# US 2019/0172567 A1 · Methods, systems, and devices for determining mutli-party collocation

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

- Number: US 2019/0172567 A1
- Document ID: US20190172567A1
- Status: Published application
- Kind: A1
- Application number: 16/152,157
- Filed: 2018-10-04
- Published: 2019-06-06
- Earliest claimed: 2017-12-04
- Assignees: Cerner Innovation, Inc.
- Applicants: Cerner Innovation, Inc.
- Inventors: Daniel Aycock; Mario Manese; Colin Wurtz
- Web page: https://patentgrep.com/patent/US20190172567A1

## CPC

- G16H 10/60: for patient-specific data, e.g. for electronic patient records
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/00: Machine learning
- G16H 40/20: for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Methods, systems, and computer-readable media are provided for analyzing amounts of time spent by clinicians caring for patients. Clinician locations, patient locations, patient data, and clinician electronic health record activity may be used to facilitate generation of one or more predictive models. Such predictive models will provide, among other things, a decision support tool for scheduling patient-clinician appointments to minimize, at least partially, patient wait times.

## Claims

### Claim 1

A system comprising:
a wireless communication device for communicating a device identifier;
a plurality of wireless receivers, wherein each of the plurality of wireless receivers for receiving the device identifier communicated by the wireless communication device, determine a signal strength, and generate a time stamp associated with the receipt of the device identifier;
one or more processors; and
computer readable instructions that cause the one or more processors to perform operations comprising:
receiving the device identifier, the signal strengths, and the associated time stamp from a set of the plurality of wireless receivers;
based on the signal strengths and the associated time stamp, determining a location of the wireless communication device at a first time;
based on the location of the wireless communication device, determine that a provider associated with the wireless communication device is at the determined location; and
based on the time stamp, the determined location, and the provider determine that the provider and a patient are collocated.

### Claim 2 (depends on 1)

The system of claim 1, wherein the system further comprises a plurality of wireless communication devices, wherein the device identifier is different for each of the plurality of wireless communication devices.

### Claim 3 (depends on 1)

The system of claim 1, wherein the system further comprises an electronic health record system for storing medical information for a plurality of patients.

### Claim 4 (depends on 3)

The system of claim 3, wherein the computer readable instructions further cause the one or more processors to perform operations comprising:
retrieving appointment data associated with the patient, the appointment data comprising patient check-in time, patient room assignment, patient roomed time, and patient check-out time; and
based on the patient roomed time, the patient check-out time, and the patient room assignment determining that the patient was located at the location corresponding to the room assignment for the time points between the patient roomed time and the patient check-out time.

### Claim 5 (depends on 3)

The system of claim 3, wherein the system further comprises:
a plurality of remote user devices for accessing the electronic health record system; and
the computer readable instructions further cause the one or more processors to perform operations comprising:
identifying a first amount of time the provider accessed an electronic health record associated with the patient while the provider was collocated with the patient; and
identifying a second amount of time the provider accessed the electronic medical record associated with the patient while not collocated with the patient.

### Claim 6

One or more computer storage media having computer-executable instructions embodied thereon that, when executed, facilitate a method of time data analysis, the method comprising:
receiving time data, the time data comprising an amount of time a set of patients and a set of clinicians were collocated;
receiving patient data for the set of patients;
tuning a predictive model by supplying the predictive model the time data and the patient data corresponding to input variables in the predictive model;
receiving a clinic appointment request for a particular patient; and
in response to receiving the new appointment request utilizing the tuned predictive model to identify one or more recommended appointment times and durations.

### Claim 7 (depends on 6)

The computer storage media of claim 6, wherein determining a total amount of time the set of patients and the set of clinicians were collocated comprises:
determining a clinician location for each clinician of the set of clinicians at a plurality of time points;
determining a patient location at a set of the plurality of time points for each patient of the set of patients; and
identifying correlations between the clinician location and corresponding time points, and the patient location and corresponding time points, thereby identifying a set of time points in which at least one clinician is collocated with a patient.

### Claim 8 (depends on 7)

The computer storage media of claim 7, wherein determining the clinician location further comprises:
receiving a plurality of device identifiers, signal strengths, corresponding time stamps, and wireless receiver identifiers from a set of wireless receivers for receiving device identifiers broadcast by a set of devices, each device of the set of devices associated with a health care clinician;
segmenting the plurality of device identifiers, signal strengths, corresponding time stamps, and wireless receiver identifiers by device identifier thereby generating a per-device set of data for each device identifier comprising a plurality of signal strengths, corresponding time stamps, and wireless receiver identifiers; and
determining, for each per-device set of data, a location of the device associated with the device identifier at each time stamp based on the signal strengths and the wireless receiver identifiers.

### Claim 9 (depends on 7)

The computer storage media of claim 7, wherein determining the patient location further comprises:
receiving appointment data associated with the patient, the appointment data comprising patient check-in time, patient room assignment, patient roomed time, and patient check-out time; and
based on the patient roomed time, the patient check-out time, and the patient room assignment determining that the patient was located at the location corresponding to the room assignment for the time points between the patient roomed time and the patient check-out time.

### Claim 10 (depends on 6)

The computer storage media of claim 6, wherein the time data further comprises:
a first amount of time the set of clinicians accessed an electronic medical record associated with a patient of the set of patients while collocated with the patient; and
a second amount of time the set of clinicians accessed the electronic medical record associated with the patient while not collocated with the patient.

### Claim 11 (depends on 6)

The computer storage media of claim 6, wherein the input variables in the predictive model further comprise at least one of:
an appointment type;
a scheduled appointment duration;
a reason for visit;
a medication history;
a clinician identifier or name;
an indication of the patient's gender;
a weight associated with the patient;
a numerical indication of patient reported active problems;
an indication of emergency room or urgent care visits;
an indication of the patient's primary language; or
a representation of a chronologic position of the appointment.

### Claim 12 (depends on 6)

The computer storage media of claim 6, wherein utilizing the tuned predicative model comprises:
receiving a set of available appointment durations or a set of available appointment slots;
receiving patient data for the particular patient corresponding to the input variables of the predictive model; and
providing the predictive model the set of available appointment durations or the set of available appointment slots, and the patient data for the particular patient as inputs.

### Claim 13 (depends on 6)

The computer storage media of claim 6, wherein the new appointment request comprises a preferred time, a reason for the appointment, or the identity of the patient.

### Claim 14 (depends on 6)

The computer storage media of claim 6, wherein the input variables are determined by identifying Markov relationships between patient data and variations in patient appointments, variations in the duration of clinician-patient collocation, or variations in scheduled duration and actual duration.

### Claim 15

A computer implemented method for using broadcast identifiers to predictively generate optimized scheduling, the method comprising:
receiving a plurality of device identifiers, signal strengths, corresponding time stamps, and wireless receiver identifiers from a set of wireless receivers for receiving device identifiers broadcast by a set of devices, each device of the set of devices associated with a health care clinician;
segmenting the plurality of device identifiers, signal strengths, corresponding time stamps, and wireless receiver identifiers by device identifier thereby generating a per-device set of data for each device identifier comprising a plurality of signal strengths, corresponding time stamps, and wireless receiver identifiers;
determining, for each per-device set of data, a location of the device associated with the device identifier at each time stamp based on the signal strengths and the wireless receiver identifiers, thereby generating an enhanced per-device set of data;
receiving patient data for a set of patients associated with the health care provider from an electronic health records (EHR) system, wherein the set of patients are identified at least partially based on the location of the device at each time stamp;
training a predictive model by providing the clinician locations, patient location, and patient data as training inputs to the predictive model, wherein the predictive model generates appointment durations and appointment slots as output; and
receiving an appointment request for a particular patient;
receiving the particular patient's patient data from the EHR system;
scheduling the appointment for a length of time determined by the predictive model.

### Claim 16 (depends on 15)

The computer implemented method of claim 15, further comprising
receiving time data for the set of patients associated with the health care providers, wherein the time data represents the health care providers active time within the EHR of each patient of the set of patients;
identifying, based on the location information and the time data, a first subset of the time data and a second subset of the time data, the first set of time data representing the health care providers active time within the EHR and not in the presence of the patient, the second set of time data representing the health care providers active time within the EHR and within the presence of the patient; and
wherein training the predictive model further utilizes the first subset of time data as input for the model.

### Claim 17 (depends on 15)

The computer implemented method of claim 15, wherein scheduling the appointment comprises:
accessing a set of available appointment durations or a set of available appointment slots;
accessing patient data for the particular patient corresponding to the input variables of the predictive model; and
providing the predictive model the set of available appointment durations or the set of available appointment slots, and the patient data for the particular patient as inputs.

### Claim 18 (depends on 15)

The computer implemented method of claim 15, wherein the appointment request comprises a preferred time, a reason for the appointment, or the identity of the patient.

### Claim 19 (depends on 15)

The computer implemented method of claim 15, wherein the predictive model is a discrete event simulation model.

### Claim 20 (depends on 15)

The computer implemented method of claim 15, wherein the patient data for training the predictive model comprise at least one of:
an appointment type;
a scheduled appointment duration;
a reason for visit;
a medication history;
a clinician identifier or name;
an indication of the patient's gender;
a weight associated with the patient;
a numerical indication of patient reported active problems;
an indication of emergency room or urgent care visits;
an indication of the patient's primary language; or
a representation of a chronologic position of the appointment.