Neural networks as statistical methods in survival analysis

摘要: Introduction Artificial neural networks are increasingly being seen as an addition to the statistics toolkit that should be considered alongside both classical and modern statistical methods. Reviews in this light have been given by one of us (Ripley 1993, 1994a–c, 1996) Cheng & Titterington (1994) it is a point view widely accepted mainstream community. There now many texts (Hertz et al. 1991; Haykin 1994; Bishop 1995; Ripley covering wide range artificial networks; we concentrate here on methods see most appropriate generally medicine, particular for survival data not our knowledge reviewed depth (although Schwarzer (1997) large number applications oncology). In particular, out different ways classification used data, well their flaws. Most medicine problems; is, task basis measured features assign patient (or biopsy or electroencephalograph …) small set classes. Baxt (1995) gave table clinical almost all form, including those laboratories (Dybowski Gant 1995).