A multicenter study on preoperative prediction of TNBC based on multi-parameter MRI intratumoral combined with peritumoral radiomics

Motivation: Previous studies on triple-negative breast cancer(TNBC) have primarily focused on the characteristics of the tumor itself, with less emphasis on the surrounding tissue, and have mostly been conducted at a single center using a single sequence of analysis. Goal(s): Multi-sequence MRI intratumoral combined with peritumoral radiomics models can identify TNBC , providing individualized treatment plans for patients and improving their prognosis. Approach: Using image genomics methods, eight predictive models for TNAC were constructed, validated internally and externally, and the optimal model was finally selected. Results: The model constructed based on multiparametric MRI intratumor combined with peritumor radiomics features can better predict TNBC and non-TNBC. Impact: This model can predict TNBC in a non-invasive and early manner, which is of great significance to the treatment and prognosis of patients.