Back to results
US 12,597,508 B2Granted
Computerized decision support tool for post-acute care patients
AssigneeCerner Innovation, Inc.
InventorsLeonardo Ji, Andrew Roberts, Sasanka Are
Life of this patent
Estimating the term…
Abstract
A decision support tool is provided for discharging a patient by predicting the probability of a patient's readmission based on information prior to discharge. The information used to make the prediction may include labs, vitals, diagnoses, and medications from prior encounters and from the current encounter. At least some of this information may be used to compute a readmission prediction. One or more recurrent neural networks may be utilized to generate the prediction, and one or more sequences may be generated from the patient information to feed into the one or more recurrent neural networks. Based on the prediction, one or more actions may be initiated to reduce the probability of readmission.
Claims21
A non-transitory computer storage media having computer-executable instructions embodied thereon that, when executed, provide a method for predicting readmission for a patient, the method comprising:
- training a recurrent neural network based on patient event data, of a plurality of patients, to predict readmission probabilities for patients discharged from acute care facilities and having received post-acute care services, the patient event data of the plurality of patients comprising (a) healthcare visits, (b) healthcare events, and (c) patient readmission data;
- receiving patient data for a patient (i) discharged from an acute care facility and (ii) currently receiving post-acute care services, the patient data corresponding to patient care services received over time;
- generating at least two sequences based on the patient data, wherein the at least two sequences comprise:
- (a) a visits sequence indicating a plurality of healthcare visits by the patient, wherein the visits sequence is in reverse chronological order, and
- (b) an events sequence indicating a plurality of healthcare events for the patient, wherein the events sequence is in chronological order;
- applying the recurrent neural network to a combination of time series data corresponding to the at least two sequences to predict a probability of the patient being readmitted into the acute care facility within a future time interval, wherein the recurrent neural network comprises a visits layer that receives the visits sequence and an events layer that receives the events sequence;
- wherein prior to discharge of the patient from receiving the post-acute care services, an action for treatment of the patient is initiated based on the probability of the patient being readmitted into the acute care facility;
- and
- retraining the recurrent neural network based on feedback corresponding to the probability of the patient being readmitted into the acute care facility.
6 more elements
A computerized method for implementing a decision support tool for a patient receiving post-acute care services, the computerized method comprising:
- training a recurrent neural network based on patient event data, of a plurality of patients, to predict readmission probabilities for patients discharged from acute care facilities and having received post-acute care, the patient event data of the plurality of patients comprising (a) healthcare visits, (b) healthcare events, and (c) patient readmission data;
- receiving patient data for a patient (i) discharged from an acute care facility and (ii) currently receiving post-acute care, the patient data corresponding to patient care services received over time;
- generating at least two sequences based on the patient data, wherein the at least two sequences comprise:
- (a) a visits sequence indicating a plurality of healthcare visits by the patient, wherein the visits sequence is in reverse chronological order, and
- (b) an events sequence indicating a plurality of healthcare events for the patient, wherein the events sequence is in chronological order;
- applying the recurrent neural network to a combination of time series data corresponding to the at least two sequences to predict a probability of the patient being readmitted into the acute care facility within a future time interval, wherein the recurrent neural network comprises a visits layer that receives the visits sequence and an events layer that receives the events sequence;
- wherein prior to discharge of the patient from receiving post-acute care services, an action for treatment of the patient is initiated based on the probability of the patient being readmitted into the acute care facility;
- and
- retraining the recurrent neural network based on feedback corresponding to the probability of the patient being readmitted into the acute care facility.
6 more elements
A computerized system for predicting a probability of being readmitted for a patient who has received post-acute care services, the system comprising:
- one or more processors;
- and
- one or more computer storage media storing computer-useable instructions that, when used by the one or more processors, cause the one or more processors to perform a method, the method comprising:
- training a recurrent neural network based on patient event data of a plurality of patients to predict readmission probabilities for patients discharged from acute care facilities and having received post-acute care services, the patient event data of the plurality of patients comprising (a) healthcare visits, (b) healthcare events, and (c) patient readmission data;
- receiving patient data for a patient (i) discharged from an acute care facility and (ii) currently receiving post-acute care services;
- generating at least two sequences based on the patient data, wherein the at least two sequences comprise:
- (a) a visits sequence indicating a plurality of healthcare visits by the patient, wherein the visits sequence is in reverse chronological order, and
- (b) an events sequence indicating a plurality of healthcare events for the patient, wherein the events sequence is in chronological order;
- applying the recurrent neural network to a combination of time series data corresponding to the at least two sequences to predict a probability of the patient being readmitted, into the acute care facility within a future time interval, wherein the recurrent neural network comprises a visits layer that receives the visits sequence and an events layer that receives the events sequence;
- wherein prior to discharge of the patient from receiving post-acute care services, an action for treatment of the patient is initiated based on the probability of the patient being readmitted into the acute care facility;
- and
- retraining the recurrent neural network based on feedback corresponding to the probability of the patient being readmitted into the acute care facility.
9 more elements
Dependencies are inferred from the claim text.