Computer-aid system using machine learning for epileptic focus detection is having a promising future. In this study, we report some initial results on machine learning based automatic identification of epileptic focus from intracranial EEG (iEEG) data. We analyzed 90 segments from 30 minutes of iEEG from four surgical patients with focal cortical dysplasia. Frequency band were divided into δ, θ, α, β, γ, ripple, and fast ripple. Subsequently, the prominent entropies were used and calculated mutual information value from the following; Appropriate, Sample, Permutation, Shannon, Renyi, Tsallis, and phase 1 & 2. Finally, support vector machine was used for classifying the epileptic from non-epileptic foci. The accuracy of our method was evaluated by 10-fold cross validation. Appropriate and sample entropies were proper to distinguish the epileptic electrodes. Automated detection using our method resulted higher AUC from 0.63 to 0.83. Machine learning using the multiband entropy-based feature-extraction method is useful for epileptic focus detection.

Recent studies of the neural network for language function have revealed the existence of various networks such as superior longitudinal fasciculus and inferior frontooccipital fasciculus in addition to arcuate fibers connecting the anterior language area (Broca’s area) and the posterior language area (Wernicke’s area). In particular, the posterior language area has many fiber connections in the frontal, parietal, and temporal lobes. Previous studies using magnetoencephalography (MEG) have shown that beta and low gamma activities could be detected in the anterior language area in the left inferior frontal gyrus during silent reading and word recall tasks. However, detecting activity in the posterior language area has not been fully elucidated. Therefore, in this study, we investigated the imaginary coherence (IC) of the functional area with a whole-head MEG system during silent reading or word recall. In the silent reading tasks, the anterior part of the left frontal lobe and posterior part of the left temporal lobe showed high connectivity with the anterior language area, especially in the beta band and the low gamma bands. In the high gamma band, the connection with the motor area was strengthened. Areas with high connectivity with the posterior language area were found in the frontal lobe and the temporal lobe in the all bands. However, the relevant areas within these lobes were wide, and the specificity was unclear. In the word recall task, an area with high connectivity with the anterior language area was confirmed especially in the posterior superior temporal gyrus in the low gamma band. Areas with high connectivity with the posterior language area were found in surrounding areas and nonspecific parts of the frontal lobe in various bands. Though differences in language function networks between the anterior and posterior language areas were common, various patterns were revealed in each subject. This implies the importance of individual analysis in clinical evaluations. Imaginary coherence may be an effective way to detect activity in the posterior language area during silent reading and word recall tasks. Furthermore, the ease of performing the silent reading task makes it clinically applicable.

We describe an abnormal visual evoked potential (VEP) after the remission of childhood absence epilepsy in a patient with Alice in Wonderland syndrome (AIWS). A 10-year-old girl presented with abnormal visual symptoms (micropsia and macropsia). Although the VEP latencies were normal, the VEP amplitude showed significant high amplitudes of both N75–P100 and P100–N145 on Oz–Fz. Moreover, SPECT showed low cerebral blood flow in the right parietal area. These results may have been caused by cortical hyperexcitability due to the patient’s history of epilepsy. The pathophysiology of AIWS may be easy to assess with simultaneous VEP and SPECT.

The impairment of excitation-contraction (E-C) coupling have been recognized as a possible mechanism of ICU-acquired weakness (ICUAW). Although it has been reported that an E-C coupling impairment in animal models of ICUAW and muscle biopsy pathology in ICUAW patients, there has been few evidence in vivo. We developed a novel method to measure E-C coupling using accelerometer in vivo. The tibial nerve was stimulated at the popliteal fossa using a supramaximal rectangular pulse to record compound muscle action potential (CMAP) from soleus muscles. Simultaneously, movement-related potential (MRP) was recorded by accelerometer attached at the base of hallux. The E-C coupling time (ECCT) was calculated by differences between soleus CMAP and MRP. The amplitude of MRP corresponded to maximal twitch force of the soleus muscle. We found some patients revealed prolonged ECCTs and decreased MRP amplitudes without changes of CMAP templates in very early stages of ICUAW. These results indicated that the impairment of E-C coupling might occur in very early stages of ICUAW. Furthermore, the later stages may include additional mechanisms involving the complex pathologies of ICUAW.