Search

One of the major challenges of modern epileptology is the underutilization of epilepsy surgery for treatment of patients with focal, medication resistant epilepsy (MRE). Aggravating this distressing failure to deliver optimum care to these patients is the underuse of proven localizing tools, such as magnetoencephalography (MEG), a clinically validated, non‐invasive, neurophysiological method used to directly measure and localize brain activity. A sizable mass of published evidence indicates that MEG can improve identification of surgical candidates and guide pre‐surgical planning, increasing the yield of SEEG and improving operative outcomes. However, despite at least 10 common, evidence supported, clinical scenarios in MRE patients where MEG can offer non‐redundant information and improve the pre‐surgical evaluation, it is regularly used by only a minority of USA epilepsy centers. The current state of the art in MEG sensors employs SQUIDs, which require cooling with liquid helium to achieve superconductivity. This sensor technology has undergone significant generational improvement since whole head MEG scanners were introduced around in 1990s, but still has limitations. Further advances in sensor technology which may make ME G more easily accessible and affordable have been eagerly awaited, and development of new techniques should be encouraged. Of late, optically pumped magnetometers (OPMs) have received considerable attention, even prompting some potential acquisitions of new MEG systems to be put on hold, based on a hope that OPMs will usher in a new generation of MEG equipment and procedures. The development of any new clinical test used to guide intracranial EEG monitoring and/or surgical planning must address several specific issues. The goal of this commentary is to recognize the current state of OPM technology and to suggest a framework for it to advance in the clinical realm where it can eventually be deemed clinically valuable to physicians and patients. The American Clinical MEG Society (ACMEGS) strongly supports more advanced and less expensive technology and looks forward to continuing work with researchers to develop new sensors and clinical devices which will improve the experience and outcome for patients, and perhaps extend the role of MEG. However, currently, there are no OPM devices ready for practical clinical use. Based on the engineering obstacles and the clinical tradeoffs to be resolved, the assessment of experts suggests that there will most likely be another decade relying solely on "frozen SQUIDs" in the clinical MEG field. [ABSTRACT FROM AUTHOR]

Epilepsy affects 1 in 150 children under the age of 10 and is the most common chronic pediatric neurological condition; poor seizure control can irreversibly disrupt normal brain development. The present study compared the ability of different machine learning algorithms trained with resting-state functional MRI (rfMRI) latency data to detect epilepsy. Preoperative rfMRI and anatomical MRI scans were obtained for 63 patients with epilepsy and 259 healthy controls. The normal distribution of latency z-scores from the epilepsy and healthy control cohorts were analyzed for overlap in 36 seed regions. In these seed regions, overlap between the study cohorts ranged from 0.44-0.58. Machine learning features were extracted from latency z-score maps using principal component analysis. Extreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and Random Forest algorithms were trained with these features. Area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, specificity and F1-scores were used to evaluate model performance. The XGBoost model outperformed all other models with a test AUC of 0.79, accuracy of 74%, specificity of 73%, and a sensitivity of 77%. The Random Forest model performed comparably to XGBoost across multiple metrics, but it had a test sensitivity of 31%. The SVM model did not perform >70% in any of the test metrics. The XGBoost model had the highest sensitivity and accuracy for the detection of epilepsy. Development of machine learning algorithms trained with rfMRI latency data could provide an adjunctive method for the diagnosis and evaluation of epilepsy with the goal of enabling timely and appropriate care for patients. [ABSTRACT FROM AUTHOR]