In this study, Pranjić et al. conducted a systematic review of objective behavioral, neurobiological, and genetic biomarkers to increase the accuracy of pediatric ADHD diagnoses. This review was performed using three major databases (MEDLINE, PsycInfo, and Scopus) and evaluated diagnostic biomarkers within 111 studies that met eligibility criteria, including 42 behavioral, 44 neuroimaging, and 25 genetic studies. These studies attempted to differentiate ADHD through behavioral studies (neuropsychological test performances, computer-based task performance tests, behavior in virtual reality, and movement measures), neurobiological studies (electroencephalograms (EEGs), magnetic resonance imaging (MRIs), and near-infrared spectroscopy (NIRS)), and genetic analyses (candidate genes, messenger RNA, microRNA, and polygenic risk). Behavioral studies assessed many features and had variable diagnostic validity. Machine learning has been used extensively to enhance neurobiological studies. Changes in theta/beta ratios (TBRs) and event-related potentials (ERPs) measured by EEGs have reported differences in ADHD; however, discriminative validity in studies remained inconsistent. Studies with ERPs had more promising results, and the use of machine learning has helped combine these measurements with behavioral scores to significantly improve accuracy, as high as 0.93 in one study. Additional neuroimaging studies using MRIs have shown that children with ADHD have decreased total brain volume, reduced iron content in the basal ganglia, and lower cerebral blood flow in the frontal lobe and caudate nuclei in comparison to controls. These studies had stronger rates of diagnostic accuracy, exceeding 0.80. Functional MRIs (fMRIs) increased accuracy with a deep learning technique, achieving 0.93, and the use of data from various distinct tasks achieved 0.92. Studies of movement measures, including eye tracking, smart chairs, and robot-led games with deep learning models, have proved promising with classification accuracies greater than 0.80. Finally, studies using NIRS consistently exceeded 0.80 for both sensitivity and specificity, and genetic biomarkers also demonstrated variable success with miRNA profiles performing best with AUCs greater than 0.90. The discussion emphasized the need for establishing clear objective diagnostic markers for ADHD within neuropsychological tests to better classify the condition. Current diagnostic markers in neuropsychological diagnosis for ADHD have inconsistent clinical utility, and more motor skill, neuroimaging, and machine learning biomarkers can improve accuracy for the diagnosis of ADHD.