IP Library Granted Patent US 12,046,340
Granted Patent B2
US 12,046,340 · App. 16/930,525 · Granted Jul 23, 2024

System and method for fulfilling prescriptions

Inventors: Samir Pattanaik (Bangalore, IN); Laxmi Swetha Guntakandla (Bangalore, IN); Bhaskar Mishra (San Mateo, CA)
Assignee: Walmart Apollo, LLC
G16H20/10G06N3/08G06N7/00G16H30/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,046,340
App. No.
16/930,525
Granted
Jul 23, 2024
Kind
B2
Abstract

A control circuit applies images obtained from the mobile electronic device to a trained model and responsively receiving prescription information. The prescription information indicates contents of the prescription, whether a prescription transfer is needed, and whether a prescription refill is needed. The control circuit determines an action based upon an analysis of the prescription information. The action is taken to electronically instruct a human or robot to fulfill the prescription. The action includes, when a prescription transfer is needed, transmitting an electronic message to a second pharmacy currently holding the medical prescription of the user.

Claims (49)

1. A system, comprising:

an electronic network;

an electronic database, the electronic database including a mathematical model;

a control circuit, the control circuit coupled to the electronic network and the electronic database, the control circuit being configured to:

train the mathematical model to recognize prescription information by applying training images to the model;

wherein the training extracts data from training images with a confidence score, and the confidence score determines whether to rely on the training images for the training, fine-tune the model subsequent to training, and/or to discard selected ones of the training images and wherein the training images contain images of insurance cards, membership cards, and prescriptions;

wherein the membership cards and prescriptions include indicia that the membership cards and prescriptions are from or are related to competing pharmacies that are different from an existing pharmacy currently fulfilling medical prescriptions for a user;

wherein the mathematical model is a neural network comprising a structure including neural layers, and the neural layers are changed from a first structure or a first state to a second structure or a second state because of the training such that the second structure or state is automatically configured to recognize and discern information from the images of the insurance cards, membership cards and prescriptions when applied to the neural network, and the first structure or state is unable to recognize and discern information from the images of the insurance cards, membership cards, and prescriptions;

wherein subsequent to the training:

a transaction to fulfill and potentially transfer an existing prescription of the user from the existing pharmacy to a pharmacy associated with a store in which the user is currently located is initiated by a camera on a mobile electronic device of the user being actuated to obtain one or more images, the existing prescription previously having been serviced by the existing pharmacy, the one or more images including images of an existing medical prescription of the user, an existing membership card of the user, the existing membership card relating to or identifying the store the user is currently located, and/or an insurance card of the user, the mobile electronic device transmitting the one or more images to the control circuit via the electronic network after obtaining the images;

the control circuit receives the one or more images via the electronic network while the user is in a store;

the control circuit applies the one or more images obtained from the mobile electronic device to the trained model and automatically and responsively receives from the trained model prescription information, the prescription information indicating contents of the prescription, whether a prescription transfer is needed and, if needed an identity of the existing pharmacy, and whether a prescription refill is needed;

the control circuit determines an action based upon an analysis of the prescription information; and

based on the determined action the control circuit electronically controls a robot to retrieve medication from a storage area to fulfill the prescription;

wherein the action includes, when a prescription transfer is needed, transmitting an electronic message to the existing pharmacy currently holding the medical prescription of the user to transfer the existing medical prescription to the pharmacy of the store in which the user is currently located and the existing medical prescription is subsequently electronically transferred to the pharmacy of the store where the user is currently located.

2. The system of claim 1 , wherein the one or more images are a series of images of different items, a single image with multiple items, or a video.

3. The system of claim 1 , wherein the model is a convolutional neural network.

4. The system of claim 1 , wherein a computer application program on the mobile electronic device presents to the user an option as to which type of items to scan.

5. The system of claim 1 , wherein the mobile electronic device is a smart phone, a tablet, a cellular phone, a laptop computer, or a personal computer.

6. A method, comprising:

training, by a control circuit via an electronic network, a mathematical model to recognize prescription information by applying training images to the mathematical model, the mathematical model implemented on an electronic database coupled to the electronic network;

wherein the training by the control circuit includes extracting data from training images with a confidence score, wherein the confidence score is used by the control circuit to determine whether to rely on the training images for the training, fine-tune the mathematical model subsequent to the training, and/or to discard selected ones of the training images, and wherein the training images contain images of insurance cards, membership cards, and prescriptions;

wherein the membership cards and the prescriptions include indicia that the membership cards and the prescriptions are from or are related to competing pharmacies that are different from an existing pharmacy currently fulfilling medical prescriptions for a user;

wherein the mathematical model is a neural network comprising a structure including neural layers, and the neural layers are changed from a first structure or a first state to a second structure or a second state because of the training such that the second structure or state is automatically configured to recognize and discern information from the images of the insurance cards, the membership cards and the prescriptions when applied to the neural network, and the first structure or state is unable to recognize and discern information from the images of the insurance cards, the membership cards, and the prescriptions;

initiating, by the control circuit via the electronic network, a transaction to fulfill and potentially transfer an existing prescription of the user from the existing pharmacy to a pharmacy associated with a store in which user is currently located based on a camera on a mobile electronic device of the user being actuated to obtain one or more images, the existing prescription previously having been serviced by the existing pharmacy, the one or more images obtained including images of an existing medical prescription of the user, an existing membership card of the user, the existing membership card relating to or identifying the store in which the user is currently located, and/or an insurance card of the user;

receiving, by the control circuit via the electronic network, the one or more images obtained from the mobile electronic device while the user is in the store;

applying, by the control circuit, the one or more images obtained from the mobile electronic device to the trained model;

responsive to applying the one or more images obtained to the trained model, automatically receiving from the trained model prescription information, the prescription information indicating contents of the prescription, whether a prescription transfer is needed, whether a prescription refill is needed and an identity of the existing pharmacy;

analyzing, by the control circuit, the prescription information received from the trained model;

determining, by the control circuit, an action based upon an analysis of the prescription information; and

based on the determined action, electronically controlling, by the control circuit, a robot to retrieve medication from a storage area to fulfill the prescription;

wherein the action includes, when a prescription transfer is needed, transmitting an electronic message to the existing pharmacy currently holding the medical prescription of the user to transfer the existing medical prescription of the user to the pharmacy of the store in which the user is currently located and the existing medical prescription of the user is subsequently electronically transferred to the pharmacy of the store where the user is currently located.

7. The method of claim 6 , wherein the one or more images are a series of images of different items, a single image with multiple items, or a video.

8. The method of claim 6 , wherein the model is a convolutional neural network.

9. The method of claim 6 , wherein a computer application program on the mobile electronic device presents to the user an option as to which type of items to scan.

10. The method of claim 6 , wherein the mobile electronic device is a smart phone, a tablet, a cellular phone, a laptop computer, or a personal computer.

11. A device, comprising:

a control circuit, the control circuit coupled to an electronic network and an electronic database, the control circuit being configured to:

train a model implemented on the electronic database to recognize prescription information by applying training images to the model, wherein the training extracts data from training images with a confidence score, using the confidence score to determine whether to rely on the training images for the training and to fine-tune the model subsequent to training, and wherein the training images contain images of insurance cards, membership cards, and prescriptions;

wherein the membership cards and the prescriptions include indicia that the membership cards and the prescriptions are related to competing pharmacies that are different from an existing pharmacy currently fulfilling medical prescriptions for a user;

an image capture device configured to obtain one or more images associated with a user in a location and transmit the one or more images to the control circuit while the user is in the location;

the control circuit receives the one or more images while the user is in the location and apply the one or more images to the trained model to automatically receivs from the trained model prescription information, the prescription information indicating contents of the prescription, whether a prescription transfer is needed and, if needed an identity of the existing pharmacy, and whether a prescription refill is needed;

the control circuit determines an action based upon an analysis of the prescription information; and

based on the determined action the control circuit electronically controls a robot to retrieve medication from a storage area to fulfill the prescription;

wherein the action includes, when a prescription transfer is needed, transmitting an electronic message to the existing pharmacy currently holding the medical prescription of the user to transfer the existing medical prescription to the pharmacy of the store in which the user is currently located and the existing medical prescription is subsequently electronically transferred to the pharmacy of the store where the user is currently located.

12. The device of claim 11 , wherein the one or more images are a series of images of different items, a single image with multiple items, or a video.

13. The device of claim 11 , wherein the model is a convolutional neural network.

14. The device of claim 11 , wherein the device is an in-store kiosk.

15. The device of claim 11 , wherein the device is a smart phone, a tablet, a cellular phone, a laptop computer, or a personal computer.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 31, 2020
From: PATTANAIK, SAMIR; GUNTAKANDLA, LAXMI SWETHA; MISHRA, BHASKAR
To: WALMART APOLLO, LLC
Reel/Frame 053365/0838 →
Priority Claims (1)
IN 201941028718 · Jul 17, 2019 · national
Continuity (2)
Provisional Application 62898076 · Sep 10, 2019
Related Publication 20210020287A1 · Jan 21, 2021