JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Development and validation of a novel echocardiography-based nomogram for the streamlined classification of cardiac tumors in cancer patients

  • 0Department of Medical Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Quantitative Imaging in Medicine and Surgery +

|

March 31, 2025

|

March 31, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed a machine learning model using echocardiography to classify cardiac tumors in cancer patients. The model effectively differentiates benign from malignant tumors, aiding treatment planning.

Area Of Science

  • Cardiology
  • Oncology
  • Medical Imaging
  • Machine Learning

Background

  • Echocardiography is a first-line tool for cardiac tumor screening, but its specificity is limited.
  • Accurate differentiation of cardiac tumors is essential for effective treatment planning.
  • Cancer patients with extracardiac malignancies may develop cardiac tumors requiring precise diagnosis.

Purpose Of The Study

  • To develop a streamlined classification model for cardiac tumors using echocardiographic data.
  • To improve the accuracy of distinguishing between benign and malignant cardiac tumors.
  • To integrate radiomics and clinical indicators for enhanced diagnostic performance.

Main Methods

  • A cohort of 215 echocardiographic clips from 121 cancer patients with cardiac tumors was analyzed.
  • Radiomics features were extracted, and a radiomics score (Rad-score) was computed using a machine learning framework.
  • A classification model was constructed by integrating the Rad-score with non-experience-dependent indicators (NDIs) and a nomogram was developed.

Main Results

  • Significant differences in Rad-scores and NDIs (age, tumor location, size) distinguished benign from malignant tumors.
  • Malignant tumors were associated with younger age, right-sided location, larger size, and lower Rad-scores.
  • The integrated model achieved strong classification performance (AUC: 0.873), comparable to senior physicians and superior to junior physicians.

Conclusions

  • A novel nomogram integrating radiomics and objective echocardiographic indicators effectively distinguishes malignant from benign cardiac tumors.
  • This approach enhances classification accuracy and decision-making in clinical settings.
  • The developed model offers a valuable tool for improving cardiac tumor diagnosis in cancer patients.

Keywords: