The Red Blood Cells dataset

Published: 10-03-2021| Version 2 | DOI: 10.17632/rg7f6bwvb7.2
Nour Eddine Alaa,


The Red Blood Cells (RBC) data set is a multivariate data set inspired by prominent E. Evans article "Improved Measurements of the Erythrocyte Geometry" Microvascular Research, 4(4):335-347, 1972. The data set consists of fifty cells for each tonicity of three species of RBC. Attribute Information: The RBC data set has six attributes. You will be using the first five attributes to predict the species, whose class is defined in the fifth attribute of the data set. Input Variables : • Diameter in μm • Minimum thickness in μm • Maximum thickness in μm • Surface area in μm^2 • Volume in μm^3 Output Variables: Class (RBC Species) • RBC at 300 mOsm • RBC at 217 mOsm • RBC at 131 mOsm An example of the different species is given in the attached figure. The complete program for generating the RBC database in Matlab can be found in the RBCdataset.m file in the attached file. The bases generated by this program for the case N = 50 and N = 500 where N is the number of each species in the base are given here in the excel tables below.