Predicting states of subjects

A system for predicting one or more states of a subject, the system comprising:

• a training data obtaining module configured to obtain training data comprising a plurality of variables;
• a training states obtaining module configured to obtain one or more training states associated with the training data;
• and
• a predictive model forming module configured to form a predictive model according to the training data and the one or more training states, the predictive model being predictive of the one or more training states, wherein the forming of the predictive model comprises extracting one or more hidden components from the training data, wherein the extracting of the one or more hidden components comprises regression analysis comprising determining one or more relationships between the one or more hidden components and the plurality of variables, and determining one or more relationships between the one or more hidden components and the one or more training states, and wherein a number of the one or more hidden components is less than a number of the plurality of variables and greater than a number of the one or more training states;
• wherein the one or more training states comprise one or more predicted brain states of the subject comprising at least one of:
• a mental state, a cognitive state, a disease state, a response to a stimulus, a response to a physical task, a response to a mental task, a response to an emotion, and a response to a sensatory input;
• wherein the extracting of the one or more hidden components comprises reducing a regularized total loss comprising a regularized loss of a first matrix comprising the training data and a regularized loss of a second matrix comprising the one or more training states;
• wherein one or more of the obtaining of the training data, the obtaining of the one or more training states, and the forming of the predictive model are implemented as instruction code executed on a processor device.

An apparatus for predicting one or more states of a subject, the apparatus comprising:

• a memory;
• and
• a processor coupled to the memory and configured to:
• obtain training data comprising a plurality of variables;
• obtain one or more training states associated with the training data;
• and
• form a predictive model according to the training data and the one or more training states, the predictive model being predictive of the one or more training states, wherein the forming of the predictive model comprises extracting one or more hidden components from the training data, wherein the extracting of the one or more hidden components comprises regression analysis comprising determining one or more relationships between the one or more hidden components and the plurality of variables, and determining one or more relationships between the one or more hidden components and the one or more training states, and wherein a number of the one or more hidden components is less than a number of the plurality of variables and greater than a number of the one or more training states;
• wherein the one or more training states comprise one or more predicted brain states of the subject comprising at least one of:
• a mental state, a cognitive state, a disease state, a response to a stimulus, a response to a physical task, a response to a mental task, a response to an emotion, and a response to a sensatory input;
• wherein the extracting of the one or more hidden components comprises reducing a regularized total loss comprising a regularized loss of a first matrix comprising the training data and a regularized loss of a second matrix comprising the one or more training states.

An article of manufacture for predicting one or more states of a subject, the article of manufacture comprising a non-transitory computer readable storage medium tangibly embodying a computer readable program code which, when executed, causes the computer to:

• obtain training data comprising a plurality of variables;
• obtain one or more training states associated with the training data;
• and
• form a predictive model according to the training data and the one or more training states, the predictive model being predictive of the one or more training states, wherein the forming of the predictive model comprises extracting one or more hidden components from the training data, wherein the extracting of the one or more hidden components comprises regression analysis comprising determining one or more relationships between the one or more hidden components and the plurality of variables, and determining one or more relationships between the one or more hidden components and the one or more training states, and wherein a number of the one or more hidden components is less than a number of the plurality of variables and greater than a number of the one or more training states;
• wherein the one or more training states comprise one or more predicted brain states of the subject comprising at least one of:
• a mental state, a cognitive state, a disease state, a response to a stimulus, a response to a physical task, a response to a mental task, a response to an emotion, and a response to a sensatory input;
• wherein the extracting of the one or more hidden components comprises reducing a regularized total loss comprising a regularized loss of a first matrix comprising the training data and a regularized loss of a second matrix comprising the one or more training states.