[go: up one dir, main page]

WO2018058189A1 - Système d'apprentissage automatique supervisé pour optimiser la présence de patients externes dans des cliniques - Google Patents

Système d'apprentissage automatique supervisé pour optimiser la présence de patients externes dans des cliniques Download PDF

Info

Publication number
WO2018058189A1
WO2018058189A1 PCT/AU2017/051061 AU2017051061W WO2018058189A1 WO 2018058189 A1 WO2018058189 A1 WO 2018058189A1 AU 2017051061 W AU2017051061 W AU 2017051061W WO 2018058189 A1 WO2018058189 A1 WO 2018058189A1
Authority
WO
WIPO (PCT)
Prior art keywords
attendance
clinic
patient
failure probability
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/AU2017/051061
Other languages
English (en)
Inventor
Jock Lawrie
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hro Holdings Pty Ltd
Original Assignee
Hro Holdings Pty Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from AU2016903938A external-priority patent/AU2016903938A0/en
Application filed by Hro Holdings Pty Ltd filed Critical Hro Holdings Pty Ltd
Priority to AU2017301078A priority Critical patent/AU2017301078A1/en
Publication of WO2018058189A1 publication Critical patent/WO2018058189A1/fr
Priority to AU2019201079A priority patent/AU2019201079A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0499—Feedforward networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition

Definitions

  • the present invention relates to a supervised machine learning system for optimising outpatient clinic attendance.
  • the present invention seeks to provide a way to optimising outpatient clinic attendance, which will overcome or substantially ameliorate at least some of the deficiencies of the prior art, or to at least provide an alternative.
  • FTA rates may be surprisingly affected by patient specific parameters and clinic specific parameters in unintuitive ways that dynamically change over time.
  • the system comprises a trained machine module.
  • the trained machine module is configured for having as input patient specific data and clinic specific data and calculating an attendance failure probability accordingly.
  • the system further comprises a machine learning module configured for training the trained machine module.
  • the machine learning module trains the trained machine module using historical training data comprising patient specific training data representing a plurality of patients, clinic specific training data representing a plurality of clinics and attendance training data representing attendance by the plurality of patients for each of the historical clinics.
  • the machine learning module is configured for optimising the accuracy of the attendance failure probability calculation of the trained machine.
  • the trained machine is configured for calculating an attendance probability (or probabilities) for the future clinics.
  • the future clinics are overbooked by a number of patients according to the calculated attendance failure probabilities to generate attendance probability optimised future clinics.
  • the supervised machine learning system described herein is configured for intelligently overbooking outpatient clinic schedules according to future FTA rates predicted by the system to generate attendance probability optimised future booking schedules.
  • the supervised machine learning module may identify seemingly unrelated and unintuitive effects on FTA rates. For example, the supervised machine learning system may have detected that patients with diabetes who smoke generally do not attend physiotherapy clinics on Mondays if they live more than 10 km from the hospital. As such, for a clinic scheduled for next Monday, a number of the outpatients may meet such criteria and therefore the FTA rate calculated by the supervised machine learning system may be higher than normal, such as 12%.
  • the calculated FTA rate may be utilised for calculating a number of overbookings to make for next Monday's clinic schedule.
  • the number of overbookings may be configured using an FTA percentage reduction setting.
  • the FTA percentage reduction setting 50% so as to aim to reduce the FTA rate by half.
  • the overbooking number for a clinic comprising 100 patients would be 6.
  • a supervised machine learning system for optimising outpatient clinic attendance, the system comprising: trained machine module configured for: having as input patient specific data and clinic specific data for a plurality of clinics; and calculating an attendance failure probability according to the input patient specific data and clinic specific data; a machine learning module configured for training the trained machine module, wherein the machine learning module trains the trained machine module using historical training data comprising patient specific training data representing a plurality of patients; clinic specific training data representing a plurality of clinics and attendance training data representing attendance by the plurality of patients of the respective historical clinics, and wherein the machine learning module is configured for optimising the accuracy of the attendance failure probability calculation of the trained machine; and a clinic schedule module for scheduling clinics, wherein, in use: for a future clinic schedule comprising patient specific data and clinic specific data, the trained machine is configured for calculating an attendance failure probability for the future clinic schedule; the clinic schedule module is configured for overbooking the future clinic schedule by a number
  • the system may be further configured for identifying a patient for overbooking for the attendance probability optimised clinic schedule.
  • Identifying the patient may comprise identifying the patient in accordance with wait time data for the patient.
  • Identifying the patient may comprise identifying the patient in accordance with an expected failure to attend rate for the patient determined by the trained machine module.
  • the patient may be identified according to a difference of the attendance failure probability for the patient and the attendance failure probability for the future clinic schedule.
  • the machine learning module may be configured for calculating the attendance failure probability for a particular time period of the future clinic schedule.
  • the trained machine module may comprise an artificial neural network.
  • the artificial neural network may comprise input nodes for patient specific data and clinic specific data and at least one output node for the attendance failure probability calculation.
  • the neural network may comprise at least one hidden layer between the input nodes and the upper nodes.
  • the machine learning module may be configured for adjusting weightings of the artificial neural network.
  • the machine learning module may be configured for adjusting the architecture of the artificial neural network.
  • Adjusting the architecture may comprise adjusting at least one of the number of neurons and number of hidden layers.
  • the patient specific data may comprise patient demographic data.
  • the patient in a graphic data may comprise demographic data including at least one of age, gender and residential address.
  • the patient specific data further may comprise health-related data.
  • the health-related data may comprise at least one of smoking status, pregnancy status, diabetes status and diagnosis.
  • the clinic specific data may comprise at least one of clinical speciality data, health practitioner specific data, and date time data.
  • the calculated attendance failure probability may be patient specific.
  • the machine learning module may be configured for calculating an attendance failure probability for each patient of the future clinic schedule.
  • the calculated attendance failure probability may be clinic schedule specific.
  • the machine learning module may be configured for calculating an attendance failure probability for the entire clinic schedule.
  • the attendance failure probability represents an attendance failure probability distribution.
  • Calculating an attendance failure probability for the entire clinic schedule may comprise calculating an attendance failure probability for each scheduled patient of the future clinic schedule and combining the attendance failure probabilities to calculate the attendance failure probability distribution.
  • the attendance failure probability for the entire clinic schedule may be calculated using the number of patients scheduled and the attendance failure probability distribution.
  • Figure 1 shows a supervised machine learning system for optimising outpatient clinic attendance in accordance with an embodiment of the present disclosure
  • Figure 2 shows an exemplary data flow for supervised machine learning system for optimising outpatient clinic attendance in accordance with an embodiment of the present disclosure.
  • Figure 1 shows a supervised machine learning system 100 for optimising outpatient clinic attendance.
  • the system 100 takes the form of a distributed web-server architecture and therefore comprises a web server 101 in operable communication with a plurality of client terminals 102 across the Internet 124. It should be noted that not all embodiments need necessarily be limited to this distributed web-server architecture and, in embodiments, the processing functionality may be implemented herein by way of a standalone computing device for example.
  • Each of the server 101 and client terminals 102 comprise a processor 109 for processing digital data.
  • a processor 109 for processing digital data.
  • a memory device 114 In operable communication with the processor across a system bus 108 is a memory device 114.
  • the memory device 114 is configured for storing digital data including computer program code instructions.
  • the processor 109 is configured for fetching these computer code instructions from the memory 114 for interpretation and execution and wherein data results from such execution may be stored within memory 114.
  • the memory device 114 of the server 101 has been shown as having been divided into logical computer program code instruction modules. These instruction modules may comprise an operating system 107.
  • the operating system 107 may be fetched by the processor 109 during the bootstrap phase.
  • the memory device 140 may further comprise a plurality of applications including a web server application 110 such as the Apache Web server application.
  • the applications may further comprise a hypertext preprocessor 106 and a database server application 111.
  • the web server application 110 upon receiving a web requests, is able to dynamically generate webpage responses utilising the hypertext preprocessor 106 and a database server 111.
  • the memory device 114 of the server 101 comprising a plurality of software modules 103 - 105 and respective database tables 115 - 118.
  • the software modules may comprise a trained machine module 104.
  • the trained machine module 104 has as input patient specific data and clinic specific data 116 and is configured for calculating attendance failure probabilities 117 accordingly.
  • the modules may further comprise a machine learning module 103 configured for training the trained machine module 104.
  • the machine learning module 103 trains the trained machine module 104 utilising historical training data comprising patient specific training data representing a plurality of patients, clinic specific training data representing a plurality of clinics and attendance training data representing attendance by the plurality of patients of the respective historical clinics.
  • the machine learning module 103 is configured for optimising accuracy of the attendance failure probability calculation 117 of the trained machine module 104.
  • the modules may further comprise a scheduler module 118 for scheduling the clinics.
  • the client terminal 102 may have a scheduler module 120 and store patient and clinic data 121 and attendance data 122 for the respective clinic schedules.
  • Each of the server 101 and the client terminal 102 may comprise a network interface 113 for sending and receiving data across the Internet 124.
  • each of the server 101 and client terminal 102 may comprise an I/O interface 112 for interfacing with various computer peripherals including human interface and data storage peripherals.
  • the client terminal 102 may comprise a display device 123 for the display of digital data including the clinic schedules described herein.
  • FIG. 6 illustrating the supervised machine learning system data flow 200 for optimising outpatient clinic attendance.
  • the data flow 200 comprises supervised machine learning 215 comprising the machine learning model 103 and the trained machine module 104.
  • the machine learning module 103 is configured for training the trained machine module 104.
  • the machine learning module 103 trains the trained machine module 100 for using historical training data 201 which may be obtained via database interface 205.
  • the historical training data 201 comprises patient specific training data 202 representing a plurality of patients, and clinic specific training data 203 representing a plurality of clinics.
  • the historical training data 201 comprises attendance training data 204 representing attendance by the plurality of patients of the respective historical clinics.
  • the trained machine module 104 is configured for outputting an attendance failure probability population 213.
  • the machine learning module 103 is configured for optimising the accuracy of the attendance failure probability calculation 213 of the trained machine module 104 with reference to the attendance training data 204.
  • the trained machine module 104 may take the form of an artificial neural network (ANN).
  • ANN artificial neural network
  • the machine learning module 103 may generate trained data 206 comprising a plurality of weightings 208 for waiting each of the neural paths of the artificial neural network.
  • the trained data 206 may further comprise architectural modification data 207 to modify and optimise the neural network, such as by modifying the number of neurons, number of layers et cetera.
  • the trained machine module 104 is configured for receiving a query 201 comprising a plurality of future clinics schedule 209.
  • the future clinics schedule data 209 may be obtained from the scheduler module 105.
  • the server 101 may be sent, or periodically retrieve, the future clinics schedule 209 from the respective client terminals 102 across the Internet 124.
  • the future clinics schedule data 209 may comprise patient specific data 211 and, in a preferred embodiment, clinic specific data 212.
  • the trained machine module 104 is configured for calculating an attendance failure probability 213 for the future clinics schedule 209.
  • the attendance failure probability 213 may take the form of an attendance failure probability 213 percentage or probability distribution.
  • a number of patient overbookings 217 is calculated according to the attendance failure probability 213. [84] The number of overbooking 217 may be modified according to attendance settings 216 as will be described in further detail below.
  • the scheduler module 105 may be configured with the number of patient overbookings 217 to generate an attendance probability optimised clinic schedule 219.
  • the optimise schedule 219 is intelligently optimised in accordance with the patient specific data 211 and clinic specific data 212 to dynamically and intelligently mitigate against FTA rates.
  • the optimised schedule 219 may be utilised as feedback 220 for further training the machine learning module 103.
  • a clinic group having a capacity of 50 clinics may have 60 patients booked (overbooked by 10 patients) but only 49 patients actually attend, so the group is under-attended by 1 (which is a good result). Without the overbooking, the actual attendance may have been around 40, an under-attendance of 10 (or 20%), which is a bad result.
  • the client terminal 102 may be operated by a hospital.
  • the hospital is staffed by many health practitioners each having an associated clinic with associated clinics.
  • the hospital comprises five doctors each having between 5 to 10 clinics available per clinic schedule.
  • the clinic scheduling may be maintained by the scheduler 120 of the client terminal 102. Furthermore, the client terminal 102 may record the attendance data 122 which may be subsequently utilised for training in the manner described herein.
  • the client terminal is configured for sending the historical attendance data 204 across the Internet 124 to the server 101 for optimisation purposes.
  • the historical training data 201 comprises at least one of patient specific data and clinic specific data and an attendance status indication as to whether an associated previous clinic was attended by the patient.
  • the patient specific data may comprise such data as patient demographic data, such as age, gender, residential address and the like.
  • the patient specific data may further comprise health related data, such as smoking status, pregnancy, diabetes status, diagnosis and the like.
  • the patient specific data may include the time the patient has spent on the waiting list for an clinic.
  • the waiting list time patient specific data may be utilised for prioritising patients when selecting patients for overbooking.
  • Further patient specific information may relate to the clinic type such as whether the clinic is a first clinic or a review (i.e. checkup) clinic.
  • the patient specific data may further comprise other relevant data for optimising the schedule.
  • the clinic specific data may comprise various data including the clinical speciality, such as neurology, cardiology and the like.
  • the clinic specific data may further comprise health practitioner specific data.
  • the attendance status indication indicates whether the historical clinic was attended to or not. Date and time specific information for the historical clinic may be recorded also, such as the day of the week, month, time of day and the like.
  • the historical training data 201 may indicated that a 42-year-old male smoker with diabetes and living at a residential address 15 km from the hospital had failed to attend a checkup clinic at 2 PM on Monday, 26 September 2017 with cardiologist Dr John Smith.
  • the supervised machine learning 205 may comprise the trained machine module 104 being trained by the machine learning module 103 using the historical training data 201.
  • the artificial neural network may comprise input nodes for the patient specific data 211 and clinic specific data 212, a number of hidden layers and a predicted attendance failure probability 213 output node.
  • the weighting of the artificial neural network may be trained. Specifically, as the historical training data 201 is fed into the machine learning module 103, the machine learning module 103 adjusts the weights of the artificial neural network so as to reduce the output error of the calculated predicted attendance failure probability 213 when compared to the input attendance training data 204.
  • the architecture of the artificial neural network may be fixed. However, during the training phase, the machine learning module 103 may additionally optimise the architecture of the artificial neural network.
  • the output expected attendance failure probability 213 may either be specific to a particular clinic or to a group of clinics. For the former, the output attendance failure probability 213 may represent, for example, that for Dr John Smith, the expected attendance failure probability 213 for a particular clinic day/period would be 10% for the specific patient or clinic.
  • the expected attendance failure probabilities 213 could be for a group of clinics such as a group of clinics comprising five doctors wherein, for the group of clinics, there would be a combined attendance failure probability 213 of 10% for a particular day/clinic period.
  • the machine learning module 103 may calculate a probability of non-attendance for each scheduled patient. Then, these probabilities are combined to obtain the probability of non- attendances for each clinic, that is, the probability of 0 non-attendances, the probability of 1 non- attendances, the probability of 2 non-attendances, the probability of K non-attendances, where K is the number of patients booked into the clinic.
  • clinics can be grouped wherein the same approach applies to obtain non-attendance failure probabilities for the group. I.e., the probability of 0 non-attendances, the probability of K non-attendances, where K is the total number of patients booked into the group.
  • K the total number of patients booked into the group.
  • a clinic group has 1 or more clinics the distinction between a clinic and a group of clinics is immaterial, because the former is a special case of the latter.
  • Combining the patient-level probabilities into clinic or group-level probabilities may use the Poisson-Binomial distribution, calculated via fast fourier transforms.
  • the trained artificial neural network may have detected that patients with diabetes who smoke generally do not attend their physiotherapy clinics on Mondays if they live more than 10 km from the hospital. As such, for a clinic scheduled for next Monday, a number of the outpatients may meet such criteria and therefore the attendance failure probability 213 output by the artificial neural network would be higher than normal, such as 12%.
  • the attendance failure probability 213 rate 37 may be generated in the format of an attendance failure probability distribution.
  • the output attendance failure probability 213 may be utilised for calculating a number of overbookings 217 to make.
  • the number of overbookings 217 may be determined in accordance with attendance setting 216 which may include the above-described failure to attend percentage reduction setting or the over attendance rate.
  • the failure to attend percentage reduction setting may be configured at 50% so as to aim to reduce the FTA rate by half. As such, for the calculated FTA rate 37 of 12% for next Monday, the overbooking number 217 for a clinic comprising 100 clinics would be 6.
  • the scheduler module 105 would then update the outpatient schedule 118.
  • the server 101 would update the schedule of the client terminal 102 remotely such as by having access to the schedule.
  • the server 101 may send the number of overbookings 217 to the client terminal 102 such that the client terminal 102 is able to update the schedule itself.
  • the system 1 may implement a "dummy schedule" representing a schedule comprising outpatients who may attend any of the available health practitioners in a given clinic group when available.
  • the additional six outpatients may be allocated to the dummy schedule such that, at any time, should an outpatient fail to attend a particular clinic, any of the outpatients of the dummy schedule may be allocated to the relevant slot.
  • the clinics may run for predetermined time periods, such as between 9 AM and 12 PM.
  • the system 1 need only allocate the overbookings to the dummy schedule corresponding to this time period wherein those outpatients allocated to the dummy schedule may be required to wait for an available timeslot during this time period.
  • the attendance failure probability 213 may be calculated for each daily clinic schedule.
  • the attendance failure probabilities 213 may have greater time period granularity so as to aim to reduce the waiting period for outpatients on the dummy schedule.
  • the artificial neural network may have further calculated that the above exemplary diabetic outpatient group is more likely to fail to attend clinics after lunch on Mondays.
  • the scheduling module 2.3 would request that the overbooked outpatients attend the clinic at the relevant time after lunch such as around 2 PM.
  • the system 1 may calculate patient specific overbooking data.
  • the system 1 when overbooking, the system 1 is configured for selecting specific outpatients for overbooking. In one embodiment, such selection may be in accordance with waiting times wherein those outpatients being on the waiting list for longer are favoured.
  • the system 1 may select outpatients similarly in accordance with attendance failure probability 213 rates. For example, when overbooking the schedule to account for the above-described diabetics who fail to attend clinics on Mondays, the system 100 may favour outpatients having differing patient specific data so as to, for example, not select further diabetic outpatients who may themselves similarly failed to attend the allocated clinics.
  • the invention may be embodied using devices conforming to other network standards and for other applications, including, for example other WLAN standards and other wireless standards.
  • Applications that can be accommodated include IEEE 802.11 wireless LANs and links, and wireless Ethernet.
  • wireless and its derivatives may be used to describe circuits, devices, systems, methods, techniques, communications channels, etc., that may communicate data through the use of modulated electromagnetic radiation through a non-solid medium. The term does not imply that the associated devices do not contain any wires, although in some embodiments they might not. In the context of this document, the term “wired” and its derivatives may be used to describe circuits, devices, systems, methods, techniques, communications channels, etc., that may communicate data through the use of modulated electromagnetic radiation through a solid medium. The term does not imply that the associated devices are coupled by electrically conductive wires.
  • processor may refer to any device or portion of a device that processes electronic data, e.g., from registers and/or memory to transform that electronic data into other electronic data that, e.g., may be stored in registers and/or memory.
  • a "computer” or a “computing device” or a “computing machine” or a “computing platform” may include one or more processors.
  • the methodologies described herein are, in one embodiment, performable by one or more processors that accept computer-readable (also called machine-readable) code containing a set of instructions that when executed by one or more of the processors carry out at least one of the methods described herein.
  • Any processor capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken are included.
  • a typical processing system that includes one or more processors.
  • the processing system further may include a memory subsystem including main RAM and/or a static RAM, and/or ROM.
  • a computer-readable carrier medium may form, or be included in a computer program product.
  • a computer program product can be stored on a computer usable carrier medium, the computer program product comprising a computer readable program means for causing a processor to perform a method as described herein.
  • the one or more processors operate as a standalone device or may be connected, e.g., networked to other processor(s), in a networked deployment, the one or more processors may operate in the capacity of a server or a client machine in server-client network environment, or as a peer machine in a peer-to-peer or distributed network environment.
  • the one or more processors may form a web appliance, a network router, switch or bridge, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine.
  • each of the methods described herein is in the form of a computer- readable carrier medium carrying a set of instructions, e.g., a computer program that are for execution on one or more processors.
  • embodiments of the present invention may be embodied as a method, an apparatus such as a special purpose apparatus, an apparatus such as a data processing system, or a computer-readable carrier medium.
  • the computer-readable carrier medium carries computer readable code including a set of instructions that when executed on one or more processors cause a processor or processors to implement a method.
  • aspects of the present invention may take the form of a method, an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.
  • the present invention may take the form of carrier medium (e.g., a computer program product on a computer-readable storage medium) carrying computer-readable program code embodied in the medium.
  • Carrier Medium
  • the software may further be transmitted or received over a network via a network interface device.
  • the carrier medium is shown in an example embodiment to be a single medium, the term “carrier medium” should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more sets of instructions.
  • the term “carrier medium” shall also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by one or more of the processors and that cause the one or more processors to perform any one or more of the methodologies of the present invention.
  • a carrier medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media.
  • a device A connected to a device B should not be limited to devices or systems wherein an output of device A is directly connected to an input of device B. It means that there exists a path between an output of A and an input of B which may be a path including other devices or means.
  • Connected may mean that two or more elements are either in direct physical or electrical contact, or that two or more elements are not in direct contact with each other but yet still co-operate or interact with each other.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Epidemiology (AREA)
  • Educational Administration (AREA)
  • Primary Health Care (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Databases & Information Systems (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Pathology (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

L'invention concerne un module d'apprentissage automatique supervisé servant à optimiser la présence de patients externes dans des cliniques, qui surréserve de manière dynamique un calendrier de clinique selon des paramètres spécifiques de patients et spécifiques de cliniques afin d'optimiser la présence de patients externes dans des cliniques. Le système comprend un module de machine entraînée. Le module de machine entraînée est configuré pour recevoir l'entrée des données spécifiques de patients et des données spécifiques de cliniques, et pour calculer en conséquence une probabilité de défaut de présence. Le système comprend en outre un module d'apprentissage automatique configuré pour entraîner le module de machine entraînée. Le module d'apprentissage automatique entraîne le module de machine entraînée à l'aide de données d'apprentissage historiques incluant des données d'apprentissage spécifiques de patients représentant une pluralité de patients, des données d'apprentissage spécifiques de cliniques représentant une pluralité de cliniques et des données d'apprentissage de présence représentant la présence de la pluralité de patients pour chacune des cliniques historiques. Une fois que la machine entraînée a été optimisée de cette manière, lors de l'utilisation, pour une pluralité de cliniques futures, la machine entraînée est configurée pour calculer une probabilité (ou des probabilités) de présence pour les cliniques futures. Ensuite, les cliniques futures sont surréservées par un certain nombre de patients selon les probabilités de défaut de présence calculées pour générer des cliniques futures optimisées en termes de probabilité de présence.
PCT/AU2017/051061 2016-09-28 2017-09-28 Système d'apprentissage automatique supervisé pour optimiser la présence de patients externes dans des cliniques Ceased WO2018058189A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
AU2017301078A AU2017301078A1 (en) 2016-09-28 2017-09-28 A supervised machine learning system for optimising outpatient clinic attendance
AU2019201079A AU2019201079A1 (en) 2016-09-28 2019-02-15 A supervised machine learning system for optimising outpatient clinic attendance

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
AU2016903938A AU2016903938A0 (en) 2016-09-28 A machine learning system for optimising outpatient clinic attendance
AU2016903938 2016-09-28

Publications (1)

Publication Number Publication Date
WO2018058189A1 true WO2018058189A1 (fr) 2018-04-05

Family

ID=61762355

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/AU2017/051061 Ceased WO2018058189A1 (fr) 2016-09-28 2017-09-28 Système d'apprentissage automatique supervisé pour optimiser la présence de patients externes dans des cliniques

Country Status (2)

Country Link
AU (2) AU2017301078A1 (fr)
WO (1) WO2018058189A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110175943A (zh) * 2019-05-23 2019-08-27 威比网络科技(上海)有限公司 用于智能化课程管理的方法、装置和系统以及存储介质
US20200402663A1 (en) * 2019-06-18 2020-12-24 Canon Medical Systems Corporation Medical information processing apparatus and medical information processing method
IL281746A (en) * 2021-03-22 2022-10-01 Mor Research Applic Ltd Machine learning models for designating subjects for the purpose of treatment and/or assessment

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11593729B2 (en) 2020-03-13 2023-02-28 International Business Machines Corporation Cognitive tuning of scheduling constraints

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150242819A1 (en) * 2012-10-31 2015-08-27 Smart Scheduling, LLC Systems and methods for improving scheduling inefficiencies using predictive models
US20160253462A1 (en) * 2015-02-27 2016-09-01 Koninklijke Philips N.V. Novel open-access scheduling system that optimizes healthcare delivery system operation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150242819A1 (en) * 2012-10-31 2015-08-27 Smart Scheduling, LLC Systems and methods for improving scheduling inefficiencies using predictive models
US20160253462A1 (en) * 2015-02-27 2016-09-01 Koninklijke Philips N.V. Novel open-access scheduling system that optimizes healthcare delivery system operation

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
ALAEDDINI ET AL.: "A hybrid prediction model for noshows and cancellations of outpatient appointments", IIE TRANSACTIONS ON HEALTHCARE SYSTEMS ENGINEERING, vol. 5, no. 1, 16 March 2015 (2015-03-16), pages 14 - 32, XP055494955 *
ALAEDDINI ET AL.: "A probabilistic model for predicting the probability of no-show in hospital appointments", HEALTH CARE MANAGEMENT SCIENCE, vol. 14, no. 2, 1 February 2011 (2011-02-01), pages 146 - 157, XP019896569 *
HUANG ET AL.: "Using Artificial Neural Networks to Establish a Customer-cancellation Prediction Model", PRZEGLAD ELEKTROTECHNICZNY, vol. 89, no. 1b, 2013, pages 178 - 180, XP055494959 *
SAMORANI ET AL.: "Outpatient appointment scheduling given individual day-dependent no-show predictions", EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, vol. 240, no. 1, 2015, pages 245 - 257, XP055268703 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110175943A (zh) * 2019-05-23 2019-08-27 威比网络科技(上海)有限公司 用于智能化课程管理的方法、装置和系统以及存储介质
CN110175943B (zh) * 2019-05-23 2021-10-01 威比网络科技(上海)有限公司 用于智能化课程管理的方法、装置和系统以及存储介质
US20200402663A1 (en) * 2019-06-18 2020-12-24 Canon Medical Systems Corporation Medical information processing apparatus and medical information processing method
US11682491B2 (en) * 2019-06-18 2023-06-20 Canon Medical Systems Corporation Medical information processing apparatus and medical information processing method
IL281746A (en) * 2021-03-22 2022-10-01 Mor Research Applic Ltd Machine learning models for designating subjects for the purpose of treatment and/or assessment
IL281746B1 (en) * 2021-03-22 2023-04-01 Mor Research Applic Ltd Machine learning models for designation of subjects for treatment and/or evaluation
IL281746B2 (en) * 2021-03-22 2023-08-01 Mor Research Applic Ltd Machine learning models for designating subjects for the purpose of treatment and/or assessment

Also Published As

Publication number Publication date
AU2019201079A1 (en) 2019-03-07
AU2017301078A1 (en) 2018-04-12

Similar Documents

Publication Publication Date Title
US20240281207A1 (en) Intelligent Personal Agent Platform and System and Methods for Using Same
Albahri et al. Real-time fault-tolerant mHealth system: Comprehensive review of healthcare services, opens issues, challenges and methodological aspects
US11096577B2 (en) Proactive patient health care inference engines and systems
Kovalchuk et al. Distributed data-driven platform for urgent decision making in cardiological ambulance control
Malik et al. Intelligent load-balancing framework for fog-enabled communication in healthcare
Wang et al. High reliable real-time bandwidth scheduling for virtual machines with hidden Markov predicting in telehealth platform
AU2019201079A1 (en) A supervised machine learning system for optimising outpatient clinic attendance
US20190013095A1 (en) A patient procedure schedule throughput optimiser supervised machine learning system
WO2017015579A1 (fr) Prévision d'exigence pour des services de soins de santé
Safdari et al. A multi agent based approach for prehospital emergency management
Paul et al. Inventory management strategies for mitigating unfolding epidemics
CN116130074A (zh) 一种候诊等待时间预测方法、装置及存储介质
Rajagopal et al. AI augmented edge and fog computing for Internet of Health Things (IoHT)
CN111368412B (zh) 用于护理需求预测的仿真模型构建方法和装置
Landolfi et al. Intelligent value chain management framework for customized assistive healthcare devices
EP3841552B1 (fr) Traitement et analyse de systèmes de données de prestataires de soins de santé
Hosseinzadeh et al. Enhancing healthcare IoT systems for diabetic patient monitoring: Integration of Harris Hawks and grasshopper optimization algorithms
CN113724824B (zh) 慢性病患者随访方法、装置、计算机设备及可读存储介质
US20230268061A1 (en) Systems and Methods to Track and Automate Home Care Management
CN111354449B (zh) 长期护理策略分配方法、装置、计算机设备和存储介质
Roy et al. Public health in India: Leveraging technology for a brighter future
Sitlong et al. Hybrid Dynamic Programming Healthcare Cloud-Based Quality of Service Optimization
AU2017202982B1 (en) A patient procedure schedule throughput optimiser supervised machine learning system
US20250336521A1 (en) Cell manufacturing management platform using machine learning
US11735324B2 (en) Two-way questionnaire generation for medical communication

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2017301078

Country of ref document: AU

Date of ref document: 20170928

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17854251

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17854251

Country of ref document: EP

Kind code of ref document: A1