Designing an intelligent clinical decision support system to diagnose preschoolers at risk for dyslexia or reading disorder

Aim: The aim of this research was to develop and implementation of an intelligent decision support system, through computerized neuro-cognitive program which was designed by Delavarian et al. in order to identify preschoolers at risk for dyslexia.
Method: This research is a type of “research and development”, in terms of its goal, and it is a kind of descriptive research, assessment and diagnostic, in terms of its data collection method. The computerized neuro-cognitive program is a program which focuses on dyslexics’ neuro-cognitive impairments due to some neural networks and circuits’ dysfunctions and alterations, and the components of information processing theory and executive functions and the efficacy of this program was proven. The program was executed with preschoolers who were selected with cluster random sampling method. Then, these data were applied in designing a decision support system.
Results: the average accuracy of the designed multilayer perceptron neural network, was reached to 94.40% and the network’s sensitivity and specificity in diagnosis of preschoolers at risk for dyslexia obtained 90.27 and 95.28%, respectively.
Conclusion: According to the high validity and reliability of the neuro-cognitive program and the high accuracy and sensitivity of the designed decision support system, the mentioned system could be used in early diagnosis of at risk preschoolers, before entering to elementary school.