Research studies
Update this page with information about current and past research studies.
Research
Epilepsy is a chronic disabling disorder affecting over 1% of the population who suffer higher rates of physical, mental and social disability and associated costs of over $10,000 per year. Neurosurgery can be curative in many cases. Surgery can have cognitive risks, however: up to 41% patients still suffer language decline after temporal lobe surgery1 and 44% suffer verbal memory loss. There is a dearth of validated, reproducible methods to simply and accurately predict patient outcome. My research mission is to develop standardized, evidence-based protocols to accurately predict patients’ real-world functioning after epilepsy surgery.
Language fMRI
Clinical interpretation of language fMRI. To obtain an accurate picture of how clinical fMRI is used and interpreted, we surveyed 82 epilepsy surgical program directors and clinicians. In addition to using fMRI to identify the language dominant hemisphere, 44% reported relying on fMRI to guide surgical margins to preserve language function. Further, at most 21% sought to map other known language regions beyond Broca’s and Wernicke’s areas. Instances of unpredicted language decline were reported by 17%, and 54% reported cases of unexpected preservation of function. None of these cases had been published.
Technical execution of fMRI. Separately, there are no clear technical guidelines covering the tasks, training, and analytic approaches required for presurgical fMRI, and the standards of practice in each of these areas was unknown. We surveyed 63 analysts across epilepsy surgical programs and found heterogeneity in all aspects of fMRI. Multiple variants of over 15 protocols were in use, and neither of the two best-validated protocols were used by more than 10% of programs. Many sites reported completing fMRI without professionals trained in neuropsychology (64%) and radiology (34%)–disciplines with complementary skills fundamental to this method.
These data show that the interpretation and execution of fMRI in the clinic varies from the best available evidence, a fact that will lead to varying accuracy in predicting post-surgical decline. Separately, they suggest that in spite of a lack of evidence, many clinicians already use fMRI maps to guide surgical margins and that the majority only recognize two language areas when they do so. This is significant as at least four further language regions are both identifiable using fMRI and are associated with language deficits when removed or directly stimulated.
Mapping of language cortex beyond Broca’s and Wernicke’s areas. The above data suggests a clinical need for fMRI to identify critical language areas. We defined a new approach to estimate the location of six known language areas, including Broca’s & Wernicke’s areas, Exner’s Area, Supplementary Speech Area, Angular Gyrus, and the Basal Temporal Language Area. This method relies on a set of three different language tasks and a trained clinician selecting and combining data from the tasks to identify these regions. In 22 epilepsy patients, this method was reliable when used by different clinicians (78% overlap) and identified the Wada-defined language-dominant hemisphere with accuracy equivalent to the best published methods (85% of cases).12 Further, activation consistent with all six language regions consistently and more often than an automated analysis of the same data
A reliable, validated, freely-available multilingual battery (ongoing). The above work will have limited impact on patients’ lives if it cannot be reproduced, or is available only in English. Further, the above data showed that over half of surgical programs rely on closed (and often unvalidated) commercial software packages (52%) rather than freely-available published alternatives, and that 68% of US epilepsy programs use unvalidated in-house translations of tasks (most often in Spanish). We have provided the above tasks11 free for download on this site, and in over fifteen different languages (further translations pending). These protocols have been downloaded forty times to date, and have led to developing collaborations with Portuguese, Greek and South African epilepsy programs.
Visual and memory system mapping
In past work we have focused on improving diffusion tractography to avoid vision problems after temporal lobe surgery.
We examined existing methods of optic radiation tractography and showed that many technical limitations can be overcome through accurate knowledge of the visual system’s neuroanatomy. Applying this anatomical framework allowed us to resolve a misunderstanding in the literature – namely, the view that the optic radiations comprise three separate components – and instead to map the radiations as a single sheet more accurately than prior methods.
In my PhD research, I found that remembering simple spatial information activates brain regions on both sides of the brain near the hippocampus, a structure involved in memory.
To study this, I imaged the brain (fMRI) while people viewed simple visual information about where things were located in space and when they appeared. In people without epilepsy, processing spatial relationships consistently activated areas near the hippocampus on both sides of the brain. In patients with epilepsy affecting the left mesial temporal lobe, the brain appeared to reorganize: patients relied more on structures in the opposite hemisphere and on additional brain regions outside the mesial temporal lobe. This suggests the mesial temporal lobe supports fundamental associative processes – such as linking elements across space and time – and that the brain can partially reorganize these networks after injury.
Publications
Publications demonstrating the above, and others, are viewable here: