Parietal, INRIA
@Parietal_INRIA
Followers
2K
Following
141
Media
6
Statuses
369
INRIA Neurospin research team. NeuroImaging + computer science = modelling brain structure, function and variability based on high-field MRI data.
Joined September 2012
RT @Philippe_Ciuciu: Le 12 avril 13-14 de @franceinter delocalisé au @CEA_Officiel à @NeuroSpin_91 pour une journée spéciale IA et santé.….
0
10
0
RT @Philippe_Ciuciu: L'offre d' Ingénieur-chercheur CEA en neuro-imagerie computationnelle par IRM clinique à 7 Tesla et au-delà (CDI),au s….
0
8
0
RT @Parietal_INRIA: One contrast per day: moving the right hand vs moving other body parts, reproduced from the HCP motor task (10.1016/j.n….
0
3
0
One contrast per day: moving the right hand vs moving other body parts, reproduced from the HCP motor task (10.1016/j.neuroimage.2013.05.033).All IBC contrats available in and
1
3
8
RT @misovalko: The amazing performance of these two on the stage! The up and coming young @Inria's stars @mathusmassias and @tomamoral, inv….
0
4
0
RT @nicolas_courty: Latest work on Optimal Transport for comparing human cortical surfaces by @alexisthual and @sixulm1 and amazing set of….
0
1
0
RT @alexisthual: Our paper on using Optimal Transport to compare human cortical surfaces was accepted at #NeurIPS22 🥳.We implement a new OT….
0
48
0
RT @StanDehaene: Aligning two human brains, in spite of anatomical and functional variability across individuals, remains a major hurdle fo….
0
16
0
RT @tomamoral: And before going to NO, happy to see this work presented live at #NeurIPSParis!!. Great job by @Mat_Dag to fasten and simpli….
0
5
0
RT @alexandrebln: How many active voxels are there in this cluster ? This question can be addressed using Notip: Non-parametric True Discov….
0
18
0
Some take-home messages:.* Public data are great ! We **need** @Neurovault and @CognitiveAtlas these projects need to be further developed.* The most difficult aspect of open-ended decoding is cognitive labelling.* brain atlases make decoder training efficient and powerful.
0
3
1
Many terms are very acurately decoded. Some decoding failures illustrate the limitations of currently available datasets.
1
1
0
We show that mutli-resolution atlases (DiFuMo) outperform fixed-resolution atlases. The more regions, the better the decoding.
1
1
3
Our Neurovault-wide decoder outperforms previous meta-analytic tools, such as Neurosynth and GCLDA.
1
1
0
We could successfully decode about 50 terms from unseen Neurovault collections. Fors this we had to clean and improve image annotations, which proved to be the most crucial aspect of the decoding problem.
1
1
0
Huge thanks and kudos to @neurovault and @CognitiveAtlas that we relied upon to test the possibility of open-ended decoding. @russpoldrack @chrisgorgo.
1
0
2
Cognitive activity can be predicted "in the wild".For **any** cognitive neuroscience study, mental processes involved can be inferred from fMRI maps. @GaelVaroquaux @Parietal_INRIA @NeuroSpin_91
1
7
20
PhD defence of Valentin Iovene on November 23rd, 3PM at Inria Saclay Valentin Iovene will defend hi. .
0
1
5