Sex differences in the functional connectomes of individuals with ADHD
Poster No:
465
Submission Type:
Abstract Submission
Authors:
Natalia Avendano Prieto1, Keith Jamison2, Shahrooz Faghihroohi3, Amy Kuceyeski4
Institutions:
1Weill Cornell Medicine, Ithaca, NY, 2Weill Cornell Medicine, New York, NY, 3Technische Universität München, Munich, Bayern, 4Cornell, Ithaca, NY
First Author:
Co-Author(s):
Introduction:
Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental and heterogeneous condition influenced by sex, genetic, and environmental factors (Cao, Martin, & Li, 2023). It is characterized by inattention, hyperactivity, and impulsivity, causing challenges in learning and daily life (Zhang, Murray, Duval, Wang, & Jann, 2024). ADHD primarily affects children but often persists into adulthood and is diagnosed more frequently in males than in females (sex ratios: 2:1 to 10:1) (Mowlem, Rosenqvist, Martin, Lichtenstein, Asherson, & Larsson, 2019).
Despite the high prevalence of ADHD, its psychopathology remains unclear, hindering early detection and effective treatments (Zhang, Murray, Duval, Wang, & Jann, 2024). Treatment is often trial-and-error, with about 30% of patients responding poorly to common drugs like Methylphenidate (Cao, Martin, & Li, 2023), which translates to a high risk of adverse drug-related outcomes.
Understanding brain changes in ADHD through functional MRI can improve diagnosis and treatment. Specifically, identifying functional connectivity (FC) differences between ADHD and healthy controls (HC), and focusing on sex differences in the disorder, can provide insights into its biological mechanisms and guide the development of personalized treatments.
Despite the high prevalence of ADHD, its psychopathology remains unclear, hindering early detection and effective treatments (Zhang, Murray, Duval, Wang, & Jann, 2024). Treatment is often trial-and-error, with about 30% of patients responding poorly to common drugs like Methylphenidate (Cao, Martin, & Li, 2023), which translates to a high risk of adverse drug-related outcomes.
Understanding brain changes in ADHD through functional MRI can improve diagnosis and treatment. Specifically, identifying functional connectivity (FC) differences between ADHD and healthy controls (HC), and focusing on sex differences in the disorder, can provide insights into its biological mechanisms and guide the development of personalized treatments.
Methods:
We used the Healthy Brain Network (HBN) dataset which includes FC, diagnosis information, and demographics for each subject (Healthy Brain Network, n.d.). The dataset comprises 1359 subjects aged 5-21 years, 885 males and 474 females, where 939 have an ADHD diagnosis, 254 have another neurodevelopmental disorder (NoADHD), and 166 have no diagnosis (healthy controls – HC).
Logistic regression was implemented via cross validation with train and test splits stratified by sex and diagnosis to ensure that the percentage of each group was consistent. Separate models were trained on slightly different diagnosis splits: ADHD vs HC, and ADHD vs [HC+NoADHD]. Finally, we trained classifiers using 1) both male and female participants, 2) females only and 3) males only. This was done to investigate if there are sex differences in the FC of individuals with ADHD. Feature importances were compared across all three models and male models were applied to female data (and vice versa). Balanced accuracy, and F1 score for each class were used to evaluate model performance. To obtain uncertainty of the performance metrics, each model was run 100 times, and the average of each metric were calculated across different train-test splits.
Logistic regression was implemented via cross validation with train and test splits stratified by sex and diagnosis to ensure that the percentage of each group was consistent. Separate models were trained on slightly different diagnosis splits: ADHD vs HC, and ADHD vs [HC+NoADHD]. Finally, we trained classifiers using 1) both male and female participants, 2) females only and 3) males only. This was done to investigate if there are sex differences in the FC of individuals with ADHD. Feature importances were compared across all three models and male models were applied to female data (and vice versa). Balanced accuracy, and F1 score for each class were used to evaluate model performance. To obtain uncertainty of the performance metrics, each model was run 100 times, and the average of each metric were calculated across different train-test splits.
Results:
The logistic regression model trained on male and female data had slightly better balanced accuracy on the ADHD vs HC task compared to the ADHD vs [NoADHD+HC] task (Fig 1). The model trained on males only had higher accuracy when applied to both males and females compared to the model trained only on females. The female model had about the same accuracy when classifying males and females, while the male model had slightly higher accuracy for test females than test males. Model coefficients for each edge in the FC were averaged over regions and visualized on the brain (Fig 2); coefficients for the three models were correlated. There is some overlap of the female and male only model coefficients compared to the model coefficients for the combined Male+Female model, but little overlap exists for the male only and female only models. The correlation with the combined model is not surprising as they share some of the same data. In contrast, female model and male model have almost no correlation, suggesting they assign feature importance differently.
·Test balanced accuracy for the different developed models
·Brain plot and correlation of the model’s beta weights for ADHD vs HC data
Conclusions:
Logistic regression applied to FCs yields moderately good balanced accuracies in predicting ADHD diagnosis, and models trained on males and females only appear to have good generalizability to the other sex despite the feature importances being uncorrelated. This lack of correlation of feature importances may suggest distinct FC features are important for predicting ADHD class in males and females.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Lifespan Development:
Early life, Adolescence, Aging
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 2
Neuroinformatics and Data Sharing:
Informatics Other
Keywords:
Attention Deficit Disorder
Computing
Data analysis
FUNCTIONAL MRI
Machine Learning
Modeling
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Which processing packages did you use for your study?
Provide references using APA citation style.
Zhang, R., Murray, S. B., Duval, C. J., Wang, D. J. J., & Jann, K. (2024). Functional connectivity and complexity analyses of resting-state fMRI in pre-adolescents demonstrating the behavioral symptoms of ADHD. Psychiatry Research, 334, 115794. https://doi.org/10.1016/j.psychres.2024.115794
Mowlem, F. D., Rosenqvist, M. A., Martin, J., Lichtenstein, P., Asherson, P., & Larsson, H. (2019). Sex differences in predicting ADHD clinical diagnosis and pharmacological treatment. European Child & Adolescent Psychiatry, 28(4), 481-489. https://doi.org/10.1007/s00787-018-1211-3
Healthy Brain Network. (n.d.). Healthy Brain Network. Retrieved December 17, 2024, from https://data.healthybrainnetwork.org/main.php