Artificial intelligence for evaluating patient distress using facial emotion recognition
Inventors
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Assignees
MemberUniversity of MiamiUniversity of MiamiThe University of Miami, established in 1925 and based in Coral Gables, Florida, is a private research university recognized for its comprehensive academic offerings, robust research infrastructure, and strong interdisciplinary focus. Home to more than 19,000 students and with more than 400 acres of campuses across the Miami region, its mission encompasses education, research, innovation, and community service. The institution supports clinical, biomedical, marine, and atmospheric research initiatives, delivers diverse undergraduate and graduate programs, and maintains numerous research centers and institutes dedicated to scientific, medical, and societal advancements.
The University of Miami, established in 1925 and based in Coral Gables, Florida, is a private research university recognized for its comprehensive academic offerings, robust research infrastructure, and strong interdisciplinary focus. Home to more than 19,000 students and with more than 400 acres of campuses across the Miami region, its mission encompasses education, research, innovation, and community service. The institution supports clinical, biomedical, marine, and atmospheric research initiatives, delivers diverse undergraduate and graduate programs, and maintains numerous research centers and institutes dedicated to scientific, medical, and societal advancements.
Publication Number
US-12693733-B2
Publication Date
2026-07-28
Expiration Date
Abstract
Artificial intelligence for evaluating patient distress using facial emotion recognition. In an embodiment, an artificial intelligence model is applied to facial image(s) of a patient to classify each facial image into one of a plurality of emotional states based on a facial expression in the facial image. A determination may be made as to whether or not to alert a healthcare provider based on the emotional state(s) into which the facial image(s) were classified. If a determination is made to alert a healthcare provider, a notification may be transmitted to the healthcare provider.
Core Innovation
The invention provides a method and system for monitoring patients in a medical facility by using surveillance-captured facial images over a period of time. For each patient, a first artificial-intelligence model generates an output comprising a classification of each facial image into one of a plurality of emotional states based on a facial expression in the facial image, and the outputs from the first artificial-intelligence model are used to generate a history of emotional states for the patient.
The history of emotional states is provided as input into a second artificial-intelligence model that is different from the first artificial-intelligence model. The second artificial-intelligence model generates an output indicative of an emotional state of the patient, and based upon the output, the invention determines whether or not to alert one or more recipients.
When the determination is made to alert the one or more recipients, the invention transmits a notification to each of the one or more recipients. The disclosed monitoring further covers generating and evaluating the emotional-state history, applying the model outputs to alert decision logic, and storage of emotional-state outputs for patient and facility analytics and aggregation into distress scores and facility rating.
Claims Coverage
The document contains three independent claims (method, system, and non-transitory computer-readable medium). Across the independent claims, the same core inventive sequence is recited using two different artificial-intelligence models to generate a classification history of emotional states from facial images and then evaluate that history to decide whether to alert recipients and transmit notifications. The inventive features below capture the main elements recited in the independent claims.
Monitoring patients using facial images over time with two different AI models
Receives a plurality of facial images of the patient over a period of time, generates a history of emotional states by inputting each facial image into a first artificial-intelligence model that classifies the facial image into one of a plurality of emotional states based on a facial expression, and evaluates the history by inputting it into a second artificial-intelligence model different from the first to generate an output indicative of an emotional state of the patient.
Alert decision and notification transmission based on evaluated emotional-state output
Based upon the output of the second artificial-intelligence model, determines whether or not to alert one or more recipients, and upon determining to alert the one or more recipients, transmits a notification to each of the one or more recipients.
System configured with hardware and software modules for two-stage emotional-state evaluation
A system comprising at least one hardware processor and one or more software modules configured to, for each of one or more persons, receive a plurality of facial images over a period of time, generate a history of emotional states using a first artificial-intelligence model, evaluate the history using a second artificial-intelligence model different from the first, determine whether or not to alert one or more recipients, and transmit a notification to each recipient.
Non-transitory computer-readable medium instructions for two-stage emotional-state evaluation and alerting
A non-transitory computer-readable medium having instructions stored therein that, when executed by a processor, cause the processor to, for each of one or more persons, receive a plurality of facial images over a period of time, generate a history of emotional states using a first artificial-intelligence model, evaluate the history using a second artificial-intelligence model different from the first, determine whether or not to alert one or more recipients, and transmit a notification to each recipient.
The independent claims are directed to a two-model framework in which facial-image-based emotional-state classifications are accumulated into a history, then evaluated by a different second artificial-intelligence model to produce an output indicative of the patient’s emotional state that drives whether recipients are alerted and notified.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?