US 12,406,770 B2
Hybrid predictive model for alertness monitoring
Mary D. Swift, Rochester, NY (US); Donna K. Byron, Petersham, MA (US); and Ashok Kumar, North Chelmsford, MA (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Jan. 3, 2019, as Appl. No. 16/238,877.
Prior Publication US 2020/0219616 A1, Jul. 9, 2020
Int. Cl. G06N 20/00 (2019.01); G06N 5/043 (2023.01); G06N 7/01 (2023.01); G16H 10/60 (2018.01); G16H 50/20 (2018.01)
CPC G16H 50/20 (2018.01) [G06N 5/043 (2013.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01); G16H 10/60 (2018.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
progressively obtaining, by one or more processors, data related to medical treatment and a current physical state of a given patient from a patient monitoring system, an electronic medical records system, and a scheduling system, wherein the patient monitoring system comprises Internet of Things devices and radio-frequency identification tags, wherein the Internet of Things devices and the radio-frequency identification tags are located proximate to the given patient and monitor wakefulness and alertness of the given patient;
cognitively analyzing, by the one or more processors, the data to identify potential patient behavioral patterns of the given patient;
obtaining, by the one or more processors, general patient data;
correlating, by the one or more processors, the general patient data with a portion of the progressively obtained data to identify one or more elements in the general patient data with a potential impact on the identified potential patient behavioral patterns;
determining, by the one or more processors, impacts of the one or more elements on the identified potential behavioral patterns and applying the impacts to generate a data structure comprising baseline behavioral patterns, wherein the data structure comprises a predictive model to utilize in determining one or more probabilities that the given patient will exhibit one or more behaviors comprising the baseline behavioral patterns during a defined future time interval;
applying, by the one or more processors, the predictive model, to the defined future time interval, wherein the applying comprises generating a timeline comprising the future time interval, wherein the timeline is segmented into one or more sub-intervals, wherein a probability of the one or more probabilities is assigned to each sub-interval of the one or more sub-intervals, wherein the probabilities comprise likelihoods of wakefulness and availability for the given patient in each sub-interval of the one or more sub-intervals in a predicted location;
transmitting, by the one or more processors, the timeline to authorized users, via computing nodes communicatively coupled to the one or more processors via an Internet connection;
obtaining, by the one or more processors, via one or more of the computing nodes, after passage of a given sub-interval, feedback regarding veracity of one or more probabilities assigned to the given sub-interval, wherein the feedback comprises passive feedback;
continuously obtaining, by the one or more processors, during the given sub-interval, data from the Internet of Things devices indicating the wakefulness and the alertness of the given patient during the sub-interval; and
continuously updating, by the one or more processors, the predictive model based on the data and the feedback.