Improving the performance of mutation-based evolving artificial neural networks with self-adaptive mutations.

Neuroevolution is a promising approach for designing artificial neural networks using an evolutionary algorithm. Unlike recent trending methods that rely on gradient-based algorithms, neuroevolution can simultaneously evolve the topology and weights of neural networks. In neuroevolution with topological evolution, handling crossover is challenging because of the competing conventions problem. Mutation-based evolving artificial neural network is an alternative topology and weights neuroevolution approach that omits crossover and uses only mutations for genetic variation. This study enhances the performance of mutation-based evolving artificial neural network in two ways. First, the mutation step size controlling the magnitude of the parameter perturbation is automatically adjusted by a self-adaptive mutation mechanism, enabling a balance between exploration and exploitation during the evolution process. Second, the structural mutation probabilities are automatically adjusted depending on the network size, preventing excessive expansion of the topology. The proposed methods are compared with conventional neuroevolution algorithms using locomotion tasks provided in the OpenAI Gym benchmarks. The results demonstrate that the proposed methods with the self-adaptive mutation mechanism can achieve better performance. In addition, the adjustment of structural mutation probabilities can mitigate topological bloat while maintaining performance.

Subarachnoid haemorrhage as a cause of out-of-hospital cardiac arrest: a prospective computed tomography study.

AIM: Aneurysmal subarachnoid haemorrhage (SAH) is a relatively common cause of out-of-hospital cardiac arrest (OHCA). Early identification of SAH-induced OHCA with the use of brain computed tomography (CT) scan obtained immediately after resuscitation may help emergency physicians make therapeutic decision as quickly as they can. METHODS: During the 4-year observation period, brain CT scan was obtained prospectively in 142 witnessed non-traumatic OHCA survivors who remained haemodynamically stable after resuscitation. Demographics and clinical characteristics of SAH-induced OHCA survivors were compared with those with "negative" CT finding. RESULTS: Brain CT scan was feasible with an average door-to-CT time of 40.0 min. SAH was found in 16.2% of the 142 OHCA survivors. Compared with 116 survivors who were negative for SAH, SAH-induced OHCA survivors were significantly more likely to be female, to have experienced a sudden headache, and trended to have achieved return of spontaneous circulation (ROSC) prior to arrival in the emergency department less frequently. Ventricular fibrillation (VF) was significantly less likely to be seen in SAH-induced than SAH-negative OHCA (OR, 0.06; 95% CI, 0.01-0.46). Similarly, Cardiac Trop-T assay was significantly less likely to be positive in SAH-induced OHCA (OR, 0.08; 95% CI, 0.01-0.61). CONCLUSION: Aneurysmal SAH causes OHCA more frequently than had been believed. Immediate brain CT scan may particularly be useful in excluding SAH-induced OHCA from thrombolytic trial enrollment, for whom the use of thrombolytics is contraindicated. The low VF incidence suggests that VF by itself may not be a common cause of SAH-induced OHCA.