Abstract: In a method for determining if a test data set is anomalous in a deep neural network that has been trained with a plurality of training data sets resulting in back propagated training gradients having statistical measures thereof, the test data set is forward propagated through the deep neural network so as to generate test data intended labels including at least original data, prediction labels, and segmentation maps. The test data intended labels are back propagated through the deep neural network so as to generate a test data back propagated gradient. If the test data back propagated gradient differs from one of the statistical measures of the back propagated training gradients by a predetermined amount, then an indication that the test data set is anomalous is generated. The statistical measures of the back propagated training gradient include a quantity including an average of all the back propagated training gradients.

Abstract: Neural networks and learning algorithms can use a variance of gradients to provide a heuristic understanding of the model. The variance of gradients can be used in active learning techniques to train a neural network. Techniques include receiving a dataset with a vector. The dataset can be annotated and a loss calculated. The loss value can be used to update the neural network through backpropagation. An updated dataset can be used to calculate additional losses. The loss values can be added to a pool of gradients. A variance of gradients can be calculated from the pool of gradient vectors. The variance of gradients can be used to update a neural network.

Abstract: In a method of generating a neural network used to detect a feature of medical significance from a body image data input, test data images (110) are divided into patches (120). Each patch is labelled as either corresponding to the feature (124) or not corresponding to the feature (122). One trained fully connected layer (136) in a pretrained general purpose convolutional neural network (130) is replaced with a new fully connected layer. The pretrained convolutional neural network (130) is retrained with the set of labelled patches (126) to generate a feature-specific convolutional neural network that includes at least one feature-specific fully connected layer (152) that maps the body image data to the feature of medical significance when the feature of medical significance is present in the body image data input.

Abstract: A system for monitoring ocular movement can comprise a housing, a plurality of light sources, at least one imager, and a controller. The housing can define a cavity configured to allow each eye of a patient to view an interior region of the housing. The plurality of light sources can be oriented within the interior region of the housing. The at least one imager can be oriented to capture an image of an eye of a patient during an evaluation. The at least one controller can comprise at least one processor and a non-transitory computer readable medium storing instructions. The instructions can be executed by the processor and cause the controller to receive image data from the at least one imager and illuminate the plurality of light sources in a predetermined and reconfigurable sequence.

Abstract: A system for monitoring ocular movement can comprise a housing, a plurality of light sources, at least one imager, and a controller. The housing can define a cavity configured to allow each eye of a patient to view an interior region of the housing. The plurality of light sources can be oriented within the interior region of the housing. The at least one imager can be oriented to capture an image of an eye of a patient during an evaluation. The at least one controller can comprise at least one processor and a non-transitory computer readable medium storing instructions. The instructions can be executed by the processor and cause the controller to receive image data from the at least one imager and illuminate the plurality of light sources in a predetermined and reconfigurable sequence.