| CPC G16H 80/00 (2018.01) [G16H 20/00 (2018.01)] | 19 Claims |

|
1. A computer-implemented method for facilitating personalized digital therapy, the method comprising:
providing, by at least one processor, to a first device associated with a therapist and to a second device associated with a patient, remote access to a digital therapy platform that enables digital therapy to be delivered to the patient, the therapist being assigned to a patient profile of the patient;
automatically collecting, by the at least one processor, patient data from a plurality of data sources associated with the patient profile of the patient in the digital therapy platform;
processing, by the at least one processor, at least a subset of the patient data to detect a patient event;
in response to detecting the patient event, automatically generating, by the at least one processor and using at least one generative machine learning model, a personalized recommendation comprising at least one of a recommended digital therapy program for the patient, a recommended modification of a digital therapy program of the patient, or a personalized message recommended for transmission to the patient, wherein the at least one generative machine learning model comprises a transformer-based Large Language Model (LLM), and the automatic generation of the personalized recommendation comprises:
dynamically generating, by the at least one processor, a prompt for processing by the transformer-based LLM, the prompt comprising at least some of the patient data and the prompt further comprising an instruction for generating the personalized recommendation, and
providing, by the at least one processor, the prompt to the transformer-based LLM such that the transformer-based LLM processes the prompt and automatically generates the personalized recommendation;
invoking, by the at least one processor and based at least partially on the personalized recommendation, an action in the digital therapy platform;
automatically adjusting, by the at least one processor, the patient profile to reflect the action; and
causing, by the at least one processor, presentation of at least one of a first indication of the action at the first device or a second indication of the action at the second device.
|