We have developed microfluidic devices for precise characterization of hematological diseases. By means of one drop of blood and combining front microfluidics with mathematical models based on biomechanics, we analyze the properties of red blood cells and microrheological properties such as the viscosity of blood. Machine learning and statistical analysis is used to improve the diagnosis of the hematological diseases. We have used different machine learning methods as Logistic Regressions or Artificial Neural Networks (ANN) to predict if a sample of blood corresponds to healthy blood or to a blood with an hematological disease. We have obtained different performance for the different methods, some of them with very good results and an accuracy of 94% of correct prediction has been achieved with an ANN model.