The Category Forecast Model of Complete Blood Count Based on the Principal Components Analysis and BP Network

摘要

In this paper we built a model using the principal components analysis (PCA) and BP neural network, to forecast the category of the complete blood count (CBC) category forecast problem. Two forecast methods were used to discuss, one was a pure BP model, another was a combined model with PCA and BP model, and the experiment results showed that, the accuracy of the former model was 64%, and the latter was 81.3%. So the combined model could improve the accuracy significantly.