Sign in Subscribe
COVID-19

Researchers Develop Machine Learning Model to Differentiate Between Breast Cancer Metastasis and COVID-19 Vaccination Axillary Lymphadenopathy

Strategic Insights

— 1 min read

A recent study published in European Radiology has demonstrated the potential of radiomics and machine learning in distinguishing between breast cancer metastasis and COVID-19 vaccination axillary lymphadenopathy. Researchers from Chaim Sheba Medical Center in Ramat-Gan, Israel, analyzed data from 53 breast cancer patients and 46 COVID-19 mRNA vaccine recipients. They extracted radiomics features from PET/CT images, which were then used to train machine learning models to identify differences between the two groups.

Key Takeaways:

  • The study used radiomics features from PET/CT images to train machine learning models to differentiate between breast cancer metastasis and COVID-19 vaccination axillary lymphadenopathy.
  • The researchers analyzed data from 53 breast cancer patients and 46 COVID-19 mRNA vaccine recipients, with 85 axillary lymph nodes from breast cancer patients and 80 from COVID-19-vaccinated individuals being evaluated.
  • The machine learning models used, K-nearest neighbors (KNN) and random forest (RF), showed statistically significant differences between the two groups in terms of first-order features (p < 0.001).
  • The accuracy of the models was evaluated using the area under the receiver-operator characteristic curve (AUC-ROC) score, with the best performance achieved by the RF model (AUC-ROC = 0.85 ± 0.03).
  • The study concluded that radiomics features and machine learning models may have a role in differentiating between benign and malignant axillary lymphadenopathy.

Statistics:

  • 53 breast cancer patients were included in the study.
  • 46 COVID-19 mRNA vaccine recipients were included in the study.
  • 85 axillary lymph nodes from breast cancer patients were analyzed.
  • 80 axillary lymph nodes from COVID-19-vaccinated individuals were analyzed.
  • The first-order features showed statistically significant differences between the two groups (p < 0.001).
  • The AUC-ROC score for the RF model was 0.85 ± 0.03.

Sources:

  • European Radiology: "Fdg Pet/ct Radiomics As a Tool To Differentiate Between Reactive Axillary Lymphadenopathy Following Covid-19 Vaccination and Metastatic Breast Cancer Axillary Lymphadenopathy: a Pilot Study".
  • Springer: "European Radiology"
  • Chaim Sheba Medical Center
  • Michal Eifer, et al. "Fdg Pet/ct Radiomics As a Tool To Differentiate Between Reactive Axillary Lymphadenopathy Following Covid-19 Vaccination and Metastatic Breast Cancer Axillary Lymphadenopathy: a Pilot Study". European Radiology, 2022;32(9):5921-5929.

Read more