SYLVIUS project

Predict language function for optimal surgical strategy

The SYLVIUS project is a collaboration with Hospital del Mar (Barcelona) and is funded by Obra Social "la Caixa".

SYLVIUS is a workbench for the neurosciences that focuses on epilepsy surgery, in particular on computer-aided interventions for the diagnostic and treatment of therapy-resistant epilepsy.


Over the past decade, the treatment of epilepsy has undergone a revolution thanks to the introduction of robots and computer-aided frameless stereotactic surgery. Not so long ago, the implantation of electrodes in deep brain structures for the identification of the epileptogenic focus needed cumbersome stereotactic frames, which involved time-consuming manipulation and resulted in only a few electrodes per patient. Computer-aided frameless stereotactic surgery uses robots to navigate the implantation in order to pinpoint the insertion point and the exact trajectory of the electrode.

This has resulted in procedures that are safer, more efficient and faster. This allows, as in the case of the Hospital del Mar, the insertion in a single surgical session of 10 to 16 electrodes per patient, each electrode consisting of up to 15 contacts. Such exceptional electrode density allows the neurophysiologist team to monitor a broad volume of the brain and increase the chances to detect and characterize the epileptogenic network that ultimately causes the seizure.

Depending on different features of the epileptogenic network, the neurosurgeon can thus decide the best treatment approach: either taking advantage of the network of implanted electrodes, by using thermal coagulation by radiofrequency applied through one of the contacts or by a subsequent intervention, removing brain tissue containing the network responsible for the onset of the seizures.


The paradigm shift brought by robots and computer-aided frameless stereotactic surgery is forcing neurosurgeons, neurophysiologists, and neurologists to adopt new strategies for dealing with the overwhelming amount of information required for the planning of epilepsy surgeries. This requires new approaches to providing an effective communication platform where to discuss the surgical approach.

Otherwise, the multimodality and multidisciplinary nature of the problem results in improvisation, long and inefficient discussions (how to describe a 3D trajectory by email?), its consequent over-simplification and neglect of the information available, and the waste of personnel time and hospital resources.

Our experience suggests that robot-aided electrode implantation procedures may benefit from a single platform integrating the essential information and allowing the inputs and outputs of at each of the team disciplines, organizing the workflow so that each member can communicate with the precision and rigour required in this kind of surgery.


Based on the experience in Hospital del Mar, where nearly 30 robot-aided epileptic procedures are done every year, we have identified the following crucial points to consider in the design of the next generation of control platforms for computer-aided frameless stereotactic surgery:

Integration of the different neuro-imaging and functional diagnostic modalities, not only from raw sources (T1-3D, CT, angiography, PET, fMRI) but also from processed studies (cortical segmentation, MAP, tractography, connectivity maps).

Natural 3D visualization and manipulation, so that the neurosurgeon and epileptologist can visualize and evaluate the electrode implantation plan in a precise, less ambiguous and error-prone way, using a 3D platform like the zSpace.

Post-surgical follow-up through the integration of SEEG monitoring, tract validation, regional connectivity, epileptogenic network identification, seizure computational predictive models, etc.

Information sharing of the surgery plan and outcome with the team members as well as with external collaborators, cooperation centres, other institutions, possibly using different robotic platforms.

We believe that the treatment of epilepsy is at the forefront of the coming 'robotic-based' interventions and could be seen as the future of neurosurgical interventions, relying on robots to guide trajectories.

In the News