Output details

An integrated framework for risk profiling of breast cancer patients following surgery.

<24> Machine learning approaches. The significance of this paper is to define an effective risk profiling interface for use by clinical experts. The originality is to derive a rigorous calculation of confidence intervals for a regularised machine learning model of survival and also to promote interpretability by contrasting novel prognostic modelling with standard multivariate clinical risk scores in a systematic manner. The paper builds on a series of papers proposing the single-risk model (Lisboa et al. AIIM, 2003) and validating it in a multicentre study (Taktak, Lisboa et al. CBM, 2007).