US 12,073,936 B2
Methods and systems for improved therapy delivery and monitoring
Ronan Patrick Cummins, Cambridge (GB); Alan James Martin, Cambridge (GB); Mihai Valentin Tablan, Cambridge (GB); and Michael Ewbank, Cambridge (GB)
Assigned to IESO DIGITAL HEALTH LIMITED, Cambridge (GB)
Appl. No. 17/055,178
Filed by IESO DIGITAL HEALTH LIMITED, Cambridge (GB)
PCT Filed May 17, 2019, PCT No. PCT/GB2019/051380
§ 371(c)(1), (2) Date Nov. 13, 2020,
PCT Pub. No. WO2019/220144, PCT Pub. Date Nov. 21, 2019.
Claims priority of application No. 1808051 (GB), filed on May 17, 2018; and application No. 1818640 (GB), filed on Nov. 15, 2018.
Prior Publication US 2021/0202065 A1, Jul. 1, 2021
Int. Cl. G16H 20/70 (2018.01); G06N 3/08 (2023.01); G16H 80/00 (2018.01)
CPC G16H 20/70 (2018.01) [G06N 3/08 (2013.01); G16H 80/00 (2018.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented method for taking one or more action relating to therapy, the method comprising:
receiving, by a computer, text data relating to a therapy session between a therapist and a patient;
dividing, by the computer, the text data into a plurality of utterances;
assigning, by the computer, a semantic representation to each of the plurality of utterances to produce a plurality of assigned utterances;
aggregating, by the computer, the plurality of assigned utterances to form a representation of the therapy session;
providing, by the computer, in real-time during the therapy session and based at least in part on utilizing a machine learning model, an output prediction comprising a measure of a quality of therapy delivered by the therapist, based on the representation of the therapy session and optionally one or more further input, of one or more characteristic of at least one of the patient, the therapist, and the therapy, wherein providing the output prediction comprising the measure of the quality of the therapy delivered by the therapist comprises:
determining, via the machine learning model and in real-time during the therapy session, that representation of the therapy session indicates that the therapy session is currently associated with a likelihood of patient recovery that is below average; and
automatically generating an alert, during the therapy session and in real-time via an interface to the therapist, indicating (i) at least one category of utterances as a cause for the likelihood of patient recovery being below average, and (ii) an adjustment of a frequency of utterances to be delivered to the patient for the at least one category of utterances to improve the likelihood of patient recovery,
wherein the one or more action comprises providing at least one utterance in accordance with the frequency of utterances of the at least one category of utterances; and
taking, by the computer, one or more action relating to the therapy determined to improve the quality of the therapy for a medical condition, wherein the one or more action is selected based on the output prediction meeting a predetermined at least one criterion.