JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Construction and analysis of the invasive prediction model for pulmonary nodules: based on clinical, CT image and DNA methylation characteristics

  • 0Department of Thoracic Surgery, Xiamen Humanity Hospital, Fujian Medical University, Xiamen, China.
Journal of Thoracic Disease +

|

April 14, 2025

|

April 14, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed a predictive model to distinguish between microinvasive adenocarcinoma (MIA) and invasive carcinoma (IC) in pulmonary nodules. The comprehensive model, integrating clinical, CT, and methylation data, demonstrated high accuracy and sensitivity, outperforming models using single data types.

Area Of Science

  • Pulmonary medicine
  • Oncology
  • Medical imaging
  • Molecular diagnostics

Background

  • Accurate differentiation between microinvasive adenocarcinoma (MIA) and invasive carcinoma (IC) in pulmonary nodules is crucial for clinical management.
  • Current diagnostic methods may have limitations in precisely classifying these early-stage lung cancers.
  • Developing a robust predictive model can aid in treatment decisions and improve patient outcomes.

Purpose Of The Study

  • To construct and validate a predictive model for differentiating MIA or IC in pulmonary nodules.
  • To evaluate the efficacy of combining clinical, computed tomography (CT) image, and peripheral blood methylation data for improved prediction.
  • To compare the performance of a comprehensive model against models based on clinical/image features or methylation features alone.

Main Methods

  • Collected clinical, CT image, and peripheral blood methylation data from 294 patients, categorized by postoperative pathology into invasive (MIA/IC) and non-invasive groups.
  • Developed predictive models using logistic regression, analyzing clinical and image features, methylation features, and a comprehensive combination of both.
  • Validated model performance using a separate training and validation set split, assessing metrics like Area Under the Curve (AUC), accuracy, sensitivity, and specificity.

Main Results

  • The comprehensive model identified six key indicators: proportion of solid components, maximum CT value, and specific CpG sites (SH3BP5_338_CpG 4, PNPLA2_329_CpG 1, PNPLA2_329_CpG 4, ARHGAP35 476_CpG_5).
  • The comprehensive model achieved high performance with AUCs of 0.90 (training) and 0.87 (validation), and prediction accuracies of 82% in both sets.
  • The comprehensive model demonstrated superior predictive performance compared to models relying solely on clinical/image features or methylation features.

Conclusions

  • A predictive model integrating clinical, CT image, and methylation features effectively predicts the invasiveness of pulmonary nodules.
  • The developed comprehensive model shows satisfactory performance and warrants further investigation and refinement.
  • This multi-modal approach holds promise for improving the diagnostic accuracy of pulmonary nodule classification.

Keywords: