# US 12,002,575 B1 · Machine learning system and method for attendance risk mitigation

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,002,575 B1
- Document ID: US12002575B1
- Status: Granted
- Kind: B1
- Application number: 18/360,231
- Filed: 2023-07-27
- Granted: 2024-06-04
- Earliest claimed: 2023-07-27
- Assignees: Yale University
- Applicants: Yale University
- Inventors: Peter Kahn; Walter S Mathis
- Web page: https://patentgrep.com/patent/US12002575B1

## CPC

- G16H 40/20: for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
- G06Q 10/1093: Calendar-based scheduling for persons or groups
- G16H 10/60: for patient-specific data, e.g. for electronic patient records
- G16H 80/00: ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

A computer-implemented method of increasing appointment attendance comprises providing a processor and a non-transitory memory including computer program code for one or more programs, the memory and the computer program code configured to, with the at least one processor, perform steps comprising obtaining an appointment data structure, comprising an attendee and a corresponding appointment for the attendee, obtaining a set of population data, obtaining a set of appointment data, obtaining a set of external data, obtaining environmental data, inferring, using the population data, the appointment data, the external data, and the environmental data in a machine learning algorithm, a probability that the attendee will attend the appointment, and when the probability of attendance is below a threshold, performing a mitigation step to increase the probability that the attendee will attend the appointment. A system for increasing appointment attendance and a non-transitory computer-readable medium containing computing instructions are also described.

## Claims

### Claim 1

A computer-implemented method of increasing appointment attendance, comprising:
providing at least one processor; and at least one non-transitory memory including computer program code for one or more programs, the at least one non-transitory memory and the computer program code configured to, with the at least one processor, perform steps comprising:
obtaining an appointment data structure, comprising an attendee and a corresponding appointment for the attendee;
obtaining a set of electronic medical record (EMR) data related to the attendee, wherein the set of EMR data is in at least two different formats;
standardizing the at least two different formats of the set of EMR data;
at a first time:
obtaining a set of population data related to the attendee;
obtaining a set of appointment data related to the appointment;
obtaining a set of external data;
obtaining environmental data related to the region around a location of the appointment or the attendee at the first time;
inferring, using the standardized EMIR data, the population data, the appointment data, the external data, and the environmental data in a machine learning algorithm by the at least one processor and the at least one non-transitory memory, a probability that the attendee will attend the appointment; and
when the probability of attendance is below a threshold, performing a mitigation step selected from a first group of mitigation steps to increase the probability that the attendee will attend the appointment;
at a second time after the first time:
obtaining a second set of population data related to the attendee;
obtaining a second set of appointment data related to the appointment;
obtaining a second set of external data;
obtaining environmental data related to the region around a location of the appointment or the attendee at the second time;
inferring, using the standardized EMR data, the second set of population data, the second set of appointment data, the second set of external data, and the environmental data in the machine learning algorithm by the at least one processor and the at least one non-transitory memory, a probability that the attendee will attend the appointment; and
when the probability of attendance is below a threshold, performing a mitigation step selected from a second group of mitigation steps different from the first group of mitigation steps, to increase the second probability that the attendee will attend the appointment.

### Claim 2 (depends on 1)

The method of claim 1, wherein the first time is at least one day prior to a date of the appointment.

### Claim 3 (depends on 2)

The method of claim 2, wherein the first time is selected from at least one month prior to the date of the appointment, at least one week prior to the date of the appointment, or at least three days prior to the day of the appointment.

### Claim 4 (depends on 1)

The method of claim 1, wherein the environmental data comprises weather data.

### Claim 5 (depends on 1)

The method of claim 1, wherein the external data comprises economic data.

### Claim 6 (depends on 5)

The method of claim 5, wherein the economic data comprises economic data related to the region around the attendee.

### Claim 7 (depends on 5)

The method of claim 5, wherein the economic data comprises global or national economic data.

### Claim 8 (depends on 1)

The method of claim 1, further comprising the step of inferring a communication preference of the attendee.

### Claim 9 (depends on 8)

The method of claim 8, wherein the first mitigation step comprises a communication step selected from written, telephonic, text message, or electronic mail communication based on the inferred communication preference of the attendee.

### Claim 10 (depends on 1)

The method of claim 1, wherein the first mitigation step comprises providing transportation assistance to the attendee, the transportation assistance selected from transportation maps, a ride sharing credit, a public transportation credit, or an ambulance dispatch.

### Claim 11 (depends on 1)

The method of claim 1, wherein the first mitigation step comprises converting the appointment to a virtual appointment or changing a location of the appointment to a home or place of business of the attendee.

### Claim 12 (depends on 1)

The method of claim 1, wherein the appointment data comprises data selected from a provider of the appointment, a nature of the appointment, whether the appointment is related to a cancer, one or more conditions the attendee holds, one or more diagnoses the attendee is currently being treated for, whether the appointment is telemedicine or in-person, or whether the appointment is an initial consultation or a second opinion.

### Claim 13

A computer-implemented method of increasing appointment attendance, comprising:
providing at least one processor; and at least one non-transitory memory including computer program code for one or more programs, the at least one non-transitory memory and the computer program code configured to, with the at least one processor, perform steps comprising:
obtaining a set of electronic medical record (EMR) data related to an attendee in at least two different data formats;
standardizing the at least two different data formats of the set of EMR data;
obtaining a set of past appointment data structures, each comprising an attendee identity, a past appointment, and an attendance status of the past appointment;
obtaining a set of external data related to the attendees in the set of past appointment data structures;
obtaining a set of regional environmental data related to a region around a location of the appointment or the attendees in the set of past appointment data structures from a time of each past appointment;
training a machine learning algorithm with the standardized EMR data, the set of past appointment data structures, the set of external data, and the set of environmental data;
at a first time:
obtaining a pending appointment data structure comprising the attendee and a corresponding appointment for the attendee;
obtaining a set of appointment data related to the appointment;
obtaining environmental data related to the region around the location of the appointment or the attendee in the pending appointment data structure at the first time;
inferring, using the trained machine learning algorithm, a probability that the attendee will attend the appointment; and
when the probability of attendance is below a threshold, performing a first mitigation step selected from a first group of mitigation steps to increase the probability that the attendee will attend the appointment;
at a second time after the first time:
obtaining the pending appointment data structure comprising the attendee and the corresponding appointment for the attendee;
obtaining a set of appointment data related to the appointment;
obtaining second environmental data related to the region around the location of the appointment or the attendee in the pending appointment data structure at the second time;
inferring, using the trained machine learning algorithm, a second probability that the attendee will attend the appointment; and
when the second probability of attendance is below a second threshold, performing a second mitigation step selected from a second group of mitigation steps different from the first group of mitigation steps, to increase the probability that the attendee will attend the appointment.

### Claim 14 (depends on 13)

The method of claim 13, further comprising the steps of:
after the appointment in the pending appointment data structure has passed, recording an attendance status of the attendee at the appointment; and
further training the machine learning algorithm with the pending appointment data structure and the attendance status.

### Claim 15 (depends on 13)

The method of claim 13, wherein the environmental data and the regional environmental data comprise weather data.

### Claim 16 (depends on 13)

The method of claim 13, further comprising the step of inferring whether a mobility factor selected from age, disability status, or travel time is contributive above a predetermined threshold.

### Claim 17 (depends on 16)

The method of claim 16, wherein the first mitigation step comprises providing transportation assistance to the attendee, the transportation assistance selected from transportation maps, a ride sharing credit, a public transportation credit, an ambulance dispatch, or an ambulette dispatch when the mobility factor is above the predetermined threshold.

### Claim 18 (depends on 14)

The method of claim 14, wherein the attendance status of the past appointment in the appointment data structures is selected from attending, no-show, late, rescheduled, converted to telemedicine, or cancelled surgery.

### Claim 19 (depends on 13)

The method of claim 13, wherein the machine learning algorithm is a random forest.

### Claim 20

A system for increasing appointment attendance, comprising:
a non-transitory computer-readable storage medium with instructions stored thereon, which when executed by a processor, perform steps comprising:
obtaining an appointment data structure, comprising an attendee and a corresponding appointment for the attendee;
obtaining a set of electronic medical record (EMR) data in at least two different formats related to the attendee;
standardizing the at least two different formats of the set of EMR data;
at a first time:
obtaining a first set of data related to the appointment;
obtaining a first set of external data;
obtaining first environmental data related to the region around a location of the appointment or the attendee at the first time;
inferring, using the standardized EMR data, the first set of data related to the appointment, the first set of external data, and the first environmental data in a machine learning algorithm, a first probability that the attendee will attend the appointment; and
when the first probability of attendance is below a first threshold, recommending a first mitigation step selected from a first group of mitigation steps to increase the probability that the attendee will attend the appointment at a second time after the first time:
obtaining a second set of data related to the appointment;
obtaining a second set of external data;
obtaining second environmental data related to the region around a location of the appointment or the attendee at the second time;
inferring, using the standardized EMR data, the second set of data related to the appointment, the second set of external data, and the second environmental data in the machine learning algorithm, a second probability that the attendee will attend the appointment; and
when the second probability of attendance is below a second threshold, recommending a second mitigation step selected from a second group of mitigation steps different from the first group of mitigation steps to increase the probability that the attendee will attend the appointment.

### Claim 21 (depends on 20)

The system of claim 20, the instructions further comprising inferring the first probability that the attendee will attend the appointment using a trained machine learning algorithm executed on the processor.

### Claim 22 (depends on 20)

The system of claim 20, wherein the non-transitory computer-readable medium further comprises at least a portion of the database of an EMR system comprising the EMR data.

### Claim 23 (depends on 20)

The system of claim 20, the instructions further comprising performing the first mitigation step, the first mitigation step comprising sending one computer-generated electronic communication message to the attendee.

### Claim 24

A non-transitory computer-readable storage medium having stored thereon one or more program instructions which, when executed by one or more processors, cause an apparatus to at least perform the following operations:
obtain an appointment data structure, comprising an attendee and a corresponding appointment for the attendee;
obtain a set of electronic medical record (EMR) data related to the attendee, wherein the set of EMR data is in at least two different formats;
standardize the at least two different formats of the set of EMR data;
at a first time:
obtain a first set of population data related to the attendee;
obtain a first set of appointment data related to the appointment;
obtain a first set of external data;
obtain first environmental data related to the region around a location of the appointment or the attendee at first time;
infer, using the standardized EMR data, the first set of population data, the first set of appointment data, the first set of external data, and the first environmental data in a machine learning algorithm, a first probability that the attendee will attend the appointment; and
when the first probability of attendance is below a first threshold, recommending a first mitigation step selected from a first group of mitigation steps to increase the probability that the attendee will attend the appointment, at a second time after the first time:
obtain a second set of population data related to the attendee;
obtain a second set of appointment data related to the appointment;
obtain a second set of external data;
obtain second environmental data related to the region around a location of the appointment or the attendee at the second time;
infer, using the standardized EMR data, the second set of population data, the second set of appointment data, the second set of external data, and the second environmental data in the machine learning algorithm, a second probability that the attendee will attend the appointment; and
when the second probability of attendance is below a second threshold, recommending a second mitigation step selected from a second group of mitigation steps different from the first group of mitigation steps to increase the probability that the attendee will attend the appointment.

### Claim 25 (depends on 24)

The non-transitory computer readable storage medium of claim 24, further comprising at least one database including at least a subset of the first set of population data, the first set of appointment data, the first set of external data, and the first environmental data stored thereon.

### Claim 26 (depends on 24)

The non-transitory computer-readable storage medium of claim 24, the program instructions further configured to cause the apparatus to perform at least part of the first or second mitigation step via an automated process.