Predicting age at onset of type 1 diabetes in children using regression, artificial neural network and Random Forest: A case study in Saudi Arabia

• DOI: 10.1371/journal.pone.0264118
• Publication Date: 2022-02-28T18:27:19Z
• Authors: Ahood Alazwari; Mali Abdollahian; Laleh Tafakori; Alice Johnstone; Rahma A. Alshumrani; Manal T. Alhelal; Abdulhameed Y. Al...; Eman S. Almoosa; Aseel R. Alkhaldi
• Keywords: Male; Adolescent; Science; Q; R; Saudi Arabia; Models, Biological; 3. Good health; 03 medical and health sciences; Diabetes Mellitus, Type 1; 0302 clinical medicine; Predictive Value of Tests; Child, Preschool; Medicine; Humans; Female; Neural Networks, Computer; Age of Onset; Child; Research Article

The rising incidence of type 1 diabetes (T1D) among children is an increasing concern globally. A reliable estimate the age at onset T1D in would facilitate intervention plans for medical practitioners to reduce problems with delayed diagnosis T1D. This paper has utilised Multiple Linear Regression (MLR), Artificial Neural Network (ANN) and Random Forest (RF) model predict Saudi Arabia (S.A.) which ranked as 7th highest number 5th world rate De-identified data between (2010-2020) from three cities S.A. were used best subset selection criteria, coefficient determination, diagnostic tests deployed select most significant variables. efficacy models predicting was assessed using multi-prediction accuracy measures. average 6.2 years common group (5-9) years. Most sample (68%) are urban areas S.A., 75% delivered after a full term pregnancy length 31% through cesarean section. fit MLR RF R2 = (0.85 0.95), root mean square error (0.25 0.15) absolute (0.19 0.11) respectively logarithm onset. study first time MLR, ANN These can effectively aid health care providers monitor create strategies impact