Search

This paper presents a new methodology for automatically learning an optimal neurostimulation strategy for the treatment of epilepsy. The technical challenge is to automatically modulate neurostimulation parameters, as a function of the observed EEG signal, so as to minimize the frequency and duration of seizures. The methodology leverages recent techniques from the machine learning literature, in particular the reinforcement learning paradigm, to formalize this optimization problem. We present an algorithm which is able to automatically learn an adaptive neurostimulation strategy directly from labeled training data acquired from animal brain tissues. Our results suggest that this methodology can be used to automatically find a stimulation strategy which effectively reduces the incidence of seizures, while also minimizing the amount of stimulation applied. This work highlights the crucial role that modern machine learning techniques can play in the optimization of treatment strategies for patients with chronic disorders such as epilepsy.

Objective: To analyze early and mid-term outcome of endovascular treatment in patients with iliac aneurysms, comparing the results of hypogastric revascularization by branch endografting with those of hypogastric occlusion., Methods: Consecutive patients with iliac aneurysms receiving side branch endograft (Group I) were compared with those receiving endograft with hypogastric exclusion (Group II) during the interval from January 2000 to May 2008. Procedural details and outcomes were prospectively collected and were analyzed at one year to avoid mismatch in follow-up length., Results: A total of 74 patients (mean age, 75.8 years, 95% males) were treated: 32 in Group I and 42 in Group II. No differences in baseline risk factors and aneurysm diameter (40.2 +/- 7.9 mm in Group I vs. 38.4 +/- 10.8 in Group II) were found. Concurrent treatment of aortic aneurysm was performed in 25/32 (78%) of Group I and 36/42 (86%) of Group II. Fluoro time was 48 minutes (interquartile range [IQR] 31-57) in Group I vs. 31 minutes (IQR 23-38) in Group II (P = .04). The amount of contrast was similar in both Groups: 184 ml (IQR 155-210) in Group I vs. 183 ml (IQR 155-200) in Group II. No intestinal ischemia or deaths occurred. There were no significant differences in failures of hypogastric side branch deployment (2/32) compared with hypogastric coiling (3/42). Limb occlusions all occurring in the external iliac artery side were 2/32 in Group I vs. 3/42 in Group II. Reintervention rates were similar (5/32 vs. 4/42) at one year. Shrinkage of 5 mm or more was detected in 7/23 (30%) of Group I and in 13/37 (34%) of Group II. Iliac endoleak was present in eight patients (19%) in Group II and in one patient in Group I (4%) (P = .1). Similarly, buttock claudication or impotence were more frequent after hypogastric exclusion, recorded in eight patients in Group II and in one patient in Group I (P = .1)., Conclusions: Endovascular treatment of iliac aneurysm with hypogastric revascularization through side branched endografts is feasible and safe in the mid-term. When compared with hypogastric embolization, this option leads to similar technical success and reintervention rates. Endoleak and buttock claudication occur frequently in patients with iliac aneurysm treated with hypogastric exclusion, while are uncommon in those with hypogastric revascularization. Side branch endografting for iliac aneurysm may be considered a primary choice in younger, active patients with suitable anatomy, but larger studies and longer postoperative observation periods are needed.

Purpose: The anterior cingulate cortex (ACC)--which plays a role in pain, emotions and behavior--can generate epileptic seizures. To date, little is known on the neuronal mechanisms leading to epileptiform synchronization in this structure. Therefore, we investigated the role of excitatory and inhibitory synaptic transmission in epileptiform activity in this cortical area. In addition, since the ACC presents with a high density of opioid receptors, we studied the effect of opioid agonism on epileptiform synchronization in this brain region., Methods: We used field and intracellular recordings in conjunction with pharmacological manipulations to characterize the epileptiform activity generated by the rat ACC in a brain slice preparation., Results: Bath-application of the convulsant 4-aminopyridine (4AP, 50 microM) induced both brief and prolonged periods of epileptiform synchronization resembling interictal- and ictal-like discharges, respectively. Interictal events could occur more frequently before the onset of ictal activity that was contributed by N-methyl-D-aspartate (NMDA) receptors. Mu-opioid receptor activation abolished 4AP-induced ictal events and markedly reduced the occurrence of the pharmacologically isolated GABAergic synchronous potentials. Ictal discharges were replaced by interictal events during GABAergic antagonism; this GABA-independent activity was influenced by subsequent mu-opioid agonist application., Conclusions: Our results indicate that both glutamatergic and GABAergic signaling contribute to epileptiform synchronization leading to the generation of electrographic ictal events in the ACC. In addition, mu-opioid receptors appear to modulate both excitatory and inhibitory mechanisms, thus influencing epileptiform synchronization in the ACC.