Background and motivation: Combining Deep Reinforcement Learning (Deep RL) and Health Systems Simulations has significant potential, for both research into improving Deep RL performance and safety, and in operational practice. While individual toolkits exist for Deep RL and Health Systems Simulations, no framework to integrate the two has been established. Aim: Provide a framework for integrating Deep RL Networks with Health System Simulations, and to ensure this framework is compatible with Deep RL agents that have been developed and tested using OpenAI Gym. Methods: We developed our framework based on the OpenAI Gym framework, and demonstrate its use on a simple hospital bed capacity model. We built the Deep RL agents using PyTorch, and the Hospital Simulation using SimPy. Results: We demonstrate example models using a Double Deep Q Network or a Duelling Double Deep Q Network as the Deep RL agent. Conclusion: SimPy may be used to create Health System Simulations that are compatible with agents developed and tested on OpenAI Gym environments. GitHub repository of code: https://github.com/MichaelAllen1966/learninghospital

Background & Motivation: Stroke outcomes following revascularization therapy (intravenous thrombolysis, IVT, and/or mechanical thrombectomy, MT) depend critically on time from stroke onset to treatment. Different service configurations may prioritise time to IVT or time to MT. In order to evaluate alternative acute stroke care configurations, it is necessary to estimate clinical outcomes given differing times to IVT and MT. Method: Model using an algorithm coded in Python. This is available at https://github.com/MichaelAllen1966/stroke_outcome_algorithm. Results: We demonstrate how the code may be used to estimate clinical outcomes given varying times to IVT and MT. Conclusion: Python code has been developed and shared to enable estimation of clinical outcome given times to IVT and MT. Here we share pseudocode and links to full Python code.

Additive manufacturing (AM) is an emerged layer-by-layer manufacturing process. However, its broad adoption is still hindered by limited material options, different fabrication defects, and inconsistent part quality. Material extrusion (ME) is one of the most widely used AM technologies, and, hence, is adopted in this research. Low-cost metal ME is a new and AM technology used to fabricate metal composite parts using sintered metal infused filament material. Since the involved materials and process are relatively new, there is a need to investigate the dimensional accuracy of ME fabricated metal parts for real-world applications. Each step of the manufacturing process, from the material extrusion to sintering, might significantly affect the dimensional accuracy. This research provides a comprehensive analysis of dimensional changes of metal samples fabricated by the ME and sintering process, using statistical and machine learning algorithms. Machine learning (ML) methods can be used to assist researchers in sophisticated pre-manufacturing planning and product quality assessment and control. This study compares linear regression to neural networks in assessing and predicting the dimensional changes of ME made components after 3D printing and sintering process. The prediction outcomes using a neural network performed the best with the highest accuracy as compared to regression. The findings of this study can help researchers and engineers to predict the dimensional variations and optimize the printing and sintering process parameters to obtain high quality metal parts fabricated by the low-cost ME process.

We propose that the high energy Cosmic Ray particles around the spectral turn-down commonly called the {\it knee} and up to the upturn, commonly called the {\it ankle}, mostly come from Blue Super Giant star explosions. At the upturn, i.e. the {\it ankle}, Cosmic Rays probably switch to another source class, most likely extragalactic sources. To show this we recently compiled a set of Radio Supernova data to list the magnetic field, shock speed and radius scale (Biermann et al. 2018) \cite{Biermann18}. Using particle acceleration theory at shocks, those numbers can be transformed into characteristic {\it knee} and {\it ankle} energies. Without adjusting any free parameters both of these observed energies are directly indicated by the supernova data. We now proceed to the next step in the argument, and use the Supernova Remnant data of the starburst galaxy M82. Assuming that they are Blue Supergiant star explosions, the shock will race to their outer edge with a magnetic field that follows ${B (r) \, r \, \sim \, const.}$. We argue that the shock runs through the entire magnetic plasma wind region at full speed all the way out to the wind-shell, which is of order parsec scale. The speed is observed to be $\sim \, 0.1 \, c$ at about ${10^{16} \, {\rm cm}}$ radius in the plasma wind. This demonstrates how Blue Supergiant star explosions can provide the Cosmic Ray particles across the {\it knee} and up to the {\it ankle} energy range. The data from the CREAM (Cosmic Ray Energetics and Mass Experiment) mission will test this cosmic ray concept which is reasonably well grounded in two independent radio supernova data sets. The next step in developing our understanding is to obtain accurate Cosmic Ray data near to the {\it knee}, and use unstable isotopes of Cosmic Ray nuclei at high energy to probe the "piston" driving the explosion. We plan to combine these data with the physics of the budding black hole which is probably forming in each of these stars to learn more.