Systems and methods for treatment of a patient by automated patient care
Inventors
Elia, Liron • Iddan, Gavriel J.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10682288-B1
Publication Date
2020-06-16
Expiration Date
Abstract
A computer-implemented method of treating a patient's and automated enteral feeding, comprising: monitoring a plurality of reflux-related parameters and at least one reflux event while the patient is automatically enterally fed by an enteral feeding controller according to a baseline feeding profile including a target nutritional goal, training a classifier component of a model for predicting likelihood of a future reflux event according to an input of scheduled and/or predicted plurality of reflux-related parameters, the classifier trained according to computed correlations between the plurality of reflux-related parameters and the at least one reflux event, feeding scheduled and/or predicted reflux-related parameters into the trained classifier component of the model for outputting risk of likelihood of a future reflux event, and computing, by the model, an adjustment to the baseline feeding profile for reducing likelihood of the future reflux event and for meeting the target nutritional goal.
Core Innovation
A computer-implemented method monitors a plurality of gastric reflux-related parameters and at least one gastric reflux event while a patient is automatically enterally fed by an enteral feeding controller according to a baseline feeding profile including a target nutritional goal. The method uses gastric reflux-related parameters observed during automatic enteral feeding to address the occurrence and prevention of gastric reflux events.
A classifier component of a model is trained to predict likelihood of a future gastric reflux event based on an input of scheduled and/or predicted plurality of gastric reflux-related parameters. The classifier is trained according to computed correlations between the plurality of gastric reflux-related parameters and the at least one gastric reflux event, and the scheduled and/or predicted gastric reflux-related parameters are fed into the trained classifier to output risk or likelihood of a future gastric reflux event.
The model computes an adjustment to the baseline feeding profile for reducing likelihood of the future gastric reflux event and for meeting the target nutritional goal. In an automated patient-care approach, the method monitors patient-related parameters, enteral delivered substances, and gastric reflux-event parameters over a monitoring interval, creates a training dataset with feature vectors associated with time, trains a model on computed correlations, outputs instructions to adjust enteral delivered substances, and adjusts the enteral delivered substances according to the instructions.
Claims Coverage
The independent claims include two inventive methods. Both claims center on monitoring gastric reflux-related information during automatic enteral feeding, training a model using computed correlations, predicting risk/likelihood of a future gastric reflux event, and computing or outputting adjustments to a baseline feeding profile or enteral delivered substances to reduce future reflux while meeting a target nutritional goal.
Monitoring gastric reflux-related parameters and reflux events during automatic enteral feeding
Monitoring a plurality of gastric reflux-related parameters and at least one gastric reflux event while the patient is automatically enterally fed by an enteral feeding controller according to a baseline feeding profile including a target nutritional goal.
Training a classifier using computed correlations to predict future reflux likelihood
Training a classifier component of a model for predicting likelihood of a future gastric reflux event according to an input of scheduled and/or predicted plurality of gastric reflux-related parameters, the classifier trained according to computed correlations between the plurality of gastric reflux-related parameters and the at least one reflux event.
Predicting risk of a future gastric reflux event from scheduled/predicted reflux-related parameters
Feeding scheduled and/or predicted gastric reflux-related parameters into the trained classifier component of the model for outputting risk of likelihood of a future gastric reflux event.
Computing an adjustment to reduce reflux likelihood while meeting nutritional goal
Computing, by the model, an adjustment to the baseline feeding profile for reducing likelihood of the future gastric reflux event and for meeting the target nutritional goal.
Creating a time-indexed training dataset including patient parameters, substances, and reflux-event parameters
Creating a training dataset by computing a plurality of feature vectors each associated with an indication of time during the monitoring interval, each feature vector storing the plurality of patient-related parameters, the plurality of enteral delivered substances, and the plurality of gastric reflux-event parameters.
Training a model on computed correlations to output instructions for substance adjustment
Training a model adapted to receive current patient-related parameters and output instructions for adjustment of the enteral delivered substances for reducing likelihood of a future gastric reflux event, the model trained according to the training dataset based on computed correlations between the plurality of patient-related parameters, the plurality of enteral delivered substances, and the plurality of gastric reflux-event parameters.
Adjusting enteral delivered substances according to model instructions
Feeding current patient-related parameters into the trained model for outputting instructions for adjustment of the enteral delivered substances for reducing likelihood of a future gastric reflux event; and adjusting the enteral delivered substances according to the instructions.
Across both independent claims, the inventive core is monitoring gastric reflux-related data during automated enteral feeding, training a model/classifier using computed correlations, predicting risk/likelihood of a future gastric reflux event, and computing adjustments to a baseline feeding profile and/or to enteral delivered substances to reduce future reflux while meeting the target nutritional goal.
Stated Advantages
Reduce likelihood of a future gastric reflux event.
Meet a target nutritional goal while reducing likelihood of a future gastric reflux event.
Documented Applications
Automated enteral feeding treatment in which a controller uses monitored gastric reflux-related parameters and reflux events to predict future reflux likelihood and adjust feeding to reduce future reflux while meeting a target nutritional goal.
Automated patient care in which monitoring and model-driven instructions adjust enteral delivered substances (including substances tracked in the training dataset) to reduce likelihood of a future gastric reflux event.
Interested in licensing this patent?