The Functionnectome framework1,2 integrates functional and diffusion MRI by mapping fMRI timeseries onto structural connectivity priors (probability maps) derived from healthy controls. This “standardized” method is unsuitable for patients with abnormal brain anatomy. We aimed to generate subject-specific structural connectivity priors to develop an “individualized” Functionnectome.
MRI data from 26 glioma patients (9F, 43±17y) included multishell diffusion imaging3,4 and task-based fMRI for motor and language functions. Whole-brain tractograms were generated using PFT5, iFOD26 (probabilistic), and TensorDet7 (deterministic) methods, registered to the MNI, and converted into the set of probability maps, one per brain voxel, using a self-written Python code. The functionnectomes were estimated using standardized and three individualized approaches, with Z-statistics activation maps computed via general linear model8.
Anatomical priors were validated by calculating Pearson’s correlation between corresponding spatial independent components9, obtaining ~0.5 for probabilistic and ~0.2 for deterministic, which was discarded. Similarity metrics between functionnectome maps revealed that the two individualized methods had higher correlation (~0.7 vs ~0.5), with PFT matching better with the standardized case, while showing lower overlap with tumors (p<0.0001). The motor PFT-functionnectome exhibited better anatomical plausibility and cross-subject reproducibility (50-80% vs 15-50%) compared to standardized, as measured through alignment to with an external tract atlas10.
Both individualized probabilistic approaches provided reliable anatomical priors, with strong correlation between them. While individualized and standardized functionnectomes showed good agreement, individual variability significantly influenced results. The individualized method demonstrated superior anatomical accuracy and specificity. Future research should explore its relevance to determine whether the increased computational cost is justified.
References
1 Nozais V, Forkel SJ, Foulon C, Petit L, Thiebaut de Schotten M. Functionnectome as a framework to analyse the contribution of brain circuits to fMRI. Commun Biol. 2021 Sep 2;4(1):1035. doi: 10.1038/s42003-021-02530-2. PMID: 34475518; PMCID: PMC8413369.
2 Nozais V, Theaud G, Descoteaux M, Thiebaut de Schotten M, Petit L. Improved Functionnectome by dissociating the contributions of white matter fiber classes to functional activation. Brain Struct Funct. 2023 Dec;228(9):2165-2177. doi: 10.1007/s00429-023-02714-y. Epub 2023 Oct 7. PMID: 37804431.
3 Manners DN, Gramegna LL, La Morgia C, Sighinolfi G, Fiscone C, Carbonelli M, Romagnoli M, Carelli V, Tonon C, Lodi R. Multishell Diffusion MR Tractography Yields Morphological and Microstructural Information of the Anterior Optic Pathway: A Proof-of-Concept Study in Patients with Leber’s Hereditary Optic Neuropathy. Int J Environ Res Public Health. 2022 Jun 5;19(11):6914. doi: 10.3390/ijerph19116914.
4 Castellaro M, Moretto M, Baro V, Brigadoi S, Zanoletti E, Anglani M, Denaro L, Dell’Acqua R, Landi A, Causin F, d’Avella D, Bertoldo A. Multishell Diffusion MRI-Based Tractography of the Facial Nerve in Vestibular Schwannoma. AJNR Am J Neuroradiol. 2020 Aug;41(8):1480-1486. doi: 10.3174/ajnr.A6706
5 Girard G, Whittingstall K, Deriche R, Descoteaux M. Towards quantitative connectivity analysis: reducing tractography biases. Neuroimage. 2014 Sep;98:266-78. doi: 10.1016/j.neuroimage.2014.04.074
6 Tournier JD, Calamante F, Connelly A. Improved probabilistic streamlines tractography by 2nd order integration over fibre orientation distributions. Proc. Intl. Soc. Mag. Reson. Med. 2010, 18:1670
7 Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A. In vivo fiber tractography using DT-MRI data. Magn Reson Med. 2000 Oct;44(4):625-32. doi: 10.1002/1522-2594(200010)44:4<625::aid-mrm17>3.0.co;2-o
8 Woolrich MW, Ripley BD, Brady M, Smith SM. Temporal autocorrelation in univariate linear modeling of FMRI data. Neuroimage. 2001 Dec;14(6):1370-86. doi: 10.1006/nimg.2001.0931
9 O’Muircheartaigh J, Jbabdi S. Concurrent white matter bundles and grey matter networks using independent component analysis. Neuroimage. 2018 Apr 15;170:296-306. doi: 10.1016/j.neuroimage.2017.05.012.
10 Radwan AM, Sunaert S, Schilling K, Descoteaux M, Landman BA, Vandenbulcke M, Theys T, Dupont P, Emsell L. An atlas of white matter anatomy, its variability, and reproducibility based on constrained spherical deconvolution of diffusion MRI. Neuroimage. 2022 Jul 1;254:119029. doi: 10.1016/j.neuroimage.2022.119029
