre P Massion3
1Pulmonary, Allergy, and Critical Care Division, Department of Medicine, University of Pennsylvania, Philadelphia, PA; Corporal Michael J. Crescenz VA Medical Center, Department of Medicine, Philadelphia, PA.
A novel risk reclassification model (RRM) using plasma proteins and clinical factors accurately identifies patients with pulmonary nodules (PNs). This model improves risk stratification, potentially reducing unnecessary invasive testing and improving diagnostic timeliness.
Area of Science:
- Pulmonary Medicine
- Biomarker Discovery
- Machine Learning in Healthcare
Background:
- Pulmonary nodule (PN) risk assessment is crucial for timely diagnosis and avoiding unnecessary invasive procedures.
- Current risk assessment models have limitations, leading to diagnostic delays and overtreatment.
Purpose of the Study:
- To evaluate the accuracy of a novel PN risk model incorporating plasma proteins and clinical factors.
- To compare the novel model's performance against the established Mayo Clinic model.
Main Methods:
- Development of assays for seven plasma proteins using magnetic nanosensor technology.
- Application of machine learning to identify an optimal risk algorithm in a training cohort (n=429).
- Validation of the algorithm in a separate cohort (n=489) and comparison with the Mayo Clinic model.
Main Results:
- A support vector machine algorithm (Risk Reclassification Model - RRM) combining seven plasma proteins and six clinical factors achieved an AUC of 0.87.
- The RRM significantly reclassified patients, decreasing intermediate-risk cases and increasing low- and very high-risk stratification.
- The RRM demonstrated improved specificity in low-risk and sensitivity in very high-risk classifications compared to the Mayo Clinic model.
Conclusions:
- The novel RRM shows promise for enhancing pulmonary nodule risk assessment.
- Accurate reclassification into low- and very high-risk categories by the RRM can potentially optimize patient management pathways.
