System and method for projecting product movement

• US 10,296,927 B2
• Filed: 12/20/2013
• Issued: 05/21/2019
• Est. Priority Date: 12/20/2013
• Status: Active Grant

• Current Assignee: IQVIA, Inc. (IQVIA Holdings, Inc.)
• Original Assignee: IQVIA, Inc. (IQVIA Holdings, Inc.)
• Inventors: Soria, Luis, Nadarajah, Sivakumar
• Primary Examiner(s): Ng, Jonathan
• Assistant Examiner(s): Reichert, Rachelle L
• Application Number:
  US14/136,706
• Publication Number:
  US 20150178745A1
• Time in Patent Office: 1,978 Days
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A method includes receiving data encoding search results from at least one search engine, each of the at least one search engine applying an ontology in a corresponding social-media channel, the ontology refined to account for at least one healthcare product and at least one event; retrieving a database that includes relationships between identified event and healthcare indication, the database being developed from historical sales activities of healthcare products and healthcare indications for the healthcare products; and analyzing the search results by: determining at least one event identified by the search results; comparing the at least one event with the retrieved database to determine a corresponding healthcare indication; identifying a healthcare product with the determined corresponding healthcare indication; and correlating the search results about the identified healthcare product with historical sales activities of the identified healthcare product to determine a sales trend for the identified healthcare product during a projected period.

Claims

1. A computer-implemented method, the method comprising:

text mining, according to an ontology, in a social media channel that is coupled to a search engine through a feed reader interface to identify at least one event for determining an impact to demand for at least one healthcare product, the social media channel collecting and posting inputs from user mobile devices;

determining, using a distributed computer network that is configured to access a data storage device, a healthcare indication based on analysis of a database that includes information encoding historical events, contemporaneous records data of healthcare products, and data representing causal links between the healthcare indication and the healthcare products in the database, wherein the healthcare indication represents a specific medical condition that necessitates treatment using the healthcare products;

determining, using the distributed computer network and based on analysis of each historical event, impact to demands for the healthcare products whose record data is encoded, each historical event being statistically correlated with the healthcare indication that is treatable using the healthcare products;

using the determined healthcare indication and impact to demands for the healthcare products, identifying a healthcare product that has a causal link to the healthcare indication based on data representations of the database, wherein the records data of the identified healthcare product is encoded by information accessed during analysis of the database;

correlating, using a statistical forecasting engine of the distributed computer network, the identified at least one event with records data of the identified healthcare product based on statistical analysis to;

(i) combine weights and number of occurrences of terms corresponding to the identified at least one event as identified from the social-media channel, both weights and number of occurrences tapering off as time elapses since the identified at least one event, and

(ii) determine a trend for the identified healthcare product during a projected period since the identified at least one event;

comparing the determined trend for the identified healthcare product with actual data of the same identified healthcare product during the projected period when the actual data of the same identified healthcare product during the projected period becomes available in the database;

identifying a difference between the determined trend and the actual data of the same identified healthcare product during the projected period; and

based on the identified difference, enhancing a trend projection capability of the statistical forecasting engine to reduce the identified difference between the determined trend and the actual data, wherein the trend projection capability is enhanced relative to a first projection capability of the statistical forecasting engine in response to the distributed computer network iteratively updating the database to account for new events that have a correlation to the healthcare indication and the identified healthcare product, and wherein enhancing the trend projection capability of the statistical forecasting engine includes;

executing a load balancing tactic for updating a regression model of the statistical forecasting engine, the load balancing, tactic being executed to balance a computing load distributed to each computer of the distributed computer network that updates the regression model;

wherein the load balancing tactic and the undated regression model are used to analyze the new events stored at the database in response to the iterative updates of the database.

2. The method of claim 1, further comprising:

performing a regression analysis on the records data to determine the trend for the identified healthcare product, the regression analysis accounting for contributions from the identified at least one event.

3. The method of claim 1, further comprising:

dynamically presenting the determined trend for the healthcare product against a competing healthcare product with identical healthcare indication in a graphic user interface such that the determined trend is projected in response to each of the at least one identified event.

4. The method of claim 1, further comprising:

providing the determined trend to cumulatively build a profile of the identified healthcare product.

5. The method of claim 4, further comprising:

recording actual data for the identified healthcare product in the profile; and

comparing the trend for the identified healthcare product in the profile to actual sales data.

6. The method of claim 1, further comprising:

receiving the ontology for the at least one healthcare product; and

refining the ontology by incorporating reporting parameters that include at least one of;

product level data, supplier data, facility level data, or class of trade data.

7. The method of claim 1, further comprising:

receiving the ontology for the at least one healthcare product; and

refining the ontology by incorporating event-driven parameters defined to include at least one of;

a natural event, or a government event.

15. The method of claim 1, wherein enhancing the trend projection capability of the statistical forecasting, engine further includes:

executing, at the distributed computer network, the load balancing tactic to analyze data in a log file against the database, the data in the log file comprising information about recent events obtained from search results of the search engine; and

in response to executing the load balancing tactic, iteratively updating, using the computers of the distributed computer network, the database to account for the new events that have the correlation to the healthcare indication and the identified healthcare product.

16. The method of claim 15, wherein executing the load balancing tactic comprises:

obtaining a computing power for each computer in the distributed computer network;

determining a geographic location of the data in the log file and a proximity of the data to one or more computers of the distributed computer network; and

distributing a computing load for iteratively updating the database using the obtained computing power and the proximity of the data to the one or more computers.

8. A computer system, comprising at least one processor, wherein the at least one processor is configured to perform the operations of:

applying an ontology to text mine in a corresponding social-media channel where inputs from user mobile devices have been collected and posted such that at least one event is identified to determine an impact to demand for at least one healthcare product relating to a particular healthcare indication;

determining, using a distributed computer network that is configured to access a data storage device, a healthcare indication based on analysis of a database encoding information that includes historical events and data representing causal links between the healthcare indication and healthcare products in the database, the database being developed from historical events that are each statistically correlated with one or more healthcare indications, and wherein the healthcare indication represents a specific medical condition that necessitates treatment using the healthcare products; and

determining, using the distributed computer network and based on analysis of each historical event, impact to demands for the healthcare products in the database;

using the determined healthcare indication and impact to demands for the healthcare products, identifying a healthcare product that has a causal link to the healthcare indication based on data representations of the database;

correlating, using a statistical forecasting engine of the distributed computer network, the identified at least one event with historical activities of the identified healthcare product based on statistical analysis to;

(i) combine weights and number of occurrences of terms corresponding to the identified at least one event as text mined from the social-media channel, both weights and number of occurrences tapering off as, time elapses since the identified at least one event, and

(ii) determine a trend for the identified healthcare product during a projected period since the identified at least one event;

comparing the determined trend, for the identified healthcare product with actual data of the same identified healthcare product during the projected period;

identifying a difference between the determined trend and the actual data of the same identified healthcare product during the projected period; and

based on the identified difference, enhancing a trend projection capability of the statistical forecasting engine to reduce the identified difference between the determined trend and the actual data, wherein the trend projection capability is enhanced relative to a first projection capability of the statistical forecasting engine in response to the distributed computer network iteratively updating the database to account for new events that have a correlation to the healthcare indication and the identified healthcare product, and wherein enhancing the trend projection capability of the statistical forecasting engine includes;

executing a load balancing tactic for updating a regression model of the statistical forecasting engine, the load balancing, tactic being executed to balance a computing load distributed to each computer of the distributed computer network that updates the regression model;

wherein the load balancing tactic and the undated regression model are used to analyze the new events stored at the database in response to the iterative updates of the database.

9. The computer system of claim 8, further comprising:

performing a regression analysis on the historical activities to determine the trend for the identified healthcare product, the regression analysis accounting for contributions from the identified at least one event.

10. The computer system of claim 8, wherein the operations further comprise:

dynamically presenting the determined trend for the healthcare product against a competing healthcare product with identical healthcare indication in a graphic user interface such that the determined trend is projected in response to each of the at least one identified event.

11. The computer system of claim 8, wherein the operations further comprise:

providing the determined trend, to cumulatively build a profile of the identified healthcare product.

12. The computer system of claim 8, wherein the operations further comprise:

receiving the ontology for the at least one healthcare product; and

refining the ontology by incorporating reporting parameters that include at least one of;

product level data, supplier data, facility level data, or class of trade data.

13. The computer system of claim 8, wherein the operations further comprise:

receiving the ontology for the at least one healthcare product; and

refining the ontology by incorporating event-driven parameters defined to include at least one of;

a natural event, or a government event.

17. The computer system of claim 8, wherein enhancing the trend projection capability of the statistical forecasting engine further includes:

executing, at the distributed computer network, the load balancing tactic to analyze data in a log file against the database, the data in the log file comprising information about recent events obtained from the social-media channel; and

in response to executing the load balancing tactic, iteratively updating, using the computers of the distributed computer network, the database to account for the new events that have the correlation to the healthcare indication and the identified healthcare product.

18. The computer system of claim 17, wherein executing the load balancing tactic comprises:

obtaining a computing power for each computer in the distributed computer network;

determining a geographic location of the data in the log file and a proximity, of the data to one or more computers of the distributed computer network; and

distributing a computing load for iteratively updating the database using the obtained computing power and the proximity of the data to the one or more computers.

14. A non-transitory computer-readable medium comprising software instructions that, when executed by a computer, causes the computer to perform the operations of:

applying an ontology to text mine in a corresponding social-media channel where inputs from user mobile devices have been collected and posted such that at least one event is identified to determine an impact to demand for at least one healthcare product relating to a particular healthcare indication;

determining, using a distributed computer network that is configured to access a data storage device, a healthcare indication based on analysis of a database encoding information that includes, historical events and data representing causal links between the healthcare indication and healthcare products in the database, the database being developed from historical events that are each statistically correlated with one or more healthcare indications, and wherein the healthcare indication represents a specific medical condition that necessitates treatment using the healthcare products; and

determining, using the distributed computer network and, based on analysis of each historical event, impact to demands for the healthcare products in the database;

using the determined healthcare indication and impact to demands for the healthcare products, identifying a healthcare product that has a causal link to the healthcare indication based on data representations of the database;

correlating, using a statistical forecasting engine of the distributed computer network, the identified at least one event with historical activities of the identified healthcare product based on statistical analysis to;

(i) combine weights and number of occurrences of terms corresponding to the identified at least one event as text mined from the social-media channel, both weights and number of occurrences tapering off as time elapses since the identified at least one event, and

(ii) determine a trend for the identified, healthcare product during a projected period since the identified at least one event;

comparing the determined trend for the identified healthcare product with actual data of the same identified healthcare product during the projected period;

identifying a difference between the determined trend and the actual data of the same identified healthcare product during the projected period; and

based on the identified difference, enhancing a trend projection capability of the statistical forecasting engine to reduce the identified difference between the determined trend and the actual data, wherein the trend projection capability is enhanced relative to a first projection capability of the statistical forecasting engine in response to the distributed computer network, iteratively updating the database to account for new events that have a correlation to the healthcare indication and the identified healthcare product, and wherein enhancing the trend projection capability of the statistical forecasting engine includes;

executing a load balancing tactic for updating a regression model of the statistical forecasting engine, the load balancing, tactic being executed to balance a computing load distributed to each computer of the distributed computer network that updates the regression model;

wherein the load balancing tactic and the undated regression model are used to analyze the new events stored at the database in response to the iterative updates of the database.

19. The non-transitory computer-readable medium of claim 14, wherein enhancing the trend projection capability of the statistical forecasting engine further includes:

executing, at the distributed computer network, the load balancing tactic to analyze data in a log file against the database, the data in the log file comprising information about recent events obtained from the social-media channel; and

in response to executing the load balancing tactic, iteratively updating, using the computers of the distributed computer network, the database to account for the new events that have the correlation to the healthcare indication and the identified healthcare product.

20. The non-transitory computer-readable medium of claim 19, wherein executing the load balancing tactic comprises:

obtaining a computing power for each computer in the distributed computer network;

determining, a geographic location of the data in the log file and a proximity of the data to one or more computers of the distributed computer network; and

distributing a computing load for iteratively updating the database using the obtained computing power and the proximity of the data to the one or more computers.