Novel method for risk stratification of radiation-induced breast fibrosis: subgroup hypothesis verified by machine learning.

By aggregating the two subgroups with low risk and three subgroups with medium risk, patients could be assigned to three risk groups (high, medium, low) with a highly significant ( p < 0.0001) five-fold difference in risk (Table 5 and insert in Fig. 2 ).

also in 132,142 other papers

Because BMI itself showed a trend (p = 0.11) for an association with fibrosis even in the high-RILA subgroup (Supplementary Fig.

also in 53,322 other papers