Skip to main content
Fig. 2 | Diagnostic and Prognostic Research

Fig. 2

From: Risk of bias of prognostic models developed using machine learning: a systematic review in oncology

Fig. 2

Bar charts showing the risk of bias ratings for each domain and the overall judgement, for the development of 152 models (left) and external validation of 37 developed models (right). “Overall” indicates the overall risk of bias; “participants” indicates bias introduced by participants or data sources; “predictors” indicates bias introduced by predictors or their assessment; “outcome” indicates bias introduced by outcomes or their assessment; “analysis” indicates bias introduced by the analysis. Values in the bars are frequency (%). * values for risk of bias (development models - predictors) are 1(1). ** values for risk of bias (validation models - overall) are 1(3)

Back to article page