Introduction: Anion gap metabolic acidosis is a common disorder seen in the emergency department. The differential can include toxicological, renal, endocrine, infectious, and cardiogenic disorders. Ketosis, however, is one of the rarer causes of metabolic acidosis seen by the emergency physician in developed nations. Case Report: A 53-year-old female presented after starting a low-carbohydrate ketogenic diet for weight loss. She reported xerostomia, nausea with abdominal pain and a 17-pound weight loss over the previous 22 days. Labs revealed an anion-gap metabolic acidosis with ketosis. She was treated with 5% dextrose in normal saline and a sliding scale insulin coverage. Her anion gap corrected during her hospital course and was discharged on hospital day three. Discussion: The ketogenic diet typically consists of a high-fat, adequate protein and low carbohydrate diet that has previously been thought to be relatively safe for weight loss. However, when carbohydrates are completely removed from the diet an overproduction of ketones bodies results in ketoacidosis. Treatment should be aimed at halting the ketogenic process and patient education. Conclusion: Although rarely included in the differential for metabolic acidosis, diet-induced ketosis should be included by the emergency physician when faced with a patient who recently changed their eating patterns.

Still image human action recognition (HAR) is a challenging problem owing to limited sources of information and large intra-class and small inter-class variations which requires highly discriminative features. Transfer learning offers the necessary capabilities in producing such features by preserving prior knowledge while learning new representations. However, optimally identifying dynamic numbers of re-trainable layers in the transfer learning process poses a challenge. In this study, we aim to automate the process of optimal configuration identification. Specifically, we propose a novel particle swarm optimisation (PSO) variant, denoted as EnvPSO, for optimal hyper-parameter selection in the transfer learning process with respect to HAR tasks with still images. It incorporates Gaussian fitness surface prediction and exponential search coefficients to overcome stagnation. It optimises the learning rate, batch size, and number of re-trained layers of a pre-trained convolutional neural network (CNN). To overcome bias of single optimised networks, an ensemble model with three optimised CNN streams is introduced. The first and second streams employ raw images and segmentation masks yielded by mask R-CNN as inputs, while the third stream fuses a pair of networks with raw image and saliency maps as inputs, respectively. The final prediction results are obtained by computing the average of class predictions from all three streams. By leveraging differences between learned representations within optimised streams, our ensemble model outperforms counterparts devised by PSO and other state-of-the-art methods for HAR. In addition, evaluated using diverse artificial landscape functions, EnvPSO performs better than other search methods with statistically significant difference in performance.

Neuro-evolution is often used to generate the parameters, topology, and rules of artificial neural networks. This technique allows for automatic configuration of a neural network. In this paper we propose a method to generate Spiking Neural Networks (SNNs) automatically called NENG (Neuro-Evolutionary Network Generation). The aim was to help alleviate the manual construction and optimization of neural network implementations. The results show the algorithm is successful at generating and improving the design of SNNs for a Classification task. After 812 generations with a population size of 20 the algorithm converges to model the Xor gate with 100% accuracy. The results show improvements to the algorithm execution time and number of neurons over time.