BMC Medical Informatics and Decision Making

Table 7 Prediction accuracy comparison, in terms of RMSE, between the proposed model and models from related works for the OhioT1DM dataset (PH = 30 min)

From: Stacked LSTM based deep recurrent neural network with kalman smoothing for blood glucose prediction

Patient ID	RMSE w/ KS	RMSE w/o KS	RMSE of [24]	RMSE of [25]	RMSE of [56]
# 559	4.73	17.85	18.78	19.50	22.48
# 563	5.74	18.65	18.12	19.00	20.35
# 570	4.81	15.94	15.46	16.40	18.26
# 575	8.45	20.93	22.83	24.80	25.65
# 588	5.10	17.71	17.72	19.30	21.69
# 591	6.53	20.35	21.34	25.40	24.59
Mean	5.89	18.57	19.04	20.73	22.17

Here, italics values and bold values represents the best and second-best result, respectively

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com