Maria Chiara Biagi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Maria Chiara Biagi has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Giulio Ruffini
Research Department, Neuroelectrics Barcelona, Barcelona, Spain
2 shared publications - Ricardo Salvador
Research Department, Neuroelectrics Barcelona, Barcelona, Spain
2 shared publications - Aurore Thibaut
Coma Science Group, GIGA-Consciousness, University of Liège, 4000 Liège, Belgium
1 shared publication - Brad Manor
Hinda and Arthur Marcus Institute for Aging Research, Hebrew SeniorLife , Boston, Massachusetts
1 shared publication - Carla Sendra-Balcells
Neuroelectrics Barcelona
1 shared publication - Charlène Aubinet
Coma Science Group, GIGA Institute, University of Liège, Belgium
1 shared publication
Latest Publications
- Open Access
- 18 March 2020
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 9 (3) , 828
Network Mapping of Connectivity Alterations in Disorder of Consciousness: Towards Targeted Neuromodulation
Disorder of consciousness (DoC) refers to a group of clinical conditions that may emerge after brain injury, characterized by a varying decrease in the level of consciousness that can last from days to years. An understanding of its neural correlates is crucial for the conceptualization and application of effective therapeutic interventions. Here we propose a quantitative meta-analysis of the neural substrate of DoC emerging from functional magnetic resonance (fMRI) and positron emission tomography (PET) studies. We also map the relevant networks of resulting areas to highlight similarities with Resting State Networks (RSNs) and hypothesize potential therapeutic solutions leveraging network-targeted noninvasive brain stimulation. Available literature was reviewed and analyzed through the activation likelihood estimate (ALE) statistical framework to describe resting-state or task-dependent brain activation patterns in DoC patients. Results show that task-related activity is limited to temporal regions resembling the auditory cortex, whereas resting-state fMRI data reveal a diffuse decreased activation affecting two subgroups of cortical (angular gyrus, middle frontal gyrus) and subcortical (thalamus, cingulate cortex, caudate nucleus) regions. Clustering of their cortical functional connectivity projections identify two main altered functional networks, related to decreased activity of (i) the default mode and frontoparietal networks, as well as (ii) the anterior salience and visual/auditory networks. Based on the strength and topography of their connectivity profile, biophysical modeling of potential brain stimulation solutions suggests the first network as the most feasible target for tES, tDCS neuromodulation in DoC patients.
- 18 March 2020
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 9 (3) , 828
Open AccessNetwork Mapping of Connectivity Alterations in Disorder of Consciousness: Towards Targeted Neuromodulation
Disorder of consciousness (DoC) refers to a group of clinical conditions that may emerge after brain injury, characterized by a varying decrease in the level of consciousness that can last from days to years. An understanding of its neural correlates is crucial for the conceptualization and application of effective therapeutic interventions. Here we propose a quantitative meta-analysis of the neural substrate of DoC emerging from functional magnetic resonance (fMRI) and positron emission tomography (PET) studies. We also map the relevant networks of resulting areas to highlight similarities with Resting State Networks (RSNs) and hypothesize potential therapeutic solutions leveraging network-targeted noninvasive brain stimulation. Available literature was reviewed and analyzed through the activation likelihood estimate (ALE) statistical framework to describe resting-state or task-dependent brain activation patterns in DoC patients. Results show that task-related activity is limited to temporal regions resembling the auditory cortex, whereas resting-state fMRI data reveal a diffuse decreased activation affecting two subgroups of cortical (angular gyrus, middle frontal gyrus) and subcortical (thalamus, cingulate cortex, caudate nucleus) regions. Clustering of their cortical functional connectivity projections identify two main altered functional networks, related to decreased activity of (i) the default mode and frontoparietal networks, as well as (ii) the anterior salience and visual/auditory networks. Based on the strength and topography of their connectivity profile, biophysical modeling of potential brain stimulation solutions suggests the first network as the most feasible target for tES, tDCS neuromodulation in DoC patients.
- Open Access
Convolutional neural network MRI segmentation for fast and robust optimization of transcranial electrical current stimulation of the human brain
The segmentation of structural MRI data is an essential step for deriving geometrical information about brain tissues. One important application is in transcranial electrical stimulation (e.g., tDCS), a non-invasive neuromodulatory technique where head modeling is required to determine the electric field (E-field) generated in the cortex to predict and optimize its effects. Here we propose a deep learning-based model ( StarNEt ) to automatize white matter (WM) and gray matter (GM) segmentation and compare its performance with FreeSurfer , an established tool. Since good definition of sulci and gyri in the cortical surface is an important requirement for E-field calculation, StarNEt is specifically designed to output masks at a higher resolution than that of the original input T1w-MRI. StarNEt uses a residual network as the encoder (ResNet) and a fully convolutional neural network with U-net skip connections as the decoder to segment an MRI slice by slice. Slice vertical location is provided as an extra input. The model was trained on scans from 425 patients in the open-access ADNI+IXI datasets, and using FreeSurfer segmentation as ground truth. Model performance was evaluated using the Dice Coefficient (DC) in a separate subset (N=105) of ADNI+IXI and in two extra testing sets not involved in training. In addition, FreeSurfer and StarNEt were compared to manual segmentations of the MRBrainS18 dataset, also unseen by the model. To study performance in real use cases, first, we created electrical head models derived from the FreeSurfer and StarNEt segmentations and used them for montage optimization with a common target region using a standard algorithm ( Stimweaver ) and second, we used StarNEt to successfully segment the brains of minimally conscious state (MCS) patients having suffered from brain trauma, a scenario where FreeSurfer typically fails. Our results indicate that StarNEt matches FreeSurfer performance on the trained tasks while reducing computation time from several hours to a few seconds, and with the potential to evolve into an effective technique even when patients present large brain abnormalities.
Open AccessConvolutional neural network MRI segmentation for fast and robust optimization of transcranial electrical current stimulation of the human brain
The segmentation of structural MRI data is an essential step for deriving geometrical information about brain tissues. One important application is in transcranial electrical stimulation (e.g., tDCS), a non-invasive neuromodulatory technique where head modeling is required to determine the electric field (E-field) generated in the cortex to predict and optimize its effects. Here we propose a deep learning-based model ( StarNEt ) to automatize white matter (WM) and gray matter (GM) segmentation and compare its performance with FreeSurfer , an established tool. Since good definition of sulci and gyri in the cortical surface is an important requirement for E-field calculation, StarNEt is specifically designed to output masks at a higher resolution than that of the original input T1w-MRI. StarNEt uses a residual network as the encoder (ResNet) and a fully convolutional neural network with U-net skip connections as the decoder to segment an MRI slice by slice. Slice vertical location is provided as an extra input. The model was trained on scans from 425 patients in the open-access ADNI+IXI datasets, and using FreeSurfer segmentation as ground truth. Model performance was evaluated using the Dice Coefficient (DC) in a separate subset (N=105) of ADNI+IXI and in two extra testing sets not involved in training. In addition, FreeSurfer and StarNEt were compared to manual segmentations of the MRBrainS18 dataset, also unseen by the model. To study performance in real use cases, first, we created electrical head models derived from the FreeSurfer and StarNEt segmentations and used them for montage optimization with a common target region using a standard algorithm ( Stimweaver ) and second, we used StarNEt to successfully segment the brains of minimally conscious state (MCS) patients having suffered from brain trauma, a scenario where FreeSurfer typically fails. Our results indicate that StarNEt matches FreeSurfer performance on the trained tasks while reducing computation time from several hours to a few seconds, and with the potential to evolve into an effective technique even when patients present large brain abnormalities.