The pathology of melanoma as a basis for prognostic models: the UCSF experience

A patient survival model is proposed which allows visualization of a data base, and which includes only routine and commonly recorded attributes in most melanoma clinics. It is proposed that a network of such data be collected for meta-analysis (MELNET), which could make stratification within the individual subsets more significant by virtue of the large numbers. Such a network could then be fully tested in various melanoma clinics for clinical usefulness.