Abstract: A method of estimating the confidence in the output of a neural network. The method comprises providing data input to a neural network, splitting the data into two sets and training the neural network with weight and bias parameters. Plural decision vectors and plural weight vectors are paired to provide a like plurality of angle distributions. Labelled class parameters and predicted class parameters are computed and fitted to a parametric function. The parametric function is used to compute distribution parameters, which are used to compute probabilities that the values from the distribution parameters are correct. These probabilities can then be used to make a risk informed decision.

Abstract: A method of estimating the confidence in the output of a neural network. The method comprises providing data input to a neural network, splitting the data into two sets and training the neural network with weight and bias parameters. Plural decision vectors and plural weight vectors are paired to provide a like plurality of angle distributions. Labelled class parameters and predicted class parameters are computed and fitted to a parametric function. The parametric function is used to compute distribution parameters, which are used to compute probabilities that the values from the distribution parameters are correct. These probabilities can then be used to make a risk informed decision.

Abstract: Disclosed are compositions comprising a first liposome, comprising at least one lipid selected from the group consisting of saturated phospholipids, unsaturated phospholipids, mixed phospholipids, and cholesterol, and at least one vasodilator; and a second liposome, comprising at least one lipid selected from the group consisting of saturated phospholipids, unsaturated phospholipids, mixed phospholipids, and cholesterol, at least one thiazolidinedione, and at least one compound having an affinity for at least one component of an atheroma. Also disclosed are methods for forming the compositions and methods of using the compositions to ameliorate atheroma.

Abstract: Disclosed are compositions comprising a first liposome, comprising at least one lipid selected from the group consisting of saturated phospholipids, unsaturated phospholipids, mixed phospholipids, and cholesterol, and at least one vasodilator; and a second liposome, comprising at least one lipid selected from the group consisting of saturated phospholipids, unsaturated phospholipids, mixed phospholipids, and cholesterol, at least one thiazolidinedione, and at least one compound having an affinity for at least one component of an atheroma. Also disclosed are methods for forming the compositions and methods of using the compositions to ameliorate atheroma.

Abstract: The present invention provides gas-containing liposomes. In particular, the present invention provide methods of generating gas-containing liposomes where the gas is introduced under pressure, as well as gas-containing liposomes which contain a large volume of gas (e.g., 10 ul of gas per 5 mg of gas-containing liposomes). In certain embodiments, the gas-containing liposomes contain nitric oxide gas. In some embodiments, such nitric oxide containing liposomes are used to treat a medical condition that is treatable by nitric oxide gas (e.g., intimal hyperplasia).

Abstract: The present invention provides gas-containing liposomes. In particular, the present invention provide methods of generating gas-containing liposomes where the gas is introduced under pressure, as well as gas-containing liposomes which contain a large volume of gas (e.g., 10 ul of gas per 5 mg of gas-containing liposomes). In certain embodiments, the gas-containing liposomes contain nitric oxide gas. In some embodiments, such nitric oxide containing liposomes are used to treat a medical condition that is treatable by nitric oxide gas (e.g., intimal hyperplasia).