Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Brain connectome fingerprinting is rapidly rising as a novel influential field in brain network analysis. Yet, it is still unclear whether connectivity fingerprints could be effectively used for mapping and predicting disease progression from human brain data. We hypothesize that dysregulation of brain activity in disease would reflect in worse subject identification. We propose a novel framework, Clinical Connectome Fingerprinting, to detect individual connectome features from clinical populations. We show that “clinical fingerprints” can map individual variations between elderly healthy subjects and patients with mild cognitive impairment in functional connectomes extracted from magnetoencephalography data. We find that identifiability is reduced in patients as compared to controls, and show that these connectivity features are predictive of the individual Mini-Mental State Examination (MMSE) score in patients. We hope that the proposed methodology can help in bridging the gap between connectivity features and biomarkers of brain dysfunction in large-scale brain networks.
Dimitri Nestor Alice Van De Ville, Friedhelm Christoph Hummel, Gabriel Girard, Takuya Morishita, Elena Beanato, Lisa Aïcha Mireille Julie Fleury, Maximilian Jonas Wessel, Philipp Johannes Koch, Philip Egger, Andéol Geoffroy Cadic-Melchior
Silvestro Micera, Michael Lassi
Silvestro Micera, Michael Lassi