Epilepsy suRgEry aSsistEd by funCtional ulTrasound Imaging in fOcal cortical dysplasia



Focal cortical dysplasia (DFC) are responsible for severe epilepsy that often begins in infancy with progressive neurological deficits. This epilepsy is resistant to drugs treatment in 25% of cases, but can be cured by surgery if the removal of the diseased brain is complete. However, these lesions are located deep in the brain and in two thirds of cases in functional area ie in areas that are important for everyday life (movement, sight, speech). These brain regions that are affected are otherwise difficult to differentiate of the healthy brain with conventional imaging modalities or examination during operation. Therefore there is a extensive risk of incomplete removal and failure of the surgery.

New imaging tools are needed to precisely differentiate the dysplastic tissue compared to healthy tissue. We are developing a new ultrasound imaging technique with a sensitivity is 40x higher than the best current commercial systems. The identification of DFC2 has already been observed using ultrasound which confirm the relevance of our approach. We have recently obtained agreements of the health authorities to evaluate our technique on a series of 20 patients. The goal is to identify the dysplastic tissue with high precision and in real time during surgery, to delimit the area that will be subsequently removed by the neurosurgeon and to control its effective removal.



I. Focal cortical dysplasia

Focal cortical dysplasias are lesions of the cerebral cortex responsible for severe early-onset epilepsy, eligible for curative surgery. Described by Taylor et al. in 1971 (1) they are characterized by major cyto-architectural abnormalities involving loss of cortical lamination, the presence of neurons and dysmorphic giant cells (balloons cells) in the cortex and the white matter adjacent (Figure 1) . They result from an abnormality of cortical development occurred at an early stage of embryogenesis and show histological similarities with lesions in tuberous sclerosis and hemi-megalencephaly. Recent classifications have been renamed DCF type 2 (DCF2) (2).

Figure 1: Cytoarchitectural abnormalities characterizing DFC2 (haematoxylin phloxine saffron: HPS). Cortical disorganization with the presence of giant dysmorphic neurons (blue arrows) and large glial cells cytoplasm ballonisé (balloons cells, white arrows, original magnification 400x, courtesy of Professor C. Daumas-Duport, CHSA).


DCF2 are representing the third cause of epilepsy in adults treated surgically after hippocampal sclerosis and neuro-developmental tumors, and the leading cause in children (3-4). Their location is primarily non-temporal, particularly at the central frontotemporal cortex (5-7) They can be of variable size, ranging from large to microscopic lesions forms, including the entire thickness of the cortical mantle until ventricular wall. The DCF2 are characterized by an intrinsic epileptogenic activity (Figure 2), demonstrated through pre or peroperative intracerebral recordings  (5-6). These characteristics have been found in vitro in slices of dysplastic cortex of human origin (8) have shown that the giant neurons are hyperexcitable while the balloon cells are electrically silent (9).  

Figure 2: Intralesional interictal SEEG recording traces in a DCF2 (polyspikes frequency 3-10 Hz, bursts of fast spikes interspersed with depression phase of activity (according Chassoux et al, Brain 2000).

DCF2 are difficult to detect on first generation MRI were initially considered rare and with poor surgical prognosis. Progress in recent years has made DFC2 a prototype of lesional epilepsy that is surgically curable, with up to 90% of patients free of seizures (7, 10). Their recognition is essential and has transformed the prognosis of severe epilepsies, particularly in children. The typical appearance on MRI has localized cortical thickening associated with poor differentiation between white matter and gray matter and hyperintense white matter on FLAIR sequences (Figure 3). Abnormalities predominate at the bottom of grooves, which have a conical termination, tapered towards the ventricles, described as the " transmantle sign". These anomalies are detected by MRI in 60-80 % of cases, however, they require a rigorous technique guided by electro-clinical data (10-11).

Figure 3: Typical appearance of a 3 Tesla MRI DCF2. Axial, T1-weighted and FLAIR, right anterior cingulate gyral abnormality (arrow) with discrete cortical thickening, poor differentiation gray-white at the bottom of the groove, hyperintense subcortical. (courtesy of Dr. C. Mellerio CHSA).

MRI techniques based on automatic analysis of the texture and morphology of cortical offer greater sensitivity of detection (12-13) but are not commonly used. Positron emission tomography (PET) with fluoro-deoxyglucose (FDG) showed high sensitivity to identify DCF2 (14-16), further enhanced by the registration of PET images obtained by MRI (Figure 4).


Figure 4: DCF2 with MRI initially considered negative, MRI axial T1, FDG-PET, PET fusion on MRI. Focal left frontal hypometabolism by detecting gyral abnormality (cortical thickening, transmantle sign) characteristic of a DCF2. (SHFJ, CEA)


The severity of epilepsy and related disorders justifies the indication of surgical treatment that is effective in 60-80% of cases (4, 6-7, 10, 17-18). Favorable results are correlated with complete resection of the dysplastic cortex. It should be noted the absence of morphological abnormalities observed intraoperatively for the dysplastic cortex but not the healthy tissue (Figure 5).       


Figure 5: Excision of a left precentral DCF2 in a 23 years old man. A) preoperative MRI, axial sections T1, T2 amount lesion (arrow). B) MRI neuronavigation (the left side of the brain is here to the right of the screen) for the intraoperative identification of dysplastic gyrus. C) intraoperative macroscopic appearance showing the absence of intraoperative macroscopic distinction between dysplastic cortex and healthy cortex, intraoperative stimulation to identify the motor cortex, end audit excision of the persistence of motor responses. D) postoperative MRI checking the completeness of resection.


II. Functional Ultrasound an innovative modality for intra-operative imaging of brain activity in real-time

DCF2 are highly epileptogenic lesions with preferential localization in functional area making their removal by surgery particularly difficult. They are often not well defined and do not differ from healthy cortex during the intraoperative macroscopic exams. Surgery outcome are better in recent years, they are clearly correlated with a complete excision of dysplastic cortex. However, the surgery may be limited in the absence of precise images and a significant number of patients are not cured of their epilepsy, with only 50-75 % of patients with class IA Engel (7) Moreover, the postoperative morbidity remains high, almost one in two patients with post operative neurological deficit and 7% of permanent sequelae. Finally, in 10-25 % of cases interventions 2 (or 3) are performed for the removal of crises. New tools to differentiate precisely the dysplastic cortex and the healthy tissue at the lowest possible level are needed. The identification of dysplastic cortex that has a specific echogenicity has been assessed using ultrasound (B-mode) (26) . However, the value of ultrasound in identifying the boundaries of dysplastic cortex has not yet been reported on a series of patients.

The potential contribution of functional ultrasound imaging applies particularly well to DCF2 surgery because it is a focal lesion, buried deep in the cortex, difficult to discern macroscopically in healthy tissue but with specific neurophysiological, metabolic and hemodynamic characteristics that should be clearly identified during surgery using the technique that we have developed and we propose to adapt in clinic.


1. Objectives of the study

Our goal is to provide to the neurosurgeons qualitative and quantitative data in real time to accurately resect the dysplastic cortex without affecting the healthy tissue by using an innovative brain ultrasound imaging system (μDoppler) during neurosurgery.

This approach aims to improve the information provided by conventional preoperative imaging and ultrasound probes currently available in the clinic, it should offer the neurosurgeon a new intraoperative tool to improve the accuracy of its gesture and functional outcome of the surgery.

Our medical team at the CHSA has a recognized expertise in the management of patients undergoing DCF2 and has sufficient recruitment to evaluate the usefulness of this new tool. These patients should therefore be among the first beneficiaries of this technological advance, since the healing of their epilepsy depends heavily on the completeness of surgical excision. As a reminder, DCF2 is often located functional area, especially near the sensorimotor cortex. Therefore improving the distinction between functional/healthy cortex and epileptogenic dysplastic cortex offers new perspectives for patients whose seizures persist after incomplete resection.     


2. Methodology of the study

It is a single-center observational study on a reduced cohort because of the high specificity of the study and the assumed high value of the evaluated technique.

We first describe the characteristics and advantages of ultrasound imaging in DCF2, its adaptation to the clinical use, technical measures during the surgery and their validation by establishing correlations with other parameters.

  1. Development of a new ultrasound system dedicated to clinical functional brain imaging

Since one year we are developing a faster version of our imaging system termed Echo1 (Figure 8) that is designed for use in preclinical and clinical research. This device is based on the same basic principles but we performed a few technical improvements to overcome the limits imposed by the previous generation of ultrasound (low memory and reduced computational speed), while having an even greater flexibility.

This imaging system offers a number of opportunities that are not accessible to conventional ultrasound as functional imaging mode in real time as well as several improvements in the quality of image morphological mode (B-mode) and micro-Doppler. This medical device has been developed by a small multidisciplinary and passionate team with one main goal: to make a very efficient imaging system, reliable, easy to control and affordable for both doctors in hospitals and research laboratories to ensure its democratization in the medium term. To do so we started from scratch and we worked on all the elements that are present in a standard ultrasound namely electronic transmission / reception of the ultrasonic waves, the computer to generate images, ultrasound probes, Doppler sequences and imaging software.

In a first step, we completely rethought how imaging data are analyzed, stored and generated by the system. For this step, it was necessary to decouple the functions of transmission/reception supported by a specific module and the calculation of images that is assigned to a workstation whose components have been selected and assembled by our team. This configuration has an acquisition and image processing power far superior to all commercial devices currently available and more can be updated very simply as technology changes. Our strategy is based on the combined use of two technologies: the generic calculation on a GPU (GPGPU) to benefit from their massively parallel processing and the use of professional SSD drives very high transfer rate (1.5 GB/s) can quickly store large amount of data generated during the reception of the waves. Thus, for the first time in the world of functional imaging, we were able to achieve real-time imaging with high quality that opens the way to many applications in preclinical and clinical research (2 patents filed in 2014).       

Figure 8: Schematic representation of the components of our ultrasound scanner dedicated to brain imaging.


In a second step, we have designed an ultrasound probe whose technical characteristics have been specifically selected for functional imaging. Indeed, the ultrasonic sensor is the heart of a medical ultrasound imaging system. Performance of the probe will indeed determine to a large extent the quality of the images provided by the system. For these experiments, we chose a probe geometry with a linear array (1D) formed by an array of 64 piezoelectric PZT elements. A schematic of the probe is shown in Figure 5. The inter-element width, which defines the spatial resolution, was reduced to only a 100 microns step. The center frequency of 15 MHz was chosen because it allows a good compromise between resolution and depth. This custom configuration enabled us to minimize the overall size of the probe (12 x 6 mm Figure 5A) and therefore its weight in order to facilitate its use on animals of small size (rodents) and during surgery in human. Acoustic performance is not the only requirements of the ultrasound probe that we want to use for this project in the clinic. Indeed, like all medical materials, the probe meets standards ensuring the safety of physician and patient. Several aspects will be taken into account during the qualification process and manufacturing as electrical safety, thermal cycling, the acoustic output, biocompatibility and resistance to sterilization. With the aim to further increase the precision of surgery, we want to increase the resolution of our technique. For this, we designate two new ultrasonic probes with technical characteristics very close to the initial probe but with a large number of elements and/or a different geometric organization while respecting specific size constraints. These probes will be needed to analyze the dysplastic focus in the case of DFC2 located deep inside the brain (greater than 3 cm from the dura). They will also allow a gain in resolution and field of view. Specialized French companies in the field will do manufacturing of these probes.

Figure 5 : CAD 2D and 3D ultrasound probe dedicated to functional imaging data. A. Front view of the probe. B. Side view showing the fastening system of the probe C. View from above. D. 3D sensor format 2 View: 1. The values are in mm.

The third part of our work was to design a set of new ultrasound sequences suitable for functional imaging. The critical point for the Doppler imaging is the repetition rate of the pulses (several kHz) to properly sample the Doppler signal. Conventional Doppler mode scan the medium line by line with focused beams but are too slow to acquire an entire image, which involves dividing the image into several sectors scanned sequentially. Added to this constraint is added the real-time imaging problem for commercial ultrasound and therefore classical Doppler systems use few shots for each line, typically 15. The speed estimation or blood volume from such a short Doppler signal is difficult, which explain why small vessels are not detectable with this method. With the high-speed imaging, a single plane wave is required to rebuild a complete picture. The imaging frame rate is no longer a limitation since speeds of the order of 20 kHz can be achieved. To increase significantly signal/noise ratio compared to conventional Doppler mode, we have developed an imaging sequence by combining several images taken of plane waves of different impacts and the number of which can be tailored to the characteristics of the lesion (size, depth). Meanwhile, as there is no loss of time to scan the image areas, the number of pulses "seen" by each pixel is increased to typically reach 200 pulses. The combination of improved signal to noise and a longer Doppler signal is called μDoppler mode, which increases the sensitivity, by a factor of from 30 to 100 compared to Doppler mode. This work on the ultrasonic waves was supplemented by an important development in the filtering of the image that is essential to extract information corresponding to the movement of red blood cells. We use a series of so-called intelligent filters that eliminate the noise associated with cardiac and respiratory movements while enhancing the signal.

To propose a new comprehensive tool, we also worked on the software interface to control our imaging system and the development of dedicated plug-ins for doctors, researchers and physicians who use functional imaging. Indeed, it is necessary to have a precise control of the physiological parameters of the subject during the imaging experiments, including devices to monitor body temperature, respiration, heart beat (ECG), brain rhythms (EEG). We have developed a dedicated software called PILOT consists of a central application to control the system and modules dedicated capable of displaying real-time constants and record readings of all devices used in the imaging experiments. Figure 9 shows example of the interface during ​​functional imaging in preclinical experiments .

Figure 9 : Example of using the PILOT GUI. OBI credits.

This easy to use interface allows real-time monitoring, data management and offers a dedicated module for the analysis of experiments including export data in most common formats. An external acquisition card that connects different devices via a BNC, USB or serial port ensures compatibility with external devices.

Finally, we designed a set of ultrasound sequences that are completely new and suitable for clinical use in both infants and adults. These optimizations are described in Part B of this document.


B. Functional brain ultrasound imaging to improve surgery DFC2

Brain ultrasound imaging has many advantages for neurosurgery. First, the spatiotemporal resolution of this technique exceeds MRI and PET. Moreover, ultrasound would be a very effective tool in addition to the neuronavigation already used for imaging in real time and with high accuracy the area to be resected. Our goal is to apply our new highly sensitive imaging sequence to DCF2 to increase the precision of surgery.

For DCF2, brain ultrasound imaging will be used according to 2 modes:

  • a μDoppler mode to study in detail the structure and microvasculature of the area containing dysplastic tissue
  • a morphological mode to accurately assess the movement of tissues (brain shift) during resection and to guide the surgeon's in complement of neuronavigation.     


The μDoppler mode to better visualize the dysplastic tissue

The Doppler imaging is based on the detection of the movement of the red blood cells (25). The acquisition step consists in sending successive ultrasonic pulses in the medium and record the echoes produced after each pulse. In the presence of blood in the pixel, the intensity of the recorded signal fluctuates over time due to the movement of red blood cells with a characteristic frequency called Doppler frequency directly proportional to their speed. After a filtering step that eliminates unwanted movements from tissues, various information can be extracted from the signal of blood flow, called µDoppler signal, such as the speed or volume of the blood.


μDoppler brain imaging that we will use in this project is based on a new way of doing Doppler imaging as shown in Figure 10. With this new method, we can perform very precise images of the cerebral microvasculature in only 200ms and with good spatial resolution (100 μm2 in the plane of the image) that can be further improved by increasing the frequency of ultrasound. Indeed, the penetration of ultrasound allow us to view the surface of the cortex but also the deeper regions (up to 5 cm from the dura) over a very wide field imaging that is compatible with the deep cortical localization of DCF2. μDoppler brain imaging should contribute to the improvement of current surgical techniques (neuronavigation, intraoperative stimulation) that often lack precision in highly functional areas such as the motor cortex, where DCF2 are often found.      


Thus, all ultrasound images in patients will be made using the ultrasonic probe protected with a sterile sheath and applied directly on the brain surface after opening the dura. The μDoppler images of brain vasculature required no mechanical vibrations and therefore an effective fixation of the ultrasound probe to the skull. This attachment will be managed using a custom designed probe holder and will measure the precise application of the probe to the cortex before resection - and before sulcal dissection - without making any pressure on the surface, and thus without risk of cortical injury.

The morphological mode to better address DFC2

Intraoperatively, after opening the dura mater, the brain structures are deformed under the influence of several factors such as the change in the intracranial pressure, the flow of cerebrospinal fluid, the hydrostatic pressure exerted on brain volume exposed, and edematous reactions during resection. These intra-operative deformation, known as "brain shift", then make the information provided by the neuronavigation system imperfect and source of errors for the surgeon. In this context, it appears that brain ultrasonic imaging using the morphological mode could be used to accurately quantify the brain shift with a high spatial resolution. From a practical point of view, we try to follow the movement of the ultrasound probe using an neuronavigation system already present in the surgical block to calculate the 3D position and orientation of the probe (Figure 11A). This approach will allow to locate the intraoperative ultrasound image overlaid on the MRI image (Figure 11B), but also to reconstruct a 3D volume from multiple 2D slices to observe three-dimensional deformations during the resection.

Figure 11 : Interest in intraoperative imaging was morphological mode. A. Basis of intraoperative hand free  ultrasonography. Neuronavigation system which is equipped with cameras continuously calculate the position and orientation of the ultrasound probe. This technique allows to match the preoperative images (MRI) and intraoperative ultrasound images. B. Reconstruction of cutting the corresponding preoperative MRI ultrasound image taken by standard Doppler imaging. Intraoperative ultrasound can visualize hyper echogenic lesions, Fluid brain areas such as grooves or ventricles, and the falx. Adapted from Rasmussen, 2007.

The potential benefits of brain ultrasound imaging are numerous. First, it is a real-time modality, the image formation is instantaneous, unlike the MRI. In addition, ultrasound allows in-depth study of three-dimensional deformations after reconstruction. The image resolution is much higher than that of the MRI images. In addition to morphological images, it is possible to visualize in μDoppler small blood vessels, and also hemodynamic changes related to the intrinsic tissue changes in DCF2. Another important point is the comacity of this imaging system in surgical block and it’s affordable cost for most hospitals. 


3. Detail of the clinical trial

A) Measurement during the surgery

To validate the sensitivity of our new μDoppler sequence applied to DCF2, we will realize for each patient a series of detailed images of the surgical area after opening the dura, before resection but also after resection – inside and around the lesion. These measures should help to define the characteristics of dysplastic tissue and to compare them with those of healthy tissue.

The surgical protocol will not change, morphological ultrasound measurements and guidance being added to the usual operating time. After general anesthesia of the patient and fixation of its head, the usual environment is installed - operating microscope, neuronavigation, cortical stimulation device - and the intervention will include: skin incision, craniotomy adapted to the size and location dysplasia  the opening of the dura mater, the sulcal and gyral identification using neuronavigation and its validation by the identification of these structures under the operating microscope. Before sulcal dissection, μDoppler and morphological ultrasound measurement will be carried out (estimated at 10-15 min scanning time), and then the dissection is resumed; once the dysplasia is identified macroscopically - abnormal color and consistency of the cortex, spatial correlation with the pointer neuronavigation - it is resected (usually in several fragments that are labeled for subsequent correlation) until the complete resection of dysplastic tissue.  This resection is estimated at a macroscopic level and the spatial position of the border of the dysplastic tissue are explored using the neuronavigation pointer and supplemented by information provided by ultrasound imaging in morphological mode. At the end of resection, a new ultrasonic measurement will be conducted to check for any residual hyperechoic abnormal tissue and measure the brain-shift. When DCF2 is larger, occupying for example an entire gyrus, it is usually resected in a block with adjacent resection in safe area to be complete, thereby again providing healthy and pathological samples for histologic correlative study and echogenicity .

In all cases, the morphological parameters measured by this new ultrasound technique (echogenicity and structure of the microvasculature) will be correlated with data from the preoperative imaging (morphological imaging ,T1- ,T2, FLAIR, diffusion, ASL (arterial spin labeling, areas of activations in fMRI brain metabolism by FDG-PET), neurophysiological (EEG surface electrical activity in SEEG if performed) and neuropathological cortical specimen taken during surgery and oriented to define precisely the location. Particular attention will be paid to correlations between morphological parameters collected intraoperatively and analysis of anomalies cytoarchitectural characteristics DCF2 in the operating room. A ratio of the volume of dysplastic vs non-dysplastic tissue will be established whenever the resection block is possible.

B) Population studied

            Patients will be selected from the population already investigated in the neurosurgery CHSA service for partial epilepsy that is drug resistant. The selection criteria are:

Adult patient who agreed to participate in the study or patient below 18 year with parental consent and with a partial epilepsy drug resistant that is related to type 2 focal cortical dysplasia identified using preoperative MRI and or FDG-PET, possibly on the SEEG data only if the activity is characteristic while the imaging is negative.

C) Scientific outcome

            The primary endpoint will be the ability to differentiate dysplastic cortex of healthy cortex based on ultrasound images obtained intraoperatively.  Then these data will be compared with histological results by examination of cortical samples taken on the edges, superficial and deep part of the dysplasic tissue.

The secondary outcome criteria will be:

- The completeness of resection of dysplastic cortex assessed on postoperative imaging and histological examination of the surgical specimen (edge of the resection in a safe area, percentage of dysplastic tissue compared to non-dysplastic tissue in case of resection en bloc).

- The functional result of surgery (seizure outcome expressed in Engel score (25), and presence and duration of postoperative neurological deficit compared with previous results of the team (7).

D) Conduct of the study

The neurosurgery department treats approximately 40-50 patients per year for partial epilepsy that is drug resistant, 25% for DCF2. For this study, we plan to recruit 20 patients over a period of 24 months.

Care of patients will not be changed at any time during the procedure. We will only use  additional time for the completion of the ultrasound images and measures described above, corresponding to a longer operative time not exceeding 20 minutes, or about 5% of the total duration of a typical intervention without additional surgical risks. The direct benefits expected in patients compensate for this predictable lengthening operative time.

E) Expected results

Improving the quality of the resection of the dysplastic tissue during initial surgery and consequently to reduce the number of patient without epileptic crises after surgery. More precisely, the goal is to increase the number of patients with class IA Engel, no reoperations and to dimish the number of patient with neurological transient or permanent deficit after surgery.

F) Perspectives

            In case of confirmation of the contribution of brain ultrasound imaging in the surgical treatment of DCF2, expansion to other indications: partial epilepsy associated with other lesions, tumor surgery.

4. References

1 - Taylor DC, Falconer MA, Bruton CJ, et al. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiat. 1971; 34: 369-87.

2 - Blümcke I, Thom M, Aronica E, et al. The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 2011: 52: 158-74.

3 - Devaux B, Chassoux F, Guenot M, et al. Epilepsy surgery in France. Neurochirurgie 2008; 54: 453-65.

4 - Lerner JT, Salamon N, Hauptman JS, et al. Assessment and surgical outcomes for mild type I and severe type II cortical dysplasia: a critical review and the UCLA experience. Epilepsia 2009; 50: 1310-35.

5 - Palmini A, Gambardella A, Andermann F, et al. Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol. 1995; 37: 476-87.

6 - Chassoux F, Devaux B, Landré E, et al. Stereoelectroencephalography in focal cortical dysplasia: a 3D approach to delineating the dysplastic cortex. Brain 2000; 123: 1733-51.

7 - Chassoux F, Landré E, Mellerio C, et al. Type II focal cortical dysplasia: Electro-clinical phenotypes and surgical outcome related to imaging. Epilepsia, 2012; 53: 349-58.

8 - D'Antuono M, Louvel J, Köhling R, et al. GABAA receptor-dependent synchronization leads to ictogenesis in the human dysplastic cortex. Brain 2004;127: 1626-40. 

9 - Cepeda C, André VM, Flores-Hernández J, et al. Pediatric cortical dysplasia: correlations between neuroimaging, electrophysiology and location of cytomegalic neurons and balloon cells and glutamate/GABA synaptic circuits. Dev Neurosci. 2005; 27: 59-76.

10- Urbach H, Scheffler B, Heinrichsmeier T, et al. Focal cortical dysplasia of Taylor's balloon cell type: a clinicopathological entity with characteristic neuroimaging and histopathological features, and favorable postsurgical outcome. Epilepsia 2002; 43: 33-40.

11- Mellerio C, Labeyrie MA, Chassoux F, et al. Optimizing MRI detection of type 2 focal cortical dysplasia: best criteria for clinical practice. AJNR, 2012; 33:1932-8.

12- Bernasconi A, Antel SB, Collins DL, et al. Texture analysis and morphological processing of magnetic resonance imaging assist detection of focal cortical dysplasia in extra-temporal partial epilepsy. Ann Neurol. 2001; 49: 770-5.

13- Colliot O, Mansi T, Bernasconi N, et al. Segmentation of focal cortical dysplasia lesions on MRI using level set evolution. Neuroimage 2006; 32: 1621-30. 

14 - Kim YK, Lee DS, Lee SK, et al. (18)F-FDG PET in localization of frontal lobe epilepsy: comparison of visual and SPM analysis. J Nucl Med. 2002; 43: 1167-74.

15 - Salamon N, Kung J, Shaw SJ, et al. FDG-PET/MRI coregistration improves detection of cortical dysplasia in patients with epilepsy. Neurology 2008; 71:1594-601.

16 - Chassoux F, Rodrigo S, Semah F, et al. FDG-PET improves surgical outcome in negative-MRI Taylor type focal cortical dysplasias. Neurology 2010; 75: 2168-75.

17 - Fauser S, Bast T, Altenmüller DM et al. Factors influencing surgical outcome in patients with focal cortical dysplasia. J Neurol Neurosurg Psychiatry 2008; 79: 103-5.

18 - Krsek P, Maton B, Jayakar P, et al. Incomplete resection of focal cortical dysplasia is the main predictor of poor postsurgical outcome. Neurology 2009; 72: 217-23.

19 - Marnet D, Devaux B, Chassoux F, et al. Surgical resection of focal cortical dysplasias in the central region. Neurochirurgie 2008; 54: 399-408.

20 - Van Raaij ME, Lindvere L, Dorr A, He J, Sahota B, Foster FS, Stefanovic B. Functional micro-ultrasound imaging of rodent cerebral hemodynamics. Neuroimage 2011;58:100-108.

21 - Sandrin L, Catheline S, Tanter M, Hennequin X, Fink M. Time-resolved pulsed elastography with ultrafast ultrasonic imaging. Ultrason Imaging 1999;21:259-272.

22 - Bercoff J, Tanter M, Fink M. Supersonic shear imaging: a new technique for soft tissue elasticity mapping. IEEE Trans Ultrason Ferroelectr Freq Control 2004;51:396-409. 

23 - Mace E, Montaldo G, Cohen I, Baulac M, Fink M, Tanter M. Functional ultrasound imaging of the brain. Nat Methods 2011;8:662-664.

24 - Montaldo G, Tanter M, Bercoff J, Benech N, Fink M. Coherent plane-wave compounding for very high frame rate ultrasonography and transient elastography. IEEE Trans Ultrason Ferroelectr Freq Control 2009;56:489-506.

25 - Engel J, Jr., Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In Engel J, Jr. (ed). Surgical treatment of the epilepsies. 2nd Ed. Raven Press, New York, 1993, pp. 609-621.

26 – Miller D., Knake S., Bauer S., Krakow K. Pagenstecher A., Sure U., Rosenow F. Intraoperative ultrasound to define focal cortical dysplasia in epilepsy surgery. Epilepsia, 2008, 49(1): 156-158



The schedule will be organized as follows:

  • Start of the clinical trial with the enrollment of the first patient.
  • The number of operated patients is on average of 1 patient / month, the first part of the study will allow us to validate all parameters (choice of sequences) of the imaging system 
  • The table below summarizes information about the clinical trial:


Inclusion period

24 months

Duration of patient participation


12 months

Duration of treatment

 1 day

Medical monitoring

12 months

 Total duration of the clinical trial

36 months




The members participating in this study will be:

1 – Bertrand Devaux, neurosurgeon, CHSA, 30%

2 – Francine CHASSOUX, neurologist, CHSA, 25%

3 – Gabriel MONTALDO, physicist-researcher, CHSA, 50%

4 – Alan URBAN, researcher, CPN – INSERM U894, 25%