Researchers Submit Patent Application, “Systems And Methods For Building Healthcare Networks”, for Approval (USPTO 20240145069): Patent Application

2024 MAY 21 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- From Washington, D.C., NewsRx journalists report that a patent application by the inventors Felemban, Nawaf (Washington, DC, US); Haddad, Christopher E (Denver, CO, US), filed on October 26, 2022, was made available online on May 2, 2024.

No assignee for this patent application has been made.

News editors obtained the following quote from the background information supplied by the inventors: “Health insurance companies face the challenging task of developing networks of healthcare providers to offer to their customers. In the past, such networks of healthcare providers were developed simply by including every possible healthcare provider in the network although not all approaches were successful at (a) identifying every provider, or (b) reaching a contractual agreement with each party in the proposed healthcare network. Indeed, some healthcare provider networks were developed with a philosophy of “broader is better.” However, over time it was learned that there were drawbacks to building healthcare provider networks this way. For example, this approach did not yield affordable, equitable and high quality networks.

“The systems and methods disclosed herein provide solutions to this problem and others, and may provide solutions to other drawbacks of conventional techniques.”

As a supplement to the background information on this patent application, NewsRx correspondents also obtained the inventors’ summary information for this patent application: “In one aspect, a computer-implemented method for building a healthcare provider network may be provided. The method may include: (1) receiving, via one or more processors, historical healthcare provider data; (2) training, via the one or more processors, a machine learning algorithm based on the historical healthcare provider data; (3) receiving, via the one or more processors, potential healthcare provider data, the potential healthcare provider data including data of individual healthcare providers, and healthcare provider facilities; and (4) building, via the one or more processors, a healthcare provider network by inputting the potential healthcare provider data into the trained machine learning algorithm.

“In another aspect, a computer-implemented method for building a healthcare provider network may be provided. The method may include: (1) receiving, via the one or more processors, potential healthcare provider data, the potential healthcare provider data including data of individual healthcare providers, and healthcare provider facilities; and (2) building, via the one or more processors, a healthcare provider network by inputting the potential healthcare provider data into the trained machine learning algorithm, wherein the built healthcare provider network includes individual healthcare providers and healthcare provider facilities.

“In yet another aspect, a computer system configured to build a healthcare provider network may be provided. The computer system may include one or more processors configured to: (1) receive historical healthcare provider data; (2) train a machine learning algorithm based on the historical healthcare provider data; (3) receive potential healthcare provider data, the potential healthcare provider data including data of individual healthcare providers, and healthcare provider facilities; and (4) build a healthcare provider network by inputting the potential healthcare provider data into the trained machine learning algorithm.”

The claims supplied by the inventors are:

“1. A computer-implemented method for building a healthcare provider network, the method comprising: receiving, via one or more processors, historical healthcare provider data; training, via the one or more processors, a machine learning algorithm based on the historical healthcare provider data, wherein the training comprises: applying a search algorithm to identify mini healthcare provider networks in the historical healthcare provider data; and training the machine learning algorithm based on the identified mini healthcare provider networks; receiving, via the one or more processors, potential healthcare provider data, the potential healthcare provider data including data of individual healthcare providers, and healthcare provider facilities; and building, via the one or more processors, a healthcare provider network by inputting the potential healthcare provider data into the trained machine learning algorithm; and wherein the mini healthcare provider networks are smaller than the healthcare provider network.

“2. The computer-implemented method of claim 1, wherein the healthcare provider network includes individual healthcare providers having individual healthcare provider types, and wherein the individual healthcare provider types include: a primary care provider (PCP); a cardiologist; a radiologist; a pediatrician; a dermatologist; a pulmonologist; and/or an oncologist.

“3. The computer-implemented method of claim 2, wherein the trained machine learning algorithm builds the healthcare provider network to include a disproportionately large number of PCP providers compared to other healthcare provider types.

“4. The computer-implemented method of claim 1, wherein the healthcare provider network includes healthcare provider facilities having healthcare provider facility types, and wherein the healthcare provider facility types include: a hospital; a primary care provider (PCP) office; a laboratory; an urgent care facility; and/or an addiction treatment facility.

“5. (canceled)

“6. The computer-implemented method of claim 1, wherein: the healthcare provider network has a type of a Medicare network, a Medicaid network, or an exchange network; and the trained machine learning algorithm builds the healthcare provider network based on the healthcare provider network type.

“7. The computer-implemented method of claim 1, wherein: the healthcare provider network has an organization type of health maintenance organization (HMO), or preferred provider organization (PPO); and the trained machine learning algorithm builds the healthcare provider network based on the organization type.

“8. The computer-implemented method of claim 1, wherein the healthcare provider network is a single tier network.

“9. The computer-implemented method of claim 1, wherein the healthcare provider network is a tiered network.

“10. The computer-implemented method of claim 1, wherein the trained machine learning algorithm builds the healthcare provider network by optimizing a cost efficiency metric, and/or a quality of care metric.

“11. The computer-implemented method of claim 1, wherein: the healthcare provider network includes a plurality of individual healthcare providers, and a plurality of healthcare organizations; and the trained machine learning algorithm builds the healthcare provider network based on affiliations between individual healthcare providers of the plurality of healthcare providers, and healthcare organizations of the plurality of healthcare organizations.

“12. The computer-implemented method of claim 1, wherein: the potential healthcare provider data includes data of individual healthcare providers; and the building of the healthcare provider network comprises: (i) identifying, in the potential healthcare provider data: (a) relationships between individual healthcare providers, and (b) strengths of the relationships between individual healthcare providers; and (ii) building the healthcare provider network based on the identified relationships and strengths of the relationships.

“13. The computer-implemented method of claim 1, wherein the historical healthcare provider data includes healthcare facilities classified based on an urban to rural scale.

“14. The computer-implemented method of claim 1, wherein: the historical healthcare provider data includes healthcare facilities classified based on population incomes of geographic areas that the healthcare facilities are located in; and the training of the machine learning algorithm comprises including a requirement that a healthcare provider network built by the machine learning algorithm include a higher density of healthcare facilities for a geographic area with a lower income population than a density of healthcare facilities for a geographic area with a higher income population.

“15. The computer-implemented method of claim 1, wherein: the training of the machine learning algorithm comprises including a requirement that a healthcare provider network built by the machine learning algorithm include a higher healthcare facility density for a geographic area with higher population density than for a geographic area with a lower population density.

“16. The computer-implemented method of claim 1, wherein the training comprises: assigning, in the historical data, cost efficiency metrics to individual healthcare providers based on readmission rates, and/or number of recommended tests.

“17. A computer-implemented method for building a healthcare provider network, the method comprising: receiving, via the one or more processors, potential healthcare provider data, the potential healthcare provider data including data of individual healthcare providers, and healthcare provider facilities; and building, via the one or more processors, a healthcare provider network by inputting the potential healthcare provider data into a machine learning algorithm, wherein the built healthcare provider network includes individual healthcare providers and healthcare provider facilities; wherein the machine learning algorithm is trained by: applying a search algorithm to identify mini healthcare provider networks in historical healthcare provider data; and training the machine learning algorithm based on the identified mini healthcare provider networks; and wherein the mini healthcare provider networks are smaller than the healthcare provider network.

“18. The computer-implemented method of claim 17, wherein the individual healthcare providers of the built healthcare provider network have individual healthcare provider types, and wherein the individual healthcare provider types include: a primary care provider (PCP); a cardiologist; a radiologist; a pediatrician; a dermatologist; a pulmonologist; and/or an oncologist.

“19. A computer system configured to build a healthcare provider network, the computer system comprising one or more processors configured to: receive historical healthcare provider data; train a machine learning algorithm based on the historical healthcare provider data, wherein the one or more processors are configured to train the machine learning algorithm by: applying a search algorithm to identify mini healthcare provider networks in the historical healthcare provider data; and training the machine learning algorithm based on the identified mini healthcare provider networks; receive potential healthcare provider data, the potential healthcare provider data including data of individual healthcare providers, and healthcare provider facilities; and build a healthcare provider network by inputting the potential healthcare provider data into the trained machine learning algorithm; and wherein the mini healthcare provider networks are smaller than the healthcare provider network.

“20. The computer system of claim 19, wherein the one or more processors are further configured to use the trained machine learning algorithm to build the healthcare provider network by optimizing a cost efficiency metric, and/or a quality of care metric.

“21. The computer-implemented method of claim 1, wherein: the search algorithm comprises a depth-first search algorithm or a breadth-first search algorithm; and the method further comprises recommending the built healthcare provider network to an insurance company.”

For additional information on this patent application, see: Felemban, Nawaf; Haddad, Christopher E. Systems And Methods For Building Healthcare Networks. U.S. Patent Application Number 20240145069, filed October 26, 2022 and posted May 2, 2024. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(20240145069)&db=US-PGPUB&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)