Neural Networks for Survival Prediction in Medicine Using Prognostic Factors: A Review and Critical Appraisal

Figure 1

Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flowchart. Reasons for exclusion of the 214 studies in screening step 1 were: ML techniques for classification (), predictions based on individual’s images (), models with focus on feature selection (), bioinformatics/computational biology analysis only (), other ML techniques for survival analysis (), unsupervised learning (), other reasons () including ML techniques for risk group stratification (), systematic/literature review (), new prediction tool (), ML techniques for regression (), ensemble of different ML techniques (), no prediction (), letter to the editor (), model for nonhumans (), models with focus on feature reduction (), and tutorial/case study ().