Voxel‐based analysis of whole brain FLAIR at 3T detects focal cortical dysplasia

• Published in: Epilepsia (Copenhagen) Vol. 49; no. 5; pp. 786 - 793
• Main Authors: Focke, Niels K., Symms, Mark R., Burdett, Jane L., Duncan, John S.
• Format: Journal Article
• Language: English
• Published: Malden, USA Blackwell Publishing Inc 01.05.2008; Blackwell
• Subjects: Adolescent; Adult; Biological and medical sciences; Brain - pathology; Brain Mapping; Control Groups; Epilepsy, Temporal Lobe - diagnosis; Epilepsy, Temporal Lobe - pathology; Evaluation Studies as Topic; Female; Focal cortical dysplasia; Functional Laterality - physiology; Headache. Facial pains. Syncopes. Epilepsia. Intracranial hypertension. Brain oedema. Cerebral palsy; Humans; Image Interpretation, Computer-Assisted; Image Processing, Computer-Assisted - methods; Image Processing, Computer-Assisted - statistics & numerical data; Imaging, Three-Dimensional - statistics & numerical data; Linear Models; Magnetic Resonance Imaging - methods; Magnetic Resonance Imaging - statistics & numerical data; Male; Malformations of Cortical Development - diagnosis; Malformations of Cortical Development - pathology; Malformations of the nervous system; Medical sciences; Middle Aged; MRI; Nervous system (semeiology, syndromes); Neurology; T2 FLAIR; Voxel based; Dysplasia; Nervous system diseases; Focal cortical dysplasia; MRI; Epilepsy; Central nervous system; Nuclear magnetic resonance imaging; Cerebral disorder; Encephalon; T2 FLAIR; Voxel; Central nervous system disease; Voxel based
• ISSN: 0013-9580; 1528-1167
• DOI: 10.1111/j.1528-1167.2007.01474.x

Summary Background: Focal Cortical Dysplasia (FCD) is an important cause for pharmacoresistant epilepsy that can be treated surgically. The identification of the abnormal cortex on standard MRI can be difficult and computational techniques have been developed to increase sensitivity. In this study we evaluate the potential of a novel whole‐brain voxel‐based technique using normalized FLAIR signal intensity (nFSI) at 3 Tesla. Methods: Twenty‐five patients with neuroradiologically reported FCD were included and compared to 25 healthy control subjects using Statistical Parametric Mapping (SPM5). T2 FLAIR scans were intensity normalized and each individual patient was compared against the control group. Each control subject was compared against the remaining control group. Results: SPM correctly identified the FCD in 88% of cases (22/25) with only one false positive finding in the control group. In all but one of these cases the FCD was the most significant finding in the whole brain. All three missing cases could be detected at lower threshold levels but this would give rise to more false positive findings and thus reduce specificity. Conclusion: We present a novel technique that uses standard clinical T2 FLAIR scans to automatically detect FCDs. It can give supplementary information to the established T1‐based automated techniques and could be useful for additional screening test, to complement the visual reading and clinical interpretation of MRI scans.