Novel 3D video action recognition deep learning approach for near real time epileptic seizure classification.

Seizure semiology is a well-established method to classify epileptic seizure types, but requires a significant amount of resources as long-term Video-EEG monitoring needs to be visually analyzed. Therefore, computer vision based diagnosis support tools are a promising approach. In this article, we utilize infrared (IR) and depth (3D) videos to show the feasibility of a 24/7 novel object and action recognition based deep learning (DL) monitoring system to differentiate between epileptic seizures in frontal lobe epilepsy (FLE), temporal lobe epilepsy (TLE) and non-epileptic events. Based on the largest 3Dvideo-EEG database in the world (115 seizures/+680,000 video-frames/427GB), we achieved a promising cross-subject validation f1-score of 0.833±0.061 for the 2 class (FLE vs. TLE) and 0.763 ± 0.083 for the 3 class (FLE vs. TLE vs. non-epileptic) case, from 2 s samples, with an automated semi-specialized depth (Acc.95.65%) and Mask R-CNN (Acc.96.52%) based cropping pipeline to pre-process the videos, enabling a near-real-time seizure type detection and classification tool. Our results demonstrate the feasibility of our novel DL approach to support 24/7 epilepsy monitoring, outperforming all previously published methods.

Case Report: Minimal Neurological Deficit of Two Adult Patients With Weston-Hurst Syndrome Due to Early Craniectomy: Case Series and Review of Literature on Craniectomy.

Objectives: We describe two new cases of acute hemorrhagic leucoencephalitis (AHLE), who survived with minimal sequelae due to early measures against increased intracranial pressure, particularly craniotomy. The recently published literature review on treatment and outcome of AHLE was further examined for the effect of craniotomy. Methods: We present two cases from our institution. The outcome of 44 cases from the literature was defined either as good (no deficit, minimal deficit/no daily help) or poor outcome (severe deficit/disabled, death). Patients with purely infratentorial lesions (n = 9) were excluded. Fisher's exact test was applied. Results: Two cases are presented: A 43-year-old woman with rapidly progressive aphasia and right hemiparesis due to a huge left frontal white matter lesion with rim contrast enhancement. Pathology was consistent with AHLE. The second case was a 56-year-old woman with rapidly progressive aphasia and right hemiparesis. Cranial MRI showed a huge left temporo-occipital white matter lesion with typical morphology for AHLE. Both patients received craniotomy within the first 24 h and consequent immunosuppressive-immunomodulatory treatment and survived with minimal deficits. Out of 35 supratentorial reported AHLE cases, seven patients received decompressive craniotomy. Comparing all supratentorial cases, patients who received craniotomy were more likely to have a good outcome (71 vs. 29%). Conclusion: Due to early control of the intracranial pressure, particularly due to early craniotomy; diagnosis per biopsy; and immediate start of immunosuppressive-immunomodulatory therapies (cortisone pulse, plasma exchanges), both patients survived with minimal sequelae. Craniotomy plays an important role and should be considered early on in patients with probable AHLE.

Stenotrophomonas maltophilia brain abscesses after implantation of motor cortex stimulator.

We describe the first case of a patient with brain abscesses caused by Stenotrophomonas maltophilia as a complication after motor cortex stimulator implantation. Brain abscesses pose a challenge in diagnosis and treatment, because microbiological diagnosis is not always achieved, antibiotic drugs may not penetrate well into the CNS and some bacteria have resistances to typical empirical antibiotic drugs. In this case diagnosis was only made after removal of the stimulator and a long term treatment with antibiotic drugs was necessary. As neurostimulation devices become more common, formerly rare bacteria may become a more common complication. Bacteria with biofilm properties and a problematic resistance spectrum like Stenotrophomonas maltophilia should be included in the differential diagnosis, because they will not respond to the typical empirical treatment.