) and contained measurements from biopsies of 30 DLBCL patients. Each sample was stained with three antibodies,: CD3, CD5, and CD19.
The LunG data [ 44 ] consists of d = 1,000 dimensions with n = 197 lung cancer patients from k = 4 different factors.
The Prostate data [ 45 ] consists of d = 12,600 dimensions with n = 102 prostatecancer patients from k = 2 different factors. The original data set from contained 97 men who haved prostatecancer and recorded the information of the patients.
Table 3 The date information in the algorithm test
Víctor M. Pérez-García, Susan Fitzpatrick, Luis A. Pérez-Romasanta, Milica Pesic, Philippe Schucht, Estanislao Arana and Pilar Sánchez-Gómez
.g. [ 55 , 56 ] for brain tumors). From the mathematical point of view it is necessary to define meaningful and robust measures of volumes, geometries, textures, etc., to be computed on images providing biomarkers of clinical relevance. Also, images are to be used as initial data for predictive mathematical algorithms able to simulate tumor evolution and provide therapy personalization.
Another source of information are blood tests, which allow clinically relevant biomarkers (such as e.g. PSA levels in prostatecancer) to be directly measured. In recent years the