Evaluating Radiomics, Deep Learning, and Hybrid Models for Forecasting Hidden Pleural Spread in Non-Small Cell Lung Cancer Patients: A Retrospective Multicenter Analysis - Summary - MDSpire

Evaluating Radiomics, Deep Learning, and Hybrid Models for Forecasting Hidden Pleural Spread in Non-Small Cell Lung Cancer Patients: A Retrospective Multicenter Analysis

  • By

  • Tao Bao

  • Xiaoguang Li

  • Yuanlin Deng

  • Liang Chen

  • Weijie Sun

  • Mingjian Ge

  • Jigang Dai

  • Xiaolong Zhao

  • Xu Chen

  • Liang Zhang

  • Lei Bao

  • Wei Guo

  • October 29, 2025

  • 0 min

Share

Objective:

To develop and validate a noninvasive tool for preoperative identification of occult pleural dissemination (PD) in high-risk non-small cell lung cancer (NSCLC) patients, thereby preventing unnecessary surgeries and improving patient outcomes.

Key Findings:
  • The study identified independent risk factors for occult PD, including age, CEA levels, and pleural invasion, which may inform clinical decision-making.
Interpretation:

The fusion model effectively integrates complementary information from radiomics and deep learning, enhancing the preoperative detection of occult PD in NSCLC patients.

Limitations:
  • The study is retrospective and may have inherent biases, which could limit the generalizability of the findings. Additionally, the external validation cohort was limited in size, potentially affecting the robustness of the model.
Conclusion:

The developed hybrid model shows promise in accurately predicting occult pleural dissemination in NSCLC patients, potentially reducing unnecessary surgeries and paving the way for future research in this area.

Original Source(s)

Related Content