A NOVEL radiomics model utilising contrast-enhanced ultrasound (CEUS) imaging could revolutionise breast cancer diagnosis by enhancing accuracy and reducing unnecessary biopsies, according to recent research.
The study demonstrated the high performance of a combined radiomics model that integrates features from both within (intratumoral) and surrounding (peritumoral) breast tumours. This approach significantly outperformed models relying solely on intratumoral features, reducing false positives and minimising superfluous procedures.
CEUS is an imaging technique that provides detailed tumour morphology and monitors microcirculation, aiding in tumour characterisation. Despite its advantages, CEUS remains operator-dependent, necessitating quantitative and objective analysis to achieve consistent and reliable diagnostic outcomes.
The researchers developed their radiomics model using CEUS images from 333 women with breast lesions, collected between 2022 and 2024. These were divided into training and testing datasets, which included both benign and malignant cases. Breast cancer was present in 27.6% of patients across both groups.
Decision curve analysis further confirmed the combined model’s clinical value across a range of threshold probabilities, surpassing the intratumoral-only approach. The combined model demonstrated predictive probabilities from 5% to 95%, offering greater benefits compared to the intratumoral-only model.
The findings underscore the importance of analysing the peritumoral microenvironment, which plays a crucial role in tumour progression through processes like angiogenesis, immune cell activity, and stromal changes.
The researchers concluded that incorporating peritumoral imaging features could provide critical insights for diagnosing breast cancer, calling for further studies to refine radiomics approaches. By integrating these insights, clinicians could achieve earlier and more precise diagnoses, improving outcomes for patients.
Reference
Li G et al. Enhancing early breast cancer diagnosis with contrast-enhanced ultrasound radiomics: insights from intratumoral and peritumoral analysis. Clin Breast Cancer. 2024;DOI:10.1016/j.clbc.2024.11.011.
