Systems and methods for predicting recovery of a patient

A method executed by a computer system comprising at least one hardware processor, the method comprising:

• storing, in a database, multimodal data of people recovering from a brain injury;
• and
• executing, by the at least one hardware processor, instructions stored in memory that provide analysis of the multimodal data over multiple intervals that span more than two months of time and include:
• executing, by the at least one hardware processor, longitudinal heterogeneous trajectory analysis on the multimodal data;
• executing, by the at least one hardware processor, separation of heterogeneous groups of the people that include good recovery potential for recovery of the brain injury, moderate recovery potential of the brain injury, and poor recovery potential of the brain injury;
• predicting, by the at least one hardware processor and based on the longitudinal heterogeneous trajectory analysis and the separation of the heterogeneous groups, functional recovery of a person in the group from the brain injury;
• and
• outputting a display of a graph comprising plots representing the predicted functional recovery of the person over a sequence of time.

A non-transitory computer readable medium having stored thereon instructions which, when executed by a computer comprising at least one hardware processor, make the computer carry out a method comprising:

• storing, in a database, multimodal data of people recovering from a brain injury;
• executing, by the at least one hardware processor, the instructions that provide analysis of the multimodal data over multiple intervals that span more than two months of time;
• executing, by the at least one hardware processor, longitudinal heterogeneous trajectory analysis on the multimodal data;
• executing, by the at least one hardware processor, separation of heterogeneous groups of the people that include good recovery potential for recovery of the brain injury, moderate recovery potential of the brain injury, and poor recovery potential of the brain injury;
• and
• predicting, by the at least one hardware processor and based on the longitudinal heterogeneous trajectory analysis and the separation of the heterogeneous groups, functional recovery of a person in the group from the brain injury.