Crystallization behavior of milk fat,palm oil,palm kernel oil,and cocoa butter with and without the addition of cannabidiol

The objective of this study was to evaluate the effect of cannabidiol (CBD) on the crystallization behavior and physical properties of various fats. Anhydrous milk fat (AMF), palm oil (PO), palm kernel oil (PKO), and cocoa butter (CB) were chosen for this study, for their unique crystallization behaviors. CBD was added at 1 and 2.5% wt/wt to these fats, and the crystallization behavior was evaluated at 26°C for AMF and PO and at 22°C for PKO and CB. Control samples with no CBD were prepared and evaluated as well. Results show that CBD delayed the crystallization of all fats with the least effect observed for the PO. Slight increases in crystal size were observed with the addition of CBD for all samples. CBD did not affect the melting profile of AMF or CB, but it increased the peak temperature of PO and decreased the enthalpy of PKO. Similarly, hardness was only affected by CBD in PO samples, with harder materials obtained for samples containing 2.5% CBD. The same trend was observed for elasticity. In addition, the elasticity of AMF increased with the addition of CBD but not its hardness. Overall, this study indicates that the effect of CBD on fat crystallization is highly dependent on the type of fat used. Producers of fat-based products that are willing to include CBD in their formulations must carefully control processing conditions to ensure product quality.     

HalfedgeCNN for Native and Flexible Deep Learning on Triangle Meshes

We describe HalfedgeCNN, a collection of modules to build neural networks that operate on triangle meshes. Taking inspiration from the (edge-based) MeshCNN, convolution, pooling, and unpooling layers are consistently defined on the basis of halfedges of the mesh, pairs of oppositely oriented virtual instances of each edge. This provides benefits over alternative definitions on the basis of vertices, edges, or faces. Additional interface layers enable support for feature data associated with such mesh entities in input and output as well. Due to being defined natively on mesh entities and their neighborhoods, lossy resampling or interpolation techniques (to enable the application of operators adopted from image domains) do not need to be employed. The operators have various degrees of freedom that can be exploited to adapt to application-specific needs.