To identify distinct neurobiological biotypes of ADHD using morphometric similarity networks (MSNs) and data-driven clustering, which may enhance understanding and treatment of the disorder.
Key Findings:
Normative modeling revealed significant heterogeneity in ADHD presentations, indicating the need for tailored interventions.
Data-driven clustering identified distinct neurobiological biotypes with unique clinical-biological profiles, suggesting diverse treatment pathways.
MSNs provided a robust framework for understanding brain network alterations in ADHD, potentially guiding future research and clinical practices.
Interpretation:
The study suggests that ADHD is not a monolithic disorder but consists of various neurobiological subtypes that can be identified through advanced neuroimaging techniques, which may lead to improved treatment strategies.
Limitations:
Sample heterogeneity across sites may affect generalizability and introduce biases.
Limited demographic data on race and ethnicity may influence findings and their applicability.
Conclusion:
The findings support the potential of using morphometric similarity networks to better understand ADHD subtypes, which could enhance clinical decision-making and inform future research directions.
by Nanfang Pan, Yajing Long, Kun Qin, Isaac Z. Pope, Qiuxing Chen, Ziyu Zhu, Ying Cao, Lei Li, Manpreet K. Singh, Robert K. McNamara, Melissa P. DelBello, Ying Chen, Alex Fornito, Qiyong Gong
A VHA study across 11 vendors finds AI-generated primary care notes score lower than clinician-written notes, with the largest deficits in thoroughness, organization, and usefulness
