Background: Diagnosing Attention Deficit Hyperactivity Disorder (ADHD) is a challenging process currently focused on questionnaires and clinical interviews, which could be sensitive to subjective biases (Farone et al. 2005). Some studies tried machine learning (ML) algorithms with neurophysiological data such as functional magnetic resonance imaging or electroencephalography to explore objective measures for diagnosing purporses (Mikolas et al. 2022). In this study, ML algorithms are built using cheaper and simpler measures including cognitive and intelligence assessments, reducing time and resources for ADHD diagnosis. Methods: Data from the Child Mind Institute dataset (Alexander et.al, 2017) was used, analyzing 145 children (77 ADHD, 68 controls) aged 8-11 years. The included variables were cognitive and intelligence assessments (NIH, WIAT, WISC), resulting in a total of 29 features. For ML models, the data was divided into training (80%) and test (20%) sets. A LASSO algorithm was used for feature selection with 10000 iterations and five-fold cross validation. Afterwards, several supervised ML classifiers were trained with a five-fold cross validation and validated on the test set over 500 iterations. Results: The best performance was reached by gaussian Support Vector Machine (SVM), with 73.3% accuracy using 5 features selected by LASSO. Given the similarity of the training and test performance values, the results reveal that overfitting was successfully avoided, indicating the model is generalizable among subjects. Conclusion: The time and resources needed for ADHD diagnosis may be reduced by using cognitive and intelligence assessments combined with ML techniques, reaching a moderate accuracy. References: Alexander, L. et al. An open resource for transdiagnostic research in pediatric mental health and learning disorders. Scientific Data 4, 170181 (2017).Faraone, S. V. The scientific foundation for understanding attention-deficit/hyperactivity disorder as a valid psychiatric disorder. Eur. Child Adolesc. Psychiatry 14, 1–10 (2005).Mikolas, P., Vahid, A., Bernardoni, F., Süß, M., Martini, J., Beste, C., & Bluschke, A. (2022). Training a machine learning classifier to identify ADHD based on real-world clinical data from medical records. Scientific Reports, 12(1), 12934.