A level set driven by MR features of focal cortical dysplasia for lesion segmentation
ABSTRACT Focal cortical dysplasia (FCD), a malformation of cortical development, is an important cause of medically intractable epilepsy. FCD lesions are difficult to distinguish from non-lesional cortex and their de-lineation on MRI is a challenging task. This paper presents a method to segment FCD lesions on T1-weighted MRI, based on a 3D deformable model, implemented using the level set framework. The deformable model is driven by three MRI features: cortical thickness, relative intensity and gradient. These features correspond to the visual characteristics of FCD and allow to differentiate lesions from normal tissues. The proposed method was tested on 18 patients with FCD and its performance was quantitatively evaluated by comparison with the manual tracings of two trained raters. The validation showed that the similarity between the level set segmen-tation and the manual labels is similar to the agreement between the two human raters. This new approach may become a useful tool for the presurgical evaluation of patients with intractable epilepsy.
- SourceAvailable from: ncbi.nlm.nih.gov[show abstract] [hide abstract]
ABSTRACT: An unusual microscopic abnormality has been identified in the lobectomy specimens removed surgically from the brains of 10 epileptic patients. The abnormality could seldom be identified by palpation or with the naked eye. Histologically, it consisted of congregations of large, bizarre neurones which were littered through all but the first cortical layer. In most, but not in all cases, grotesque cells, probably of glial origin, were also present in the depths of the affected cortex and in the subjacent white matter. This kind of abnormality appears to be a malformation. The picture is reminiscent of tuberous sclerosis but too many distinguishing features, both in the clinical and in the pathological aspects, make this diagnosis untenable. The cases are therefore looked on provisionally (since all but one are still alive) as comprising a distinct form of cortical dysplasia in which localized, exotic populations of nerve cells underlie the electrical and clinical manifestations of certain focal forms of epilepsy.Journal of Neurology Neurosurgery & Psychiatry 09/1971; 34(4):369-87. · 4.92 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Malformations of cortical development (MCD) are responsible for many cases of refractory epilepsy in adults and children. The results of surgical treatment are difficult to assess from the published literature. Judging from the limited number of adequately reported cases, approximately 40% of all cases of MCD treated surgically may be rendered seizure-free over a minimum 2-year follow-up period. This figure is the same for focal cortical dysplasia (FCD), the most common variety of MCD in surgical reports. In comparison with outcome for epilepsy associated with hippocampal sclerosis, this figure is low. Part of the difference may be artificial and related to limited reporting. Much of the difference is likely to relate to the complex underlying biology of MCD. Analysis of epileptogenesis in MCD has been undertaken. Different types of MCD have different sequelae. Some varieties are intrinsically epileptogenic; these include FCD and heterotopia. Although in most cases, the visualized MCD lies within the region of brain responsible for generating seizures (the epileptogenic zone), it may not constitute the entire epileptogenic zone in all cases. For polymicrogyria and schizencephaly in particular, the visualized abnormalities are probably not the most important component of the epileptogenic zone. There is evidence that the epileptogenic zone is spatially distributed and also, in some cases, temporally distributed. These findings may explain poor surgical outcome and the inadequacy of current presurgical evaluative methods. New preoperative techniques offer the opportunity of improved presurgical planning and selection of cases more likely to be rendered seizure-free by current surgical techniques. Of paramount importance is improved reporting. The establishment of a central registry may facilitate this aim. Specific recommendations are made for surgical strategies based on current experience and understanding.Brain 07/2000; 123 ( Pt 6):1075-91. · 9.92 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Malformations of cortical development (MCD) are a recognized cause of epilepsy. Their special significance lies in the fact that, once detected and delineated, they are amenable to surgical removal. However, diagnosis from high-resolution MRI is still difficult, time-consuming, and highly dependent on individual expertise. We have recently proposed a simple procedure to detect cortical dysplasias, using automated procedures available within SPM99 (Wellcome Department, University College London, UK). Here, we aimed to systematically determine the best combination of processing parameters, using an optimized voxel-based morphometry approach. We included 20 patients with a known MCD and compared them to a normal database of 53 healthy, age- and gender-matched controls. The approaches taken during spatial normalization and a number of other parameters were systematically altered in order to find the best combination of parameters. Overall, 99 different approaches were evaluated in different ways. As far as possible, automatic processing and evaluation steps were used. With the number of candidate regions for MCD limited to five per patient, the best approaches resulted in the correct identification of up to 16 of 20 malformations. However, a number of approaches failed to perform well. The reasons for these failures and the implications this has for other studies are discussed. We conclude that voxel-based morphometry is able to detect cortical malformations with a high degree of accuracy. However, specific problems seem to arise when using an optimized protocol for voxel-based morphometry, indicating that this protocol may not be optimal for all voxel-based studies on brain morphology. Our approach, involving systematic alterations of parameters and evaluation, may be useful for other studies.NeuroImage 10/2003; 20(1):330-43. · 6.25 Impact Factor
A level set driven by MR features of focal cortical dysplasia for
?, T. Mansi, N. Bernasconi, V. Naessens, D. Klironomos, A. Bernasconi
McConnell Brain Imaging Center, Montreal Neurological Institute, McGill University, Montreal, Canada
Abstract. Focal cortical dysplasia (FCD), a malformation of cortical development, is an important cause of
medically intractable epilepsy. FCD lesions are difficult to distinguish from non-lesional cortex and their de-
lineation on MRI is a challenging task. This paper presents a method to segment FCD lesions on T1-weighted
MRI, based on a 3D deformable model, implemented using the level set framework. The deformable model is
driven by three MRI features: cortical thickness, relative intensity and gradient. These features correspond to
the visual characteristics of FCD and allow to differentiate lesions from normal tissues. The proposed method
was tested on 18 patients with FCD and its performance was quantitatively evaluated by comparison with the
manual tracings of two trained raters. The validation showed that the similarity between the level set segmen-
tation and the manual labels is similar to the agreement between the two human raters. This new approach may
become a useful tool for the presurgical evaluation of patients with intractable epilepsy.
Malformations of cortical development (MCD) have been increasingly recognized as an important cause of med-
ically intractable focal epilepsy. Focal cortical dysplasia (FCD) , a malformation due to abnormal neuroglial
proliferation, is the most frequent MCD in patients with intractable extra-temporal epilepsy . Epilepsy surgery,
consisting in the removal of the FCD lesion, is an effective treatment for these patients. However, freedom from
seizures after surgery is closely related to the resection of the whole lesion . The precise delineation of lesions
is thus important for surgical planning in epilepsy.
Although magnetic resonance imaging (MRI) has allowed the recognition of FCD in an increased number of
patients, standard radiological evaluation fails to identify lesions in a large number of cases . Moreover, the
spatial extension of the lesions is difficult to define on the MRI. The segmentation of FCD is thus a challenging
image analysis application as the lesions are often subtle, difficult to differentiate from the normal cortex, of
variable size, position and shape, and with ill-defined boundaries. Recently, image analysis techniques have been
developed to detect FCD lesions automatically on MRI, relying on different types of voxel-wise analysis [4,5]. In
particular, computational models of FCD characteristics  and a Bayesian classifier for lesion detection  were
previously proposedby our group. While these approaches successfully identify the FCD in a majority of patients,
they provide a very limited coverage of the lesion (about 20%) and thus cannot be considered as segmentation
This paper presents a method for segmenting focal cortical dysplasia (FCD) lesions on T1-weighted MRI, based
on a level set deformable model driven by MR features of these lesions. This method partly relies on our previous
detection approaches [4,6]. However, our target application is FCD segmentation and not detection. The compu-
tational models of FCD features are used to drive a level set deformable model and the FCD classifier is used only
to obtain a starting point for the segmentation procedure.
Our approach relies on a 3D deformable model, based on the level set method. The level set is guided by a
probabilitymapderivedfromFCD features. ThesefeaturescorrespondtothevisualcharacteristicsofFCD: cortical
thickening, a blurred transition between gray matter (GM) and white matter (WM), and hyperintensesignal within
the dysplastic lesion . Additionally, it is necessary to provide a starting point for the level set evolution. To this
purpose, we made use of our previously developed FCD classifier , under supervision of an expert user.
2.1Probabilistic Modeling of FCD Features
To quantitatively evaluate the visual MR characteristics of FCD, we relied on our previous computational mod-
els . A cortical thickness map, denoted as
??, is computed by solving Laplace’s equation over the cortical
?Corresponding author. email: email@example.com
ribbon. Hyperintensesignal is representedusing a relative intensity index defined as
GM/WM transition is modeled with a gradient magnitude map, denoted as
a vector-valued feature map
???? is the intensity at voxel
?is the boundary intensity between GM and WM. Blurring of the
??. These three characteristics define
???????????????????????? at each point
? in the image space.
We then performed a supervised learning to estimate the probability of different tissue classes in the brain given
the feature vector
cerebro-spinal fluid (CSF) and the FCD lesion (L). Normal tissues were segmented using a histogram-based ap-
proach with automated threshold, while the FCD lesions were painted by trained observers (see Section 3). Con-
using the maximum likelihood on a learning set of patients. The posterior probabilities
tained by Bayes’ rule. As the size of FCD lesions is variable, we assumed equal prior probabilities for the different
classes. Figure 1 presents an example of the three feature maps and of the posterior probability maps in a patient
?. Four different classes, denoted as
?, were considered: gray matter (GM), white matter (WM),
????????? for each class
? were modeled using a trivariate normal distribution and estimated
????????? were then ob-
Figure 1. Probabilistic modeling of FCD features. Upper panels: T1-weighted MRI where the FCD lesion is
indicated by the arrow (A), cortical thickness map (B), relative intensity map (C), gradient map (D). The lesion is
characterized by higher cortical thickness, higher relative intensity and lower gradient. Lower panels: probability
maps of the lesion class (E), GM (F), WM (G) and CSF (H).
2.2Feature-based Level Set
Based on the previous features, the deformable model was designed to separate the lesion from the non-lesional
regions. The regioncompetitionapproachproposedby Zhu and Yuille  is well adaptedto our purpose. It aims at
segmenting an image into several regions by moving the interfaces between them. The evolution of the interfaces
is driven by functions indicating the membership to each region. In our case, these functions can be derived from
the FCD features.
We intended to isolate the FCD lesion from the non-lesional region, which is composed of three different classes
(GM, WM, CSF). However, the boundaries between these three non-lesional classes were of no interest for our
application. Thus, region competition occurred in each point between the lesion class and the most probable
non-lesional class. The membership to the lesional region was defined as
previously computed posterior probability of the lesion class. The non-lesional region was modeled by
????????????? which is the
The feature-based deformable model describes the evolution of the interface (or surface in 3D)
region, according to those membership functions and a regularization term. The motion of a point
? of the lesional
? belonging to
? is defined as:
?is the inward normal to
? at point
? (directed towards the interiorof the lesion),
?is the mean curvature
???? is a feature-based term
? are weighting coefficients. In the previous equation,
?is a regularity term producing a smooth surface. If
???, meaning that the most probable
Figure 2. Results of FCD segmentation: level set segmentation (A), initialization (B), manual tracing
class for point
? is the lesion, the surface
? will be expanded, in order to include this point. On the contrary, if
???, meaning that this point should belong to one of the three non-lesional classes, the surface will
The motion equation was implemented using the level set method . The principle of this method is to define
the classical signed distance to the surface
curve motion to level set evolution , the feature-based level set can be described by:
? as the zero level set of an implicit function
???????????. As an implicit function
?, we chose
?, with negative values in the interior of
?. Using the derivation from
The previous equation was implemented using the numerical scheme proposed in [8, chap.6]. To reduce the
computational complexity, we made use of the narrow-band method .
3Experiments and Results
Subjects and Image Preparation
visible FCD. The Ethics Board of the MNI approvedthe study, and written informedconsent was obtainedfrom all
participants. 3D MR images were acquired on a 1.5T scanner using a T1-fast field echo sequence with an isotropic
voxel size of
dardization , automatic registration into stereotaxic space  and brain extraction . Classification of brain
tissue in GM, WM and CSF was done using an histogram-based method with automated threshold .
We selected 24 patients (13 males, mean age
???? ) with MRI-
?. All images underwent automated correction for intensity non-uniformity and intensity stan-
in 18 (18/24=75%) patients. We assessed the possibility of segmenting the six undetected lesions with a manual
initialization of the procedure. However, the segmentation failed in these cases because their features where not
sufficiently discriminant. The evaluation was thus done on the 18 detected lesions.
The FCD classifier is used to initialize the deformable model. It successfully identified the lesion
The corresponding manual labels are further denoted as
For the 18 manual labels, the mean interrater similarity index was
Lesions were delineated independently on 3D MRI by two trained raters (VN and DK).
?. Interrater agreement was assessed using the
? denote two labels), which is a special case of kappa statistic .
????????? (range=???? to
of a given patient, this patient was excluded from the learning set (Section 2.1). This approach avoids the introduc-
tion of bias in the result. Moreover, we computed the similarity obtained with the FCD classifier to evaluate the
added value of the level set. Results are reported in Table 1. Figures 2 and 3 present the segmentations obtained in
two patients with FCD.
We comparedtheautomatedsegmentationstothesets ofmanuallabelsusingthesimilar-
? presented above. The evaluation was performed using a leave-one-out approach: for the segmentation
Table 1. The table presents the similarity indices for the level set and the FCD classifier with respect to the two
manualtracings,aswellas theinterratersimilarity. Results arereportedasmean?SDwiththerangeinparentheses.
????????? (???? to
????????? (???? to
????????? (???? to
????????? (???? to
????????? (???? to
Figure 3. Results of FCD segmentation. Left panels: level set segmentation (A), initialization (B), manual tracing
?(C), manual tracing
?(D). Right panel: 3D rendering of the FCD lesion segmentation together with the
In this study, we proposed a method for segmenting FCD lesions on MRI. There is no available gold standard for
evaluating the delineation of these lesions. For this reason, we compared the level set segmentation to the manual
tracings of two trained observers. The interrater similarity was
in particular when keeping in mind the difficulty of FCD segmentation. The level set segmentations achieved a
degree of similarity of
The similarities achieved by the level set are also very close to the interrater agreement. A significant portion of
the remaining differences between automated and manual labels is probably due to the interrater variability rather
than to the unability of the level set to recover the full extension of lesions. This can be seen in Figure 3 where the
two raters decided to exclude different parts of the lesion (Panels C and D) while these parts were included in the
automated segmentation (Panel A). Moreover, compared to our previously developed FCD classifier, the present
method achieved a similarity twice as large and therefore constitutes a significant improvement.
???? which correspondsto a substantial agreement,
???? with the two sets of manual labels, which again constitutes a good agreement.
In conclusion, this paper demonstrates the effectiveness of a feature-based level set approach for the segmentation
of FCD lesions. It has the potential to reduce user subjectivity and, more importantly, to unveil lesional areas that
could be overlooked by visual inspection. This new method may become a useful tool for surgical planning in
grant 203707) and by the Scottish Rite Charitable Foundation of Canada. OC is recipient of the Epilepsy Canada
Clinical Sciences Fellowship.
This work was supported by a grant of the Canadian Institutes of Health research (CIHR-
1. D. Taylor, M. Falconer, C. Bruton et al. “Focal dysplasia of the cerebral cortex in epilepsy.” J Neurol Neurosurg Psychiatry
34, pp. 369–387, 1971.
2. S. Sisodiya. “Surgery for malformations of cortical development causing epilepsy.” Brain 123, pp. 1075–1091, 2000.
3. L. Tassi, N. Colombo, R. Garbelli et al. “Focal cortical dysplasia: neuropathological subtypes, EEG, neuroimaging and
surgical outcome.” Brain 125(Pt 8), pp. 1719–1732, 2002.
4. S. Antel, D. Collins, N. Bernasconi et al. “Automated detection of focal cortical dysplasia lesions using computational
models of their MRI characteristics and texture analysis.” NeuroImage 19(4), pp. 1748–59, 2003.
5. M. Wilke, J. Kassubek, S. Ziyeh et al. “Automated detection of gray matter malformations using optimized voxel-based
morphometry: a systematic approach.” NeuroImage 20(1), pp. 330–343, September 2003.
6. S. Antel, A. Bernasconi, N. Bernasconi et al. “Computational models of MRI characteristics of focal cortical dysplasia
improve lesion detection.” NeuroImage 17(4), pp. 1755–60, 2002.
7. S. Zhu & A. Yuille. “Region competition: Unifying snakes, region growing, and bayes/MDL for multiband image seg-
mentation.” IEEE Transactions on Pattern Analysis and Machine Intelligence 18(9), pp. 884–900, 1996.
8. J. Sethian. Level-set methods and fast marching methods. Cambridge University Press, second edition, 1999.
9. J. Sled, A. Zijdenbos & A. Evans. “A nonparametric method for automatic correction of intensity nonuniformity in MRI
data.” IEEE Transaction on Medical Imaging 17(1), pp. 87–97, February 1998.
10. D. Collins, P. Neelin, T. Peters et al. “Automatic 3D intersubject registration of MR volumetric data in standardized
Talairach space.” Journal of Computer-Assisted Tomography 18, pp. 192–205, 1994.
11. S. Smith. “Fast robust automated brain extraction.” Human Brain Mapping 17(3), pp. 143–155, November 2002.
12. A. Zijdenbos, B. Dawant, R. Margolin et al. “Morphometric analysis of white matter lesions in MR images: method and
validation.” IEEE Transactions on Medical Imaging 13(4), pp. 716–724, 1994.