This paper presents the design, the development and the evaluation of a personalized glucose prediction model for patients with Type 1 Diabetes Mellitus (T1DM). The personalized model is based on neuro-fuzzy techniques in order to capture the metabolic behavior of a patient with T1DM. Moreover, wavelets are applied as activation functions in order to enhance the prediction performance and avoid local minimum during training stage. The model receives as input, data from sensors which record in real time glucose levels and physical activity, and provides with future glucose levels. The proposed model is evaluated using data from the medical records of 6 patients with T1DM for the time being on CGMSs and physical activity sensors. The obtained results demonstrate the ability of the proposed model to capture the metabolic behavior of a patient with T1DM and to handle intra- and inter-patient variability.