JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Prediction of the benign and malignant nature of masses in COPD background based on Habitat-based enhanced CT radiomics modeling: A preliminary study

  • 0College of Traditional Chinese Medicine, Hubei University of Chinese Medicine, Wuhan, Hubei, China.
Technology and Health Care : Official Journal of the European Society for Engineering and Medicine +

|

March 22, 2024

|

March 22, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Differentiating chronic obstructive pulmonary disease (COPD)-peripheral bronchogenic carcinoma (COPD-PBC) from inflammatory masses is challenging. A combined model using clinical data and Habitat-based enhanced CT radiomics (HECT radiomics) effectively predicts COPD-PBC, offering a novel diagnostic approach.

Area Of Science

  • Pulmonology
  • Radiology
  • Oncology

Background

  • Distinguishing chronic obstructive pulmonary disease (COPD)-peripheral bronchogenic carcinoma (COPD-PBC) from inflammatory masses presents a clinical challenge.
  • Accurate differentiation is crucial for appropriate patient management and treatment planning.

Purpose Of The Study

  • To develop and validate a predictive model for COPD-PBC using clinical data and preoperative Habitat-based enhanced CT radiomics (HECT radiomics).
  • To assess the diagnostic performance of HECT radiomics in differentiating COPD-PBC from inflammatory masses.

Main Methods

  • A retrospective analysis of 232 patients with pathologically confirmed PBC or inflammatory masses was performed.
  • Predictive models were built using clinical data and HECT radiomics features, including texture analysis of enhanced CT areas.
  • A combination model integrating clinical factors and radiomics was developed and validated in training and testing sets.

Main Results

  • Univariate analysis identified female gender, tumor morphology, CEA, Cyfra21-1, CT enhancement pattern, and Habitat-Radscore B/C as significant predictors of COPD-PBC.
  • The combination model demonstrated superior predictive performance (AUC: 0.894) compared to clinical data alone (AUC: 0.758) and radiomics alone (AUC: 0.828).
  • Decision curve analysis confirmed the superior clinical utility of the combination model, which was further validated in the external testing set.

Conclusions

  • A combined model integrating clinical data and HECT radiomics offers a non-invasive and efficient method for differentiating COPD-PBC.
  • This approach can aid in diagnosis, treatment selection, and clinical decision-making for patients with suspected COPD-PBC.

Keywords: