Data-modelling and visualisation in chronic kidney disease (CKD): a step towards personalised medicine

Predicting the function of transplanted kidney in long-term care processes: Application of a hybrid model

“…Therefore, smoothing eGFR based on each patient's own records with the MAT can help predict the trend of TK's function with less noise and fluctuation. This can be considered as a step forward towards personalized medicine [29]. Also, the AIC showed that the HB model had the best performance, thus, its selection helped minimize information loss.…”

“…Therefore, the smoothed value of the MAT could produce a better prediction for TK function. A population-based study showed that GFR is a noisy variable with a very fluctuating trend in the general population [29]. Therefore, smoothing eGFR based on each patient's own records with the MAT can help predict the trend of TK's function with less noise and fluctuation.…”

“…Using MAT in our research came from the weighted exponential moving average method, which was successfully used for time series analysis [28]. The eGFR is a 'noisy' variable with a great deal of fluctuation in a general population; thus, considering it as a solitary indicator for declining TK function at the individual patient level may not be very informative [29]. MAT presented a smoothed yet similar value to a real target eGFR and was calculated by formulas (1) to (6).…”