JDS Journal of Data Science 1683-86021680-743X1680-743X School of Statistics, Renmin University of China JDS1128 10.6339/24-JDS1128 Data Science in Action A Multi-Model Framework to Explore ADHD Diagnosis from Neuroimaging Data Yavuz OzdemirYagmuryzy0096@auburn.edu1∗ NukalaNaga Chandra Padmini1 MolinariRoberto1 DeshpandeGopikrishna2345678910 Department of Mathematics and Statistics, Auburn University, Auburn, Alabama, United States AU MRI Research Center, Department of Electrical and Computer Engineering, Auburn University, Auburn, Alabama, United States Department of Psychological Sciences, Auburn University, Auburn, Alabama, United States Alabama Advanced Imaging Consortium, Auburn, Alabama, United States Center for Neuroscience, Auburn University, Auburn, Alabama, United States School of Psychology, Capital Normal University, Beijing, China Key Laboratory for Learning and Cognition, Capital Normal University, Beijing, China Department of Psychiatry, National Institute of Mental Health and Neurosciences, Karnataka, India Centre for Brain Research, Indian Institute of Science, Bangalore, India Department of Heritage Science and Technology, Indian Institute of Technology, Hyderabad, India Corresponding author. Email: yzy0096@auburn.edu. 2024252024222191207 Supplementary Material

All of our code is open source in the following GitHub repository https://github.com/yagmuryavuzozdemir/SDSS_SWAG_ADHD. One can find the necessary codes and the datasets used in the analysis of our work in this folder.

3172023342024 2024 The Author(s). Published by the School of Statistics and the Center for Applied Statistics, Renmin University of China.2024 Open access article under the CC BY license.

Attention Deficit Hyperactivity Disorder (ADHD) is a frequent neurodevelopmental disorder in children that is commonly diagnosed subjectively. The objective detection of ADHD based on neuroimaging data has been a complex problem with low ranges of accuracy, possibly due to (among others) complex diagnostic processes, the high number of features considered and imperfect measurements in data collection. Hence, reliable neuroimaging biomarkers for detecting ADHD have been elusive. To address this problem we consider a recently proposed multi-model selection method called Sparse Wrapper AlGorithm (SWAG), which is a greedy algorithm that combines screening and wrapper approaches to create a set of low-dimensional models with good predictive power. While preserving the previous levels of accuracy, SWAG provides a measure of importance of brain regions for identifying ADHD. Our approach also provides a set of equally-performing and simple models which highlight the main feature combinations to be analyzed and the interactions between them. Taking advantage of the network of models resulting from this approach, we confirm the relevance of the frontal and temporal lobes as well as highlight how the different regions interact to detect the presence of ADHD. In particular, these results are fairly consistent across different learning mechanisms employed within the SWAG (i.e. logistic regression, linear and radial-kernel support vector machines) thereby providing population-level insights, as well as delivering feature combinations that are smaller and often perform better than those that would be used if employing their original versions directly.

 automated detection functional magnetic resonance imaging interpretability prediction accuracy SWAG References ADHD-200 C (2012). The adhd-200 consortium: A model to advance the translational potential of neuroimaging in clinical neuroscience. Frontiers in Systems Neuroscience, 6: 62. Arbabshirani MR, Plis S, Sui J, Calhoun VD (2017). Single subject prediction of brain disorders in neuroimaging: Promises and pitfalls. NeuroImage, 145: 137–165. https://doi.org/10.1016/j.neuroimage.2016.02.079 Bellec P, Chu C, Chouinard-Decorte F, Benhajali Y, Margulies DS, Craddock RC (2017). The neuro bureau adhd-200 preprocessed repository. NeuroImage, 144: 275–286. https://doi.org/10.1016/j.neuroimage.2016.06.034 Branca M, Orso S, Molinari RC, Xu H, Guerrier S, Zhang Y, et al. (2018). Is nonmetastatic cutaneous melanoma predictable through genomic biomarkers? Melanoma Research, 28(1): 21–29. https://doi.org/10.1097/CMR.0000000000000412 Breiman L (2001). Random forests. Machine Learning, 45: 5–32. https://doi.org/10.1023/A:1010933404324MR3874153 Carmona S, Vilarroya O, Bielsa A, Tremols V, Soliva J, Rovira M, et al. (2005). Global and regional gray matter reductions in adhd: A voxel-based morphometric study. Neuroscience Letters, 389(2): 88–93. https://doi.org/10.1016/j.neulet.2005.07.020 Castellanos FX, Aoki Y (2016). Intrinsic functional connectivity in attention-deficit/hyperactivity disorder: A science in development. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 1(3): 253–261. Chandrashekar G, Sahin F (2014). A survey on feature selection methods. Computers & Electrical Engineering, 40(1): 16–28. https://doi.org/10.1016/j.compeleceng.2013.11.024 Craddock RC, James GA, Holtzheimer III PE, Hu XP, Mayberg HS (2012). A whole brain fmri atlas generated via spatially constrained spectral clustering. Human Brain Mapping, 33(8): 1914–1928. https://doi.org/10.1002/hbm.21333 Curatolo P, D’Agati E, Moavero R (2010). The neurobiological basis of adhd. Italian Journal of Pediatrics, 36(1): 1–7. https://doi.org/10.1186/1824-7288-36-79 Deshpande G, Wang P, Rangaprakash D, Wilamowski B (2015). Fully connected cascade artificial neural network architecture for attention deficit hyperactivity disorder classification from functional magnetic resonance imaging data. IEEE Transactions on Cybernetics, 45(12): 2668–2679. https://doi.org/10.1109/TCYB.2014.2379621 Draghici S, Khatri P, Eklund AC, Szallasi Z (2006). Reliability and reproducibility issues in dna microarray measurements. Trends in Genetics, 22(2): 101–109. https://doi.org/10.1016/j.tig.2005.12.005 Eloyan A, Muschelli J, Nebel MB, Liu H, Han F, Zhao T, et al. (2012). Automated diagnoses of attention deficit hyperactive disorder using magnetic resonance imaging. Frontiers in Systems Neuroscience, 6: 61. Fisher A, Rudin C, Dominici F (2019). All models are wrong, but many are useful: Learning a variable’s importance by studying an entire class of prediction models simultaneously. Journal of Machine Learning Research, 20(177): 1–81.MR4048988 Friedman J, Hastie T, Tibshirani R (2010). Regularization paths for generalized linear models via coordinate descent. Journal of Statistical Software, 33(1): 1. https://doi.org/10.18637/jss.v033.i01 Guerrier S, Mili N, Molinari R, Orso S, Avella-Medina M, Ma Y (2016). A predictive based regression algorithm for gene network selection. Frontiers in Genetics, 7: 97. Huf W, Kalcher K, Boubela RN, Rath G, Vecsei A, Filzmoser P, et al. (2014). On the generalizability of resting-state fmri machine learning classifiers. Frontiers in Human Neuroscience, 8: 502. Kelly C, Biswal BB, Craddock RC, Castellanos FX, Milham MP (2012). Characterizing variation in the functional connectome: Promise and pitfalls. Trends in Cognitive Sciences, 16(3): 181–188. https://doi.org/10.1016/j.tics.2012.02.001 Kobel M, Bechtel N, Specht K, Klarhöfer M, Weber P, Scheffler K, et al. (2010). Structural and functional imaging approaches in attention deficit/hyperactivity disorder: Does the temporal lobe play a key role? Psychiatry Research: Neuroimaging, 183(3): 230–236. https://doi.org/10.1016/j.pscychresns.2010.03.010 Lanka P, Rangaprakash D, Dretsch MN, Katz JS, Denney TS, Deshpande G (2020). Supervised machine learning for diagnostic classification from large-scale neuroimaging datasets. Brain Imaging and Behavior, 14: 2378–2416. https://doi.org/10.1007/s11682-019-00191-8 Lin H, Haider SP, Kaltenhauser S, Mozayan A, Malhotra A, Constable RT, et al. (2023). Population level multimodal neuroimaging correlates of attention-deficit hyperactivity disorder among children. Frontiers in Neuroscience, 17: 1138670. https://doi.org/10.3389/fnins.2023.1138670MR4644006 Loh HW, Ooi CP, Barua PD, Palmer EE, Molinari F, Acharya UR (2022). Automated detection of adhd: Current trends and future perspective. Computers in Biology and Medicine, 146: 105525. https://doi.org/10.1016/j.compbiomed.2022.105525 Meinshausen N, Yu B (2009). Lasso-type recovery of sparse representations for high-dimensional data.MR2488351 Miglioli C, Bakalli G, Orso S, Karemera M, Molinari R, Guerrier S, et al. (2022). Evidence of antagonistic predictive effects of mirnas in breast cancer cohorts through data-driven networks. Scientific Reports, 12(1): 5166. https://doi.org/10.1038/s41598-022-08737-5 Mili N, Molinari R, Ma Y, Guerrier S (2016). Differentiating inflammatory bowel diseases by using genomic data: dimension of the problem and network organization. In: HUMAN GENOMICS, volume 10. BIOMED CENTRAL LTD 236 GRAYS INN RD, FLOOR 6, LONDON WC1X 8HL, ENGLAND. Molinari R, Bakalli G, Guerrier S, Miglioli C, Orso S, Karemera M, et al. (2020). Swag: A wrapper method for sparse learning. arXiv preprint: https://arxiv.org/abs/2006.12837. Nielsen JA, Zielinski BA, Fletcher PT, Alexander AL, Lange N, Bigler ED, et al. (2013). Multisite functional connectivity mri classification of autism: Abide results. Frontiers in Human Neuroscience, 7: 599. Olivetti E, Greiner S, Avesani P (2012). Adhd diagnosis from multiple data sources with batch effects. Frontiers in Systems Neuroscience, 6: 70. https://doi.org/10.3389/fnsys.2012.00070 Parisi N, Janier-Dubry A, Ponzetto E, Pavlopoulos C, Bakalli G, Molinari R, et al. (2020). Non applicability of validated predictive models for intensive care admission and death of covid-19 patients in a secondary care hospital in belgium. medRxiv, 2020–11. Rubia K, Criaud M, Wulff M, Alegria A, Brinson H, Barker G, et al. (2019). Functional connectivity changes associated with fmri neurofeedback of right inferior frontal cortex in adolescents with adhd. NeuroImage, 188: 43–58. https://doi.org/10.1016/j.neuroimage.2018.11.055 Rudin C (2019). Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence, 1(5): 206–215. https://doi.org/10.1038/s42256-019-0048-x Sayal K, Prasad V, Daley D, Ford T, Coghill D (2018). Adhd in children and young people: Prevalence, care pathways, and service provision. The Lancet Psychiatry, 5(2): 175–186. https://doi.org/10.1016/S2215-0366(17)30167-0 Schnack HG, Kahn RS (2016). Detecting neuroimaging biomarkers for psychiatric disorders: Sample size matters. Frontiers in Psychiatry, 7: 50. Semenova L, Rudin C, Parr R (2022). On the existence of simpler machine learning models. In: Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency, Association for Computing Machinery, 1827–1858. Soman SM, Vijayakumar N, Thomson P, Ball G, Hyde C, Silk TJ (2023). Functional and structural brain network development in children with attention deficit hyperactivity disorder. Human Brain Mapping, 44(8): 3394–3409. Sowell ER, Thompson PM, Welcome SE, Henkenius AL, Toga AW, Peterson BS (2003). Cortical abnormalities in children and adolescents with attention-deficit hyperactivity disorder. The Lancet, 362(9397): 1699–1707. https://doi.org/10.1016/S0140-6736(03)14842-8 van der Ploeg T, Austin PC, Steyerberg EW (2014). Modern modelling techniques are data hungry: A simulation study for predicting dichotomous endpoints. BMC Medical Research Methodology, 14(1): 1–13. https://doi.org/10.1186/1471-2288-14-1 Vats D, Baraniuk R (2013). When in doubt, swap: High-dimensional sparse recovery from correlated measurements. Advances in Neural Information Processing Systems, 26. Vaughan L, Chen Y (2015). Data mining from web search queries: A comparison of Google trends and baidu index. The Journal of the Association for Information Science and Technology, 66(1): 13–22. https://doi.org/10.1002/asi.23201 Visser SN, Danielson ML, Bitsko RH, Holbrook JR, Kogan MD, Ghandour RM, et al. (2014). Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003–2011. Journal of the American Academy of Child and Adolescent Psychiatry, 53(1): 34–46. https://doi.org/10.1016/j.jaac.2013.09.001 Wang G, Li W, Zuluaga MA, Pratt R, Patel PA, Aertsen M, et al. (2018). Interactive medical image segmentation using deep learning with image-specific fine tuning. IEEE Transactions on Medical Imaging, 37(7): 1562–1573. https://doi.org/10.1109/TMI.2018.2791721 Wang T (2019). Gaining free or low-cost interpretability with interpretable partial substitute. In: International Conference on Machine Learning, Journal of Machine Learning Research, 6505–6514. PMLR. Wu GR, Liao W, Stramaglia S, Ding JR, Chen H, Marinazzo D (2013). A blind deconvolution approach to recover effective connectivity brain networks from resting state fmri data. Medical Image Analysis, 17(3): 365–374. https://doi.org/10.1016/j.media.2013.01.003 Yan C, Zang Y (2010). Dparsf: A Matlab toolbox for “pipeline” data analysis of resting-state fmri. Frontiers in Systems Neuroscience, 4: 1377. Zhang Z, Xu Y, Yang J, Li X, Zhang D (2015). A survey of sparse representation: Algorithms and applications. IEEE Access, 3: 490–530. https://doi.org/10.1109/ACCESS.2015.2430359