Pulse Diagnosis Signals Analysis of Fatty Liver Disease and Cirrhosis Patients by Using Machine Learning

Figure 16

Classification by PCA. The accuracy that we cannot classify the signals of the 2 groups by the 1st principal component is 0.22, which means that the accuracy to classify this 2 groups is 78%, and the blue points (patients with cirrhosis) mainly appeared in the left side of the red point and the green ones (patients with FLD) appeared in the right side. And the accuracy by the 4th principal component and the 5th principal component is 87%.