No abstract

No abstract

Diabetes is a chronic disease that severely degrades the human health. Hence, the blood glucose estimation plays an important role for monitoring the diabetic condition. In order to better estimate the blood glucose values, the multi-regression models are employed. It is worth noting that increasing the total number of the regression models would decrease the regression error. Therefore, this paper proposes a method for fusing the various regression models together based on the histogram information of the blood glucose values in the training set. The computer numerical simulation results show that the regression error yielded by our proposed method is significantly lower than those yielded by the existing methods. Also, our proposed method is also applicable for other regression applications.

Diabetes is a chronic disease that severely degrades the human health. Hence, the blood glucose estimation plays an important role for monitoring the diabetic condition. In order to better estimate the blood glucose values, the multi-regression models are employed. It is worth noting that increasing the total number of the regression models would decrease the regression error. Therefore, this paper proposes a method for fusing the various regression models together based on the histogram information of the blood glucose values in the training set. The computer numerical simulation results show that the regression error yielded by our proposed method is signi cantly lower than those yielded by the existing methods. Also, our proposed method is also applicable for other regression applications.

No abstract