Abstract: Chemical formulations for chemical products can be represented by digital formulation graphs for use in machine learning models. The digital formulation graphs can be input to graph-based algorithms such as graph neural networks to produce a feature vector, which is a denser description of the chemical product than the digital formulation graph. The feature vector can be input to a supervised machine learning model to predict one or more attribute values of the chemical product that would be produced by the formulation without actually having to go through the production process. The feature vector can be input to an unsupervised machine learning model trained to compare chemical products based on feature vectors of the chemical products. The unsupervised machine learning model can recommend a substitute chemical product based on the comparison.

Abstract: Chemical formulations for chemical products can be represented by digital formulation graphs for use in machine learning models. The digital formulation graphs can be input to graph-based algorithms such as graph neural networks to produce a feature vector, which is a denser description of the chemical product than the digital formulation graph. The feature vector can be input to a supervised machine learning model to predict one or more attribute values of the chemical product that would be produced by the formulation without actually having to go through the production process. The feature vector can be input to an unsupervised machine learning model trained to compare chemical products based on feature vectors of the chemical products. The unsupervised machine learning model can recommend a substitute chemical product based on the comparison.

Abstract: Chemical formulations for chemical products can be represented by digital formulation graphs for use in machine learning models. The digital formulation graphs can be input to graph-based algorithms such as graph neural networks to produce a feature vector, which is a denser description of the chemical product than the digital formulation graph. The feature vector can be input to a supervised machine learning model to predict one or more attribute values of the chemical product that would be produced by the formulation without actually having to go through the production process. The feature vector can be input to an unsupervised machine learning model trained to compare chemical products based on feature vectors of the chemical products. The unsupervised machine learning model can recommend a substitute chemical product based on the comparison.

Abstract: A vehicle having a first configuration for road use and a second configuration for air use, comprising: road wheels; a traction drive propulsion unit for driving road wheels when in contact with the ground; a thrust propulsion unit for driving a propeller to drive the vehicle through the air; and wherein, when the vehicle is configurable into any of: the first configuration, in which the traction drive propulsion unit is engaged to drive the road wheels on the ground, and the thrust propulsion unit is disengaged from the propeller; the second configuration, in which the traction drive propulsion unit is disengaged such that no drive is provided to the road wheels, and the thrust propulsion unit is engaged with the propeller to drive the vehicle thorough the air; or an intermediate configuration in which the traction drive propulsion unit is engaged to drive the road wheels on the ground, and the thrust propulsion unit is engaged with the propeller to drive the vehicle through the air.

Abstract: A bromination process includes contacting fly ash with liquid bromine to increase the mercury adsorbing ability of the fly ash. The resultant brominated fly ash can be used to adsorb mercury in a high temperature combustion gas.

Abstract: A bromination process includes contacting fly ash with liquid bromine to increase the mercury adsorbing ability of the fly ash. The resultant brominated fly ash can be used to adsorb mercury in a high temperature combustion gas.

Abstract: A support catheter that can puncture a heart septum with a guidewire without buckling and that can access difficult to reach arteries such as carotid or renal arteries from the aorta under difficult circumstances is provided. The support catheter comprises a catheter body having a proximal end, a distal end, and an interior wall surface defining a first lumen extending from the proximal end toward the distal end. A second lumen is disposed between the interior wall surface and an outer surface of the catheter body. A plurality of stiffening members is disposed around the second lumen, the plurality of stiffening members being engageable to selectively stiffen the catheter body. In another configuration, the catheter further comprises a second plurality of stiffening members disposed between the first lumen and the second lumen.