Hypothetical scenario evaluation in infectious disease dynamics based on similar regions

Patent metadata

Field	Value
Publication number	US-12062456-B2
Application number	US-202117332356-A
Country	US
Kind code	B2
Filing date	May 27, 2021
Priority date	May 27, 2021
Publication date	Aug 13, 2024
Grant date	Aug 13, 2024

Abstract

Mechanisms are provided to hypothetical scenario evaluations with regard to infectious disease dynamics based on similar regions. A user definition of a hypothetical scenario for a target region is received which specifies scenario features affecting an infectious disease spread amongst a population within the target region. Other predefined regions, in the set of predefined regions, having similar region characteristics to the target region are identified and attributes of the other predefined regions corresponding to the scenario features are identified. Modified model parameter(s) for an infectious disease computer model are derived based on the identified attributes. An instance of the infectious disease computer model is configured with the modified model parameter(s) and the instance is executed on case report data for the target region to generate a prediction for an infectious disease spread in the target region according to the hypothetical scenario, which is then output.

First claim

The invention claimed is:

1. A method, in a data processing system comprising at least one processor and at least one memory coupled to the at least one processor and having instructions executed by the at least one processor to specifically configure the at least one processor to execute the method comprising:
executing, by a machine learning engine, a machine learning training of an infectious disease computer model on regional infectious disease and population (RIDP) database data for a set of predefined regions, to thereby train the infectious disease computer model to predict at least one characteristic of an infectious disease spread, wherein the RIDP database data comprising initializer ranges for hyperparameters of the infectious disease computer model and population data for the regions in the set of predefined regions, wherein the infectious disease computer model is a neural network computer model, and wherein the machine learning training comprises iteratively:
processing, by the infectious disease computer model, input features from the RIDP database data based on a configuration of hyperparameters of the infectious disease computer model according to the initializer ranges, to generate a prediction of at least one characteristic of an infectious disease spread,
comparing the predicted characteristic of the infectious disease spread to a ground truth for the characteristic of the infectious disease spread to determine an error, and
modifying at least one operational parameter of the infectious disease computer model based on the error, wherein the processing, comparing, and modifying operations are repeated iteratively until the error is equal to or below a predetermined threshold error or a predetermined number of iterations have occurred;

receiving, via a user interface of the data processing system, a user definition of a hypothetical scenario for a target region, in the set of predefined regions, wherein the definition of the hypothetical scenario specifies one or more scenario features affecting the infectious disease spread amongst a population within the target region;
identifying one or more other predefined regions, in the set of predefined regions, having second region characteristics corresponding to first region characteristics of the target region at least by executing clustering computer logic on region characteristics of the set of predefined regions to generate a plurality of clusters of regions based on one or more distance valuation algorithms applied to the region characteristics, and identifying a cluster associated with the target region that comprises the one or more other predefined regions;
identifying one or more attributes of the one or more other predefined regions corresponding to the one or more scenario features based on learned changes in historical data corresponding to the one or more other predefined regions;
deriving at least one modified model parameter for the trained infectious disease computer model based on the identified one or more attributes that correspond to the one or more scenario features, wherein the at least one modified model parameter comprises at least one hyperparameter of the infectious disease computer model;
configuring an instance of the trained infectious disease computer model with the at least one modified model parameter;
executing the instance of the trained infectious disease computer model on case report data for the target region to generate at least one prediction for the infectious disease spread in the target region according to the hypothetical scenario; and
outputting the at least one prediction in a report output to a user computing device.

2. The method of claim 1 , wherein the user definition is a natural language definition, and wherein processing the user definition comprises performing natural language processing on the natural language definition to extract natural language features of the natural language definition that are mapped to at least one of infectious disease characteristics or characteristics of interventions that affect infectious disease characteristics.

3. The method of claim 1 , wherein the one or more attributes of the one or more other predefined regions comprise attributes corresponding to implementation or relaxing of interventions in corresponding predefined regions of the one or more other predefined regions.

4. The method of claim 1 , wherein the one or more attributes of the one or more other predefined regions are levels of infectious disease characteristics corresponding to hypothetical levels of infectious disease state characteristics or transmission characteristics corresponding to the one or more scenario features.

5. The method of claim 1 , wherein deriving at least one modified model parameter for the trained infectious disease computer model based on the identified one or more attributes comprises, for the one or more other predefined regions:
generating a time ordered curve of case report data for the one or more other predefined regions;
identifying one or more inflection points in the time ordered curve;
correlating the one or more inflection points in the time ordered curve with one or more intervention entries specified in time stamped infectious disease intervention data for the one or more other predefined regions, the one or more intervention entries specifying interventions implemented by authorities to control spread of the infectious disease in the one or more other predefined regions; and
identifying the at least one modified model parameter based on results of correlating the one or more inflection points with the one or more intervention entries.

6. The method of claim 5 , wherein the at least one modified model parameter comprises at least one hyperparameter of the trained infectious disease computer model specifying at least one of a transmission rate of the infectious disease or a mobility of a population of the target region.

7. The method of claim 1 , wherein the one or more scenario features are scenario features specifying a potential implementation or relaxing of an intervention to control spread of the infectious disease in the target region.

8. The method of claim 1 , wherein the infectious disease computer model is a compartmental computer model comprising a plurality of compartments, each compartment corresponding to a state of the infectious disease and having a corresponding set of one or more differential equations modeling a portion of a population associated with the corresponding compartment, and wherein the compartmental computer model comprises one or more mobility isolation and countermeasure (MIC) compartments associated with corresponding other compartments of the compartmental computer model, and wherein the MIC compartments model an isolation of a portion of a population of a corresponding other compartment, based on mobility data for the population.

9. A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed in a data processing system, causes the data processing system to:
execute, by a machine learning engine, a machine learning training of an infectious disease computer model on regional infectious disease and population (RIDP) database data for a set of predefined regions, to thereby train the infectious disease computer model to predict at least one characteristic of an infectious disease spread, wherein the RIDP database data comprising initializer ranges for hyperparameters of the infectious disease computer model and population data for the regions in the set of predefined regions, wherein the infectious disease computer model is a neural network comp

Assignees

• Merative Us L P

Inventors

• Gopalakrishnan Vishrawas
• Deshpande Ajay Ashok
• Navalekar Sayali
• Kaufman James H
• Bianco Simone
• Hu Kun
• Liu Xuan
• Miller Jacob Ora
• Srinivasan Raman
• Ding Pan

Classifications

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G16H15/00

  ICT specially adapted for medical reports, e.g. generation or transmission thereof · CPC title

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F40/20

  Natural language analysis (semantic analysis of natural language G06F40/30) · CPC title

• Y02A90/10

  Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation · CPC title