How to teach neural networks to mesh: Application on 2-D simplicial contours.

A machine learning meshing scheme for the generation of 2-D simplicial meshes is proposed based on the predictions of neural networks. The data extracted from meshed contours are utilized to train neural networks which are used to approximate the number of vertices to be inserted inside the contour cavity, their location, and connectivity. The accuracy of the scheme is evaluated by comparing the quality of the mesh generated by the neural networks with that generated by a reference mesher. Based on an element quality metric, after conducting tests on contours for a various number of edges, the results show a maximum average deviation of 15.2% on the mean quality and 27.3% on the minimum quality between the elements of the meshes generated by the scheme and the ones generated from the reference mesher; the scheme is able to produce good quality meshes that are suitable for meshing purposes. The meshing scheme is also applied to generate larger scale meshes with a recursive implementation. The findings encourage the adaption of the scheme for 3-D mesh generation.

An original valveless artificial heart providing pulsatile flow tested in mock circulatory loops.

PURPOSE: We present the test bench results of a valveless total artificial heart that is potentially compatible with the pediatric population. METHODS: The RollingHeart is a valveless volumetric pump generating pulsatile flow. It consists of a single spherical cavity divided into 4 chambers by 2 rotating disks. The combined rotations of both disks produce changes in the volumes of the 4 cavities (suction and ejection). The blood enters/exits the spherical cavity through 4 openings that are symmetrical to the fixed rotation axis of the first disk.Mock circulatory system: The device pumps a 37% glycerin solution through 2 parallel circuits, simulating the pulmonary and systemic circulations. Flow rates are acquired with a magnetic inductive flowmeter, while pressure sensors collect pressure in the left and right outflow and inflow tracts.In vitro test protocol: The pump is run at speeds ranging from 20 to 180 ejections per minute. The waveform of the pressure generated at the inflow and outflow of the 4 chambers and the flow rate in the systemic circulation are measured. RESULTS: At an ejection rate of 178 min-1, the RollingHeart pumps 5.3 L/min for a systemic maximal pressure gradient of 174 mmHg and a pulmonary maximal pressure gradient of 75 mmHg. The power input was 14 W, corresponding to an efficiency of 21%. CONCLUSIONS: The RollingHeart represents a new approach in the domain of total artificial heart. This preliminary study endorses the feasibility of a single valveless device acting as a total artificial heart.