Breast cancer: Intravoxel incoherent movement-based habitat imaging for predicting pathologic complete response to neoadjuvant chemotherapy

Motivation: Radiomics research based on whole tumors is limited by the unclear biological significance of radiomics features, which therefore lack clinical interpretability. Goal(s): We aimed to determine whether features extracted from subregions defined by habitat imaging could identify breast cancer patients who will benefit from neoadjuvant chemotherapy (NAC). Approach: We builded models for predicting pCR based on whole-tumor radiomics (ModelWH), intravoxel incoherent motion-based habitat imaging (ModelHabitats), conventional MRI features (ModelCF), and immunohistochemical findings (ModelIHC). Results: In the prediction of pCR, ModelWH, ModelHabitats, ModelCF, ModelIHC, ModelHabitats+CF, and ModelHabitats+CF+IHC achieved AUCs of 0.549, 0.708, 0.700, 0.788, 0.745, and 0.891 respectively, in the test set. Impact: The habitat model combined with conventional MRI features and IHC findings accurately predicted pCR before NAC.