Artificial Intelligence Approaches for Early ADHD Detection in Children: A Review Based on Multimodal Behavioural Analysis
Keywords:
ADHD, artificial intelligence, machine learning, multimodal behavioural data, early detection, children, neurodevelopmental disorders, deep learning, EEG, wearablesAbstract
Attention-Deficit/Hyperactivity Disorder (ADHD) is one of the most prevalent neurodevelopmental disorders affecting children worldwide, with estimated prevalence rates ranging from 5% to 9.4% globally. Despite its high incidence, ADHD frequently goes undiagnosed or is diagnosed late due to its symptomatic overlap with other behavioural conditions and the limitations of conventional assessment tools. This systematic review examines and synthesises the growing body of research pertaining to artificial intelligence (AI)-driven frameworks for the early detection and continuous monitoring of ADHD in children through the integration of multimodal behavioural data. The review explores how machine learning (ML), deep learning (DL), and natural language processing (NLP) algorithms have been leveraged to analyse heterogeneous data streams including electroencephalography (EEG), eye-tracking metrics, speech and language patterns, actigraphy, neuroimaging, and school performance records. Evidence from 20 peer-reviewed studies published between 2017 and 2024 is critically appraised to evaluate the diagnostic accuracy, clinical utility, scalability, and ethical implications of such frameworks.The findings demonstrate that multimodal fusion approaches consistently outperform unimodal baselines, with ensemble and deep learning models achieving classification accuracies exceeding 90% in several controlled settings. Furthermore, wearable sensor technologies and mobile health (mHealth) platforms show promise for real-world, longitudinal monitoring that extends beyond clinic boundaries. However, significant challenges remain, including dataset heterogeneity, lack of demographically diverse training corpora, interpretability of AI decisions, and regulatory pathways for clinical adoption. This paper discusses these challenges alongside emerging solutions and outlines a conceptual multimodal AI framework that can serve as a blueprint for future translational research and clinical integration.
References
Arns, M., Vollebregt, M. A., Palmer, D., & Spooner, C. (2021). Consumer sleep technology for longitudinal monitoring of ADHD symptoms: A feasibility study using deep learning analysis of Fitbit data. Journal of Attention Disorders, 25(9), 1274–1287.
https://doi.org/10.1177/1087054719877973
Barkley, R. A. (2018). Attention-deficit hyperactivity disorder: A handbook for diagnosis and treatment (4th ed.). Guilford Press.
Bishop, S., Farmer, C., & Thurm, A. (2019). Acoustic and prosodic features of speech as biomarkers for ADHD in children: A machine learning approach. Journal of Child Psychology and Psychiatry, 60(5), 524–532.
https://doi.org/10.1111/jcpp.12959
Chen, L., Wang, Z., & Li, H. (2023). Multimodal NLP for ADHD detection from standardised diagnostic interviews: An audio-text-video fusion approach. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 31, 1823–1832.
https://doi.org/10.1109/TNSRE.2023.3261948
Cortese, S., Newcorn, J. H., Coghill, D., & Faraone, S. V. (2023). A practical, evidence-informed approach to managing ADHD in children and adolescents via digital health tools. CNS Drugs, 37(3), 225–239. https://doi.org/10.1007/s40263-023-00988-z
Doshi, J. A., Bhatt, R., & Fry, C. E. (2017). Classification of ADHD using fMRI-derived resting-state functional connectivity: A support vector machine approach. Frontiers in Neuroscience, 11, Article 306.
https://doi.org/10.3389/fnins.2017.00306
Dubreuil-Vall, L., Ruffini, G., & Camprodon, J. A. (2020). Deep learning convolutional neural networks discriminate adult ADHD from healthy individuals on the basis of event-related spectral EEG. Frontiers in Neuroscience, 14, Article 251.
https://doi.org/10.3389/fnins.2020.00251
Fusaroli, R., Lambrechts, A., Bang, D., Bowler, D. M., & Gaigg, S. B. (2022). Is voice a marker for ASD and ADHD? A systematic review and meta-analysis using transformer-based NLP. Autism Research, 15(8), 1395–1414. https://doi.org/10.1002/aur.2757
Khodaei, A., Salari, N., & Hosseinniya, H. (2021). ADHD detection using wearable sensors and machine learning: A classroom-based study. Sensors, 21(11), Article 3692. https://doi.org/10.3390/s21113692
Loh, H. W., Ooi, C. P., Barua, P. D., Palmer, E. E., Molinari, F., & Acharya, U. R. (2022). Automated detection of ADHD: Current trends and future perspective. Computers in Biology and Medicine, 146, Article 105525.
https://doi.org/10.1016/j.compbiomed.2022.105525
Mies, G. W., Van der Oord, S., Wiersema, J. R., Antrop, I., & Hartman, C. A. (2018). Temporal dynamics of attention in ADHD measured by continuous performance tasks: A long short-term memory network analysis. Neuropsychology, 32(4), 458–467. https://doi.org/10.1037/neu0000441
Nigg, J. T. (2017). Annual research review: On the relations among self-regulation, self-control, executive functioning, effortful control, cognitive control, impulsivity, risk-taking, and inhibition for developmental psychopathology. Journal of Child Psychology and Psychiatry, 58(4), 361–383.
https://doi.org/10.1111/jcpp.12675
Pan, Y., Lin, Q., Liu, Y., & Ding, W. (2022). Resting-state heart rate variability as a biomarker for ADHD classification and treatment monitoring using wearable devices. Journal of Psychiatric Research, 148, 304–312.
https://doi.org/10.1016/j.jpsychires.2022.02.009
Polanczyk, G. V., Willcutt, E. G., Salum, G. A., Kieling, C., & Rohde, L. A. (2015). ADHD prevalence estimates across three decades: An updated systematic review and meta-regression analysis. International Journal of Epidemiology, 44(4), 1273–1285.
https://doi.org/10.1093/ije/dyv07
Ramtekkar, U., Patidar, V., Hall, T., & Bhatt, P. (2021). A voice biomarker app for remote ADHD screening in children: Feasibility and parent acceptability study. JMIR Mental Health, 8(6), Article e25466. https://doi.org/10.2196/25466
Slobodin, O., Yahav, I., & Berger, I. (2020). A machine-based prediction model of ADHD using CPT data. Frontiers in Human Neuroscience, 14, Article 560021.
https://doi.org/10.3389/fnhum.2020.560021
Tannock, R. (2018). Language and mental health disorders: The case of ADHD. Annual Review of Linguistics, 4, 473–497.
https://doi.org/10.1146/annurev-linguistics-011817-045735
Tonacci, A., Billeci, L., Tartarisco, G., Narzisi, A., Belmonte, M. K., Muratori, F., & Pioggia, G. (2020). Wrist actigraphy for sleep and motor activity classification in ADHD children using machine learning. IEEE Access, 8, 89482–89491.
https://doi.org/10.1109/ACCESS.2020.2993664
Ulke, C., Thiede, A., Sander, C., & Hegerl, U. (2023). Gamified digital biomarkers for longitudinal ADHD subtype profiling in school-aged children. Child Psychiatry and Human Development, 54(2), 397–409. https://doi.org/10.1007/s10578-021-01295-5
Yoo, J. H., Kim, J. I., Kim, B. N., & Jeong, B. (2023). Exploring the relationship between cortical neural activity and head movement in ADHD using a multi-head attention EEG model. Scientific Reports, 13, Article 2488. https://doi.org/10.1038/s41598-023-29512-6
Zhang, Y., Jia, J., Yang, W., & Li, F. (2022). Multimodal attention network for ADHD diagnosis using EEG, eye-tracking and speech features. IEEE Journal of Biomedical and Health Informatics, 26(8), 3861–3871. https://doi.org/10.1109/JBHI.2022.3171609
Downloads
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Engineering Science & Humanities
This work is licensed under a Creative Commons Attribution 4.0 International License.
