Mechanical and Computational Fluid Dynamic Models for Magnesium-Based Implants.

Today, mechanical properties and fluid flow dynamic analysis are considered to be two of the most important steps in implant design for bone tissue engineering. The mechanical behavior is characterized by Young's modulus, which must have a value close to that of the human bone, while from the fluid dynamics point of view, the implant permeability and wall shear stress are two parameters directly linked to cell growth, adhesion, and proliferation. In this study, we proposed two simple geometries with a three-dimensional pore network dedicated to a manufacturing route based on a titanium wire waving procedure used as an intermediary step for Mg-based implant fabrication. Implant deformation under different static loads, von Mises stresses, and safety factors were investigated using finite element analysis. The implant permeability was computed based on Darcy's law following computational fluid dynamic simulations and, based on the pressure drop, was numerically estimated. It was concluded that both models exhibited a permeability close to the human trabecular bone and reduced wall shear stresses within the biological range. As a general finding, the proposed geometries could be useful in orthopedics for bone defect treatment based on numerical analyses because they mimic the trabecular bone properties.

Towards global flood mapping onboard low cost satellites with machine learning.

Spaceborne Earth observation is a key technology for flood response, offering valuable information to decision makers on the ground. Very large constellations of small, nano satellites- 'CubeSats' are a promising solution to reduce revisit time in disaster areas from days to hours. However, data transmission to ground receivers is limited by constraints on power and bandwidth of CubeSats. Onboard processing offers a solution to decrease the amount of data to transmit by reducing large sensor images to smaller data products. The ESA's recent PhiSat-1 mission aims to facilitate the demonstration of this concept, providing the hardware capability to perform onboard processing by including a power-constrained machine learning accelerator and the software to run custom applications. This work demonstrates a flood segmentation algorithm that produces flood masks to be transmitted instead of the raw images, while running efficiently on the accelerator aboard the PhiSat-1. Our models are trained on WorldFloods: a newly compiled dataset of 119 globally verified flooding events from disaster response organizations, which we make available in a common format. We test the system on independent locations, demonstrating that it produces fast and accurate segmentation masks on the hardware accelerator, acting as a proof of concept for this approach.