FIELD OF THE INVENTION

[0001] The present invention relates to management systems and in particular to Artificially Intelligent management systems.

[0002] The invention has been developed primarily as an Artificially Intelligent management system for medical services and will be described hereinafter with reference to this application. However, it will be appreciated that the invention is not limited to this particular field of use.

BACKGROUND OF THE INVENTION

[0003] Any discussion of prior art throughout the specification should in no way be considered as an admission that such prior art is widely known or forms part of the common general knowledge.

[0004] Healthcare is a vitally important field of endeavor. Medical science seeks to improve the quality of healthcare to try to ensure that human beings (and animals) are able to live healthy lives. Healthcare is concerned with the prevention, diagnosis, treatment, and cure of physical and mental ailments in healthcare consumers (i.e., "patients").

[0005] Healthcare involves various specialised fields of medical practice including general practice, cardiology, neurology, oncology, pediatrics, gynecology, physiology, dermatology, psychology, psychiatry, allied practice, and others. Healthcare also involves the organisation and management of various healthcare practitioners (i.e., "doctors") and support staff, healthcare consumers, and healthcare providers (i.e., healthcare businesses, and other relevant entities). The scope of a healthcare system can range from international, national, state, to local. Quality healthcare organisation and management supports quality outcomes for healthcare consumers.

[0006] Information and data form significant aspects of medical science as well as healthcare. Information and data support quality healthcare outcomes by making healthcare organisation and management more efficient and effective. Information and

SUBSTITUTE SHEET (RULE 26) Communication Technology-based ("ICT") healthcare management systems make use of information and data to implement healthcare management in the form of healthcare management systems, which is specialised client relationship management software.

[0007] Existing healthcare management systems address niche segments of healthcare and do not aggregate and leverage data in a way that could be more useful to healthcare practitioners, healthcare providers, and consumers. The main advantages of existing healthcare management systems are planning and retrospection.

[0008] Artificial Intelligence and Machine Learning is increasingly being applied to help solve various problems and to enhance the performance of various systems including healthcare.

Machine Learning normally involves feeding usually known inputs ("independent variables") into a Machine Learning Algorithm to train the Algorithm until a Machine Learning Model emerges capable of providing desired outputs ("dependent variables") when it is fed unknown inputs.

[0009] Various types of Machine Learning Algorithms exist and one or more Machine Learning Algorithms can be used to create a bespoke Machine Learning architecture for solving a particular problem. Typical problems a Machine Learning Algorithm can solve are regression and classification. Machine Learning Algorithms are often classified as Linear and Non- Linear. The application of Machine Learning in healthcare management is currently limited by being applied in niche segments of healthcare.

[0010] It is desirable to overcome or ameliorate at least one of the disadvantages of the prior art, or to provide a useful alternative.

[0011] It is desirable to provide a healthcare management system that integrates disparate and niche healthcare management systems into a single, real-time capable, holistic healthcare management system that aggregates and leverages data in a way that is useful to healthcare practitioners and consumers and results in good healthcare outcomes.

SUMMARY OF THE INVENTION

[0012] According to the invention there is provided an artificially intelligent management system for healthcare services, the system including: one or more software structures implementing

SUBSTITUTE SHEET (RULE 26) one or more of the following: one or more task management; one or more consultation management; one or more patient record management; one or more allergies and/or reactions monitor; one or more current medication management; one or more prescription management; one or more master treatment assessment plan management; one or more drug recommendation system; one or more dashboard management; one or more administration task bar; one or more communication task bar; one or more treatment assessment plan risk assessment; and one or more Rapid Antigen Test monitor; one or more databases in operative communication with the one or more software structures; and the one or more software structures and/or one or more databases being in operative communication with any one or more of the following elements: one or more machine learning model; one or more mathematical equation; one or more algorithm; and one or more function, wherein any one or more of the foregoing elements is adapted to enhance the operation of the one or more software structures.

[0013] Preferably, the one or more software structures further implement one or more of the following: one or more mental health management; one or more orthopedics recommender system; one or more anesthetics risk management; one or more gynecology recommender system algorithm; and one or more obesity medicine/bariatric surgery recommender system.

[0014] Preferably, the one or more task management system software structures is in operative communication with one or more forecasting machine learning model; the one or more consultation management software structures is in operative communication with one or more genetic algorithm machine learning model; the one or more patient record management software structures is in operative communication with one or more data validation check function; the one or more allergies and/or reactions monitor software structures is in operative communication with one or more data validation check function; the one or more current medication management software structures is in operative communication with one or more predictive machine learning model; the one or more prescription management software structures is in operative communication with one or more predictive machine learning model; the one or more master treatment assessment plan management software structures is in operative communication with one or more predictive machine learning model; the one or more drug recommendation system software structures is in operative communication with one or more of the following: one or more

SUBSTITUTE SHEET (RULE 26) recommender system algorithm; one or more predictive machine learning model; one or more virtual stock function; and one or more mathematical equation; the one or more dashboard management software structures is in operative communication with one or more predictive machine learning model; the one or more administration task bar software structures is in operative communication with one or more prompting function; the one or more communication task bar software structures is in operative communication with one or more input prompting function; the one or more treatment assessment plan risk assessment software structures is in operative communication with one or more predictive machine learning model; and the one or more Rapid Antigen Test monitor software structures is in operative communication with one or more machine vision model.

[0015] Preferably, the one or more mental health management software structures is in operative communication with one or more or more of the following: one or more predictive machine learning models; and one or more recommender system algorithms; the one or more orthopedics recommender system is in operative communication with one or more of the following: one or more recommender system algorithms; one or more classification machine learning model; and

[0016] one or more predictive machine learning model; the one or more anesthetics risk management is in operative communication with one or more of the following: one or more predictive machine learning models; and one or more recommender system algorithms; the one or more gynecology recommender system is in operative communication with one or more of the following: one or more recommender system algorithm; one or more predictive machine learning models; and one or more classification machine learning models; and the one or more obesity medicine/bariatric surgery recommender system is in operative communication with one or more of the following: one or more recommender system algorithm; and one or more predictive machine learning models.

[0017] Preferably, the one or more task management system software structures is in operative communication with one or more Polynomial Regression Algorithm-based forecasting machine learning models; the one or more consultation management software structures is in operative communication with one or more Optimization Genetic Algorithm machine learning models; the one or more current medication management software structures is in

SUBSTITUTE SHEET (RULE 26) operative communication with one or more Polynomial Regression Algorithm-based predictive machine learning models; the one or more prescription management software structures is in operative communication with one or more Polynomial Regression Algorithm-based predictive machine learning models; the one or more master treatment assessment plan management software structures is in operative communication with one or more Polynomial Regression Algorithm-based predictive machine learning models; the one or more drug recommendation system software structures is in operative communication with one or more of the following: one or more Collaborative Filteringbased recommender system algorithms; one or more Polynomial Regression Algorithmbased predictive machine learning models; and one or more mathematical equation:

Sim a, b) = . = - ; the one or more dashboard management software z(r _a p-r _a ) j S{r _a f ₎ -r ₀ - )'~ structures is in operative communication with one or more Polynomial Regression Algorithm-based predictive machine learning models; the one or more treatment assessment plan risk assessment software structures is in operative communication with one or more Polynomial Regression Algorithm-based predictive machine learning models; and the one or more Rapid Antigen Test monitor software structures is in operative communication with one or more machine vision model, the machine vision model including one or more of the following: one or more Optical Character Scanner; one or more Object Recognition; and one or more Image Recognition.

[0018] Preferably, the one or more mental health management software structures is in operative communication with one or more or more of the following: one or more Random Forest Algorithm-based predictive machine learning models; one or more Convolutional & Recurrent Neural Network-based predictive machine learning models; one or more Jaccard Similarity-based recommender system algorithms; and one or more Recurrent Neural Network-based predictive machine learning models; the one or more orthopedics recommender system is in operative communication with one or more of the following: one or more Collaborative Filtering-based recommender system algorithms; one or more Convolutional Neural Network-based classification machine learning models; one or more Recurrent Neural Network-based predictive models; and the one or more anesthetics risk management is in operative communication with one or more of the following: one or more Polynomial Regression Algorithm-based predictive machine learning models; and one or more Jaccard Similarity-based recommender system algorithms; one or more K Nearest Neighbor Algorithm-based recommender system algorithms; the one or more gynecology

SUBSTITUTE SHEET (RULE 26) recommender system is in operative communication with one or more of the following: one or more Jaccard Similarity-based recommender system algorithms; one or more K Nearest Neighbor Algorithm-based recommender system algorithms; one or more Collaborative Filtering-based recommender system algorithms; one or more Convolutional Neural Network-based classification machine learning models; one or more Recurrent Neural Network-based classification machine learning models; the one or more obesity medicine/bariatric surgery recommender system is in operative communication with one or more of the following: one or more Collaborative Filtering-based recommender system algorithms; and one or more Polynomial Regression Algorithm-based predictive machine learning models.

[0019] Preferably, the machine learning management sub-system wherein the sub-system includes: one or more online real-time machine learning model feature database; one or more offline machine learning model feature database; one or more repositories for one or more machine learning models; one or more machine learning model performance feedback loops; one or more machine learning model drift feedback loops; one or more alarm managers; one or more machine learning model re-training scheduler; and one or more machine learning model lineage tracker.

[0020] In an aspect of the invention, the system is implemented on one or more distributed/cloud systems.

[0021] Preferably, there is provided one or more GUI for interacting with one or more system software structures.

[0022] In an aspect of the invention there is provided a Machine-readable code containing one or more sets of instructions for implementing the system.

[0023] In an aspect of the invention there is provided a method is performed by executing one or more sets of steps for implementing an artificially intelligent management system for medical services.

BRIEF DESCRIPTION OF THE DRAWINGS

SUBSTITUTE SHEET (RULE 26) [0024] A preferred embodiment of the invention will now be described, by way of example only, with reference to the accompanying drawings in which:

[0025] Figure 1 is a system diagram depicting the basic architecture of the invention.

DETAILED DESCRIPTION OF THE INVENTION

[0026] Referring to the drawing, Figure 1 shows a cloud-based artificially intelligent management system 2 for health care services accessed by users 4A and 4B by means of a computing device 8 via a cloud network 10. The management system 2 has a database 6 containing system data and user information. The database 6 is in operative communication 12 with various software structures which are depicted in Figure 1 as blocks.

[0027] The management system can be integrated with a larger healthcare services management system (not shown) or in parallel to other healthcare services management systems.

[0028] In a preferred embodiment, the management system is integrated with a healthcare services management application (not shown) such as a client relationship management application.

[0029] Throughout this specification, and unless otherwise indicated references to the "X-axis" indicate independent variables/events (i.e., "inputs") while references to the "Y-axis" indicate dependent variables/predictions/classifications/recommendations (i.e., "outputs"). The outputs can be stored in the database 6 as necessary or required the operation of the system 2.

[0030] The healthcare services making up the management system can be broadly grouped in to general services and specific services. A description of the general services follows. The specific services will be described after the general services.

[0031] A task management system software structure 14 is in operative communication 12 with a Polynomial Regression Algorithm-based forecasting machine learning model 16 adapted (i.e., "trained") to predict the amount of time needed to complete different medical tasks by medical practitioners (i.e., "Doctors") to aid in time management. The inputs (i.e.,

SUBSTITUTE SHEET (RULE 26) "independent variables", "X-axis") used to train the model 16 and the output predictions (i.e., "independent variables", "outputs", "Y-axis") include:

The type of task being completed, such as a risk assessment, could act as an independent variable (X-axis);

Time taken for individual doctors to complete particular types of tasks (X-axis); Time taken for doctors to complete particular types of tasks in general (X-axis), starting point for time allocation of tasks but the model gains accuracy in predicting the appropriate amount of time allocation for tasks as it learns the behaviour of the individual doctors it's catering to over time;

Experience level of doctor (X-axis);

Individual doctor completing said task (X-axis); and Time allocation for each task (Y-axis).

[0032] A consultation management software structure 18 is in operative communication 12 with an Optimization Genetic Algorithm machine learning model 20 adapted to continuously improve time management and successful meeting scheduling. The data inputs to train the model 20 and the output predictions include:

Cancellation Rate of meetings (X-axis);

Reschedule rate of meetings (X-axis);

Future availability and calendar of participants - including geographic location of appointments for efficiency (X-axis);

Consultant's history and rates (X-axis); and

Newly Determined generated most effective schedule (Y -axis).

[0033] A patient management software structure 22 is in operative communication 12 with a data validation check function 24 adapted to continuously check the validity of a consumer's recent records with previously confirmed accurate records of said consumer, as well as validating the information of the database by performing a data matching check with any other databases that have been implemented. The function 24 input/output data points include:

Consumer ID number (X-axis);

Consumer recent records (X-axis);

History of consumer's previous medical records assessments (X -axis);

SUBSTITUTE SHEET (RULE 26) Compares data in recent consumer record with previously confirmed data of consumer to confirm accuracy (Y-axis);

Current doctor's Confirmation of consumer's recent record and data to then be updated to the consumer's medical records; and

Consumer records received from third party databases (X-axis).

[0034] The data validation check 24 is implemented as a software function that is continuously run in an infinite loop to activate as soon as a new entry is detected.

[0035] An allergies and/or reactions monitor software structure 26 is in operative communication with a data validation check function 28 adapted to to continuously check the validity of a consumer's recent allergy and drug interaction record with previously confirmed accurate records and history of said consumer, as well as validating the information of the database by performing a data matching check with other databases that have been implemented. The function 28 input/output data points include:

Consumer ID number (X-axis);

Consumer's recent records (X-axis);

History of consumer's Allergy and Drug Interaction records assessments (X -axis); Consumer's Current Lifestyle Habits, smoking, diet, Alcohol usage etc., (X-axis); Compares data in recent consumer record with previously confirmed data of consumer to confirm accuracy (Y-axis); and

Current doctor's Confirmation of consumer's recent Allergy and Drug interaction record and data to then be updated to the patient's medical records.

[0036] The data validation check 28 is implemented as a software function that is continuously run in an infinite loop to activate as soon as a new entry is detected.

[0037] A current medication management software structure 30 is in operative communication 12 with a Polynomial Regression Algorithm-based predictive machine learning model 32 adapted to accurately predict the risk of using newly prescribed medication of a particular consumer based on the consumer's medical history, allergy and drug interacting history. The data inputs to train the model 32 and the output predictions include:

Various Static Factors of patient including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history],

SUBSTITUTE SHEET (RULE 26) pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

Current Medication properties and side-effects (X-axis);

Verified Medical history of patients (X-axis);

Verified Allergy History of Patient (X-axis);

Verified History of drug interactions of patient (X-axis);

Predicted Risk score (Categorised to High, Medium and Low risk) from the assessment of currently prescribed medication of a patient (Y-axis);

Details of Potential risks on the patient in short-, medium-, and long-term usage (Y- axis);

Doctors inputs on potential risks that the algorithm missed could also act as another independent data point (X-axis); and

Current patient lifestyle choices such as, Smoking, diet, weekly number of hours spent exercising, Sleep Patterns, Work hours, Illicit and alcohol usage, etc., (X-axis).

[0038] A prescription management software structure 34 is in operative communication 12 with a Polynomial Regression Algorithm-based predictive machine learning model 36 adapted to predict the risk of using the current list of prescribed medications on each particular patient based on the patient's medical history, allergy and drug interacting history. The data inputs to train the model 36 and the output predictions include:

Static Factors of patient including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

All Medications, properties and side-effects (X-axis);

Verified History of drug interactions of patient (X-axis);

Risk score of each medication individually (X-axis);

SUBSTITUTE SHEET (RULE 26) Predicted Risk score (Categorised to High, Medium and Low risk) from the assessment of currently prescribed medications of a patient (Y-axis);

Details of Potential risks on the patient in short-, medium- and long-term usage (Y- axis);

Doctor inputs on potential risks that the algorithm missed could also act as another independent data point (X-axis);

Current patient lifestyle choices such as, Smoking, diet, weekly number of hours spent exercising, Sleep Patterns, Work hours, Illicit and alcohol usage etc., (X-axis); Updated Lifestyle choices, as seen from consumer diary (X-axis); and

Effects of prescriptions on consumer, as seen from consumers Diary and Pain questionnaire (X-axis).

[0039] A master treatment assessment plan management software structure 38 is in operative communication 12 with a Polynomial Regression Algorithm-based predictive machine learning model 40 adapted to predict the risk/trajectory of a consumer developing further complications if given a current treatment plan. The data inputs to train the model 40 and the output predictions include:

Various Static Factors of consumer including, Sex, Height, family medical history [first-degree & second-degree familial history], pharmacogenetics, which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

Current treatment plan detail, properties and side-effects (X-axis);

Treatment Compliance rate (X-axis);

Verified Medical history of consumer (X-axis);

Verified Allergy History of consumer (X-axis);

Verified History of drug interactions of consumer (X-axis);

Consumer history of risk assessments and risk profile (X-axis);

Predicted Risk score (Categorized to High, Medium and Low risk) from the assessment of currently implemented treatment plan of consumer (Y-axis); Details of Potential risks on the consumer in the short, medium and long term (Y- axis);

SUBSTITUTE SHEET (RULE 26) Doctor's inputs on potential risks that the algorithm missed could also act as another independent data point (X-axis);

Current lifestyle choices such as, Smoking, diet, weekly number of hours spent exercising, Sleep Patterns, Work hours, etc.;

Clinical Questionnaire measurement/Assessment Tool Outputs from current prescription for response with 55% reduction, recovery 67% reduction, remission seen as 50% increase in symptoms (X-axis); and

Updated lifestyle choices as seen in consumer treatment diary e.g., change in smoking habit, change in alcohol intake etc., (X-axis).

[0040] A drug recommendation system software structure 42 is in operative communication 12 with a Collaborative Filtering-based recommender system algorithm 44 adapted to recommend a set of medication choices to include in the consumer's treatment plan based on the consumer's drug interaction history and potential risk of developing long-term effects. The algorithm 44 input/output data points include:

Various Static Factors including, consumer Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use, weekly number of hours spent exercising, Sleep Patterns, Work hours, which could all act as independent variables (X-axis);

All current Medications, properties and side-effects (X-axis);

Recommended set of medications (Y-axis);

Updated Lifestyle choices, as seen from consumer diary (X-axis); and Effects of prescriptions on consumer, as seen from Patient diary and Pain questionnaire (X-axis).

[0041] The drug recommendation system software structure 42 is in operative communication 12 with a Polynomial Regression Algorithm-based predictive machine learning model 46 adapted to predict the risk of using the newly recommended medication in combination with medications already prescribed to the consumer and the individual consumer's risk profile and other factors. The data inputs to train the model 46 and the output predictions include:

SUBSTITUTE SHEET (RULE 26) Various Static Factors of consumer including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

All Medications, properties and side-effects (X-axis);

Verified Medical history of patients (X-axis);

Verified Allergy History of Patient (X-axis);

Verified History of drug interactions of patient (X-axis);

Properties and Side effect of newly recommended medication (X-axis);

Risk of using new recommended medication in combination with already prescribed medication;

Predicted Risk score of using the new combination of medication being used on this particular consumer (Categorized to High, Medium and Low risk) from the assessment of currently prescribed medications of a patient (Y-axis);

Details of Potential risks on the consumer in short-, medium- and long-term usage (Y-axis);

Doctor's inputs on potential risks that the algorithm missed could also act as another independent data point (X-axis);

Current patient lifestyle choices such as, smoking, diet, weekly number of hours spent exercising, Sleep Patterns, Work hours, Illicit and alcohol usage, etc. (X-axis); Updated Lifestyle choices, as seen from consumer diary and prompts (X-axis); and Effects of prescriptions on patient, as seen from patient's diary and Pain questionnaire (X-axis).

[0042] The drug recommendation system software structure 42 is in operative communication 12 with a virtual stock function (not shown) adapted to keep a virtual tally of the amount of medication remaining with the patient through their updates and diary. The function input/output datapoints include:

Medications prescribed to patient (X-axis);

Amount allocated to patient for each medication (X-axis);

Store a running total (X-axis);

SUBSTITUTE SHEET (RULE 26) Updates from patient on dosages taken each day, which is then subtracted from running total;

Prompt user when medications are running low

Time; and

Display current estimated running stock of each medication (Y-axis).

[0043] The virtual stock check (not shown) is implemented as a software function activated every time a certain criterion is met. For example, one trigger would be a time the consumer updates their treatment diary, such as when the consumer confirms daily prescribed dosage has been met. The function would then update the running count of how much of the medication the user should have left if they have been honest/accurate with their daily medication intake.

[0044] The drug recommendation system software structure 42 is also in operative communication a mathematical equation function 48 adapted to calculate the consumer's treatment compliance rate based on their interaction with the system, the formular being as follows:

[0045] The mathematical function 48 input/output datapoints include:

Log of amount of times consumer has used the app (X-axis);

Track date and time consumer logged taking their prescribed daily dosage & daily medication reconciliation (X-axis);

Track compliance with in-app treatment activities eg., physiotherapy exercises [time track of video watched and physical activity trackers from devices for movement], logging of food intake in food diary, psychological interventions eg., mindfulness [time track of video watched and physical activity trackers from devices eg., HR, Rate or respiration, blood pressure, "breathe app" on iwatch & apple devices] (X-axis);

Compare the time the log was updated with when the consumer claimed to have taken the medication to then generate a likelihood of consumer claims being accurate (X-axis);

Frequency of missed dosage (X-axis);

Potential activity/compliance inhibiting factors, school, work, pleasure, schedule, etc. (X-axis);

SUBSTITUTE SHEET (RULE 26) Calculate a compliance score from all variables, and submit this to the doctor (Y- axis); and

Updated diary updates the score.

[0046] Details of the equation 48 can be found in Calculating Medication Compliance, Adherence and Persistence in Administrative Pharmacy Claims Databases August 2008 Pharmaceutical Programming 1(1):13-19.

[0047] The mathematical equation 48 is used calculate the consumer's compliance rate by using the appropriately weighted independent variables (i.e., "inputs"). The equation 48 is implemented in a software function. The software function is triggered in one or more time interval (i.e., daily) to generate a new compliance rate from an updated consumer diary.

[0048] A dashboard management software structure 50 is in operative communication 12 a Polynomial Regression Algorithm-based predictive machine learning model 52 adapted model to predict the amount of time needed to complete different medical care tasks by doctors to aid in time management. The data inputs to train the model and the output predictions include:

The type of task being completed, such as a risk assessment, could act as an independent variable (X-axis);

Time taken for individual doctors to complete particular types of tasks (X-axis);

Time taken for doctors to complete particular types of tasks in general (X-axis), starting point for time allocation of tasks but the model becomes more accurate in predicting the appropriate amount of time allocation for tasks as it learns the behaviour of the individual doctor it is recommending to; and

Experience level of doctor (X-axis); and Time allocation for each task (Y-axis).

[0049] An administrative task bar software structure 54 is in operative communication 12 with a prompting function 56 adapted to remind the doctor to send updates in the patient treatments to other doctors who are involved in the consumer's treatment. The function attributes include:

Log any changes made by the doctor to the consumer's treatment;

SUBSTITUTE SHEET (RULE 26) For each change activate the prompt for the doctor to notify all other relevant doctors;

If prompt is accepted redirect doctor to email; and

Log the frequency of the doctor accepting the prompt and who was notified.

[0050] A communication task bar software structure (not shown) is in operative communication 12 with an input prompting function (not shown) adapted to remind the consumer to to update their diary with relevant information, and this update is then sent to the doctor who can also keep track of the consumer's progress and treatment. The function attributes are:

Send a notification/prompt to consumer at time intervals, ie., every 24 hours;

Redirect the client to the diary for new entries, for example, Medications used that day, effects felt during day, symptom changes, etc.;

Log the frequency of notification being answered; and Submit updates to the doctor.

[0051] The prompting function is implemented as a software function that is activated every time a certain criterion is met, in this case being a time interval (i.e., a day), informing the user to update their Treatment diary.

[0052] A treatment assessment plan risk assessment software structure 58 is in operative communication 12 with a Polynomial Regression Algorithm-based predictive machine learning model 60 adapted to predict the risk of using a particular treatment plan for a particular medical condition for a consumer, to thus aid doctors in completing risk assessments for long term complications associated with the condition. The data inputs to train the model and the output predictions include:

Various Static Factors including, consumer Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

The proposed treatment plan details, properties and side-effects (X-axis); Treatment Compliance rate (X-axis);

SUBSTITUTE SHEET (RULE 26) Customer history of risk assessments and risk profile (X-axis);

Predicted Risk score (Categorized to High, Medium and Low risk) from the assessment of currently implemented treatment plan of the consumer (Y-axis); Details of Potential risks for the consumer in the short, medium and long-term (Y- axis);

Doctor's inputs on potential risks that the algorithm missed could also act as another independent data point (X-axis);

Current lifestyle choices such as, Smoking, diet, weekly number of hours spent exercising, Sleep Patterns, Work hours can be used (X-Axis); and

Use of Clinical Assessment Tools [questionnaires] to be used to qualify current symptoms, to assist with realising values for response, recovery, remission & relapse (X-axis);

Updated lifestyle choices as seen in consumer's treatment diary e.g., change in smoking habit, change in alcohol intake etc., (X-axis).

[0053] A Rapid Antigen Test monitor software structure 62 is in operative communication 12 with a machine vision model-based Optical Character Scanner system 64 adapted to uniquely identify a Rapid Antigen Test Device & record as "used/unused" to prevent re-use. The Optical Character Scanner system datapoints and/or attributes include:

An image from a camera

Recording specific identifiers and barcode seen on test, test-kit wrapping, or test box Determining the number of tests allocated to those identifiers

Cross-referencing the number of times these specifiers have been recorded vs the number of tests allocated to this package.

[0054] The Optical Character Scanner system 64 is sourced from the Tesseract platform, but any other suitable Optical Character Scanner-type system can be used.

[0055] A Rapid Antigen Test monitor software structures 62 is in operative communication 12 with a machine vision model-based Object 66 and Image Recognition system 68 adapted for validating reporting accuracy [Positive/Negative/Inconclusive] as a means to protect against false reporting. The data inputs to train the model(s) 66 and 68 and the output predictions include:

Camera images; and

SUBSTITUTE SHEET (RULE 26) Database of RAT test images.

[0056] A description of the specific healthcare services making up the management system are now described.

[0057] A mental health (or i.e., "psychiatry") management software structure 70 is in operative communication 12 with a Random Forest Algorithm-based predictive machine learning model adapted 72 to predict he mental state deteriorations/fluctuations Mood disorders [Depression - Mania] & Factors affecting individual mood fluctuations. The data inputs to train the model 72 and the output predictions include:

DASS-21 rating scales [determinant of severity of symptoms], (X-axis);

Daily Alcohol intake, (X-axis);

Daily Marijuana consumption [in g/day], (X-axis);

Daily Quality of sleep [ based on an 8hours/ day requirement], (X-axis);

Daily Exercise [>20 minutes per day, >3 times per week];

Treatment Compliance Rate, (X axis);

Time (Y-Axis) [0, 2 weeks, 6 weeks, 12 weeks, 6 months, 9 months, 12 months, etc.,]; and

Look for the intersection of factors affecting mental state decline.

[0058] The mental health management software structure 70 is in operative communication 12 with a Convolutional Neural Network-based predictive machine learning model 74 adapted to predict effective treatments based on doctors' activities, consumer activities, doctor's decisions, and getting main factors influencing effective treatments. The data inputs to train the model 74 and the output predictions include:

Treatment Compliance factors [ plotted as total percentage compliance rate] including medication compliance/self-cessation of medication/treatment due to external [work schedule conflicts] & internal intersecting factors [adverse effects, non-remission of symptoms] i.e., looking at factors affecting non-compliance behaviours, (X-axis);

HP response to triggered events as a result of negative reporting by HC as either negative outcomes or significant non-compliance measured as an increase in symptoms on monitoring [using clinical assessment measures], decrease in treatment compliance rates, significant changes in physical symptoms (X-axis);

SUBSTITUTE SHEET (RULE 26) The actions taken by doctor in response to these triggers - treatment optimisation through changes in medication regimens, recommendations/utilisation of psychotherapy, use of lifestyle interventions [exercise, community services support, etc., (X-axis);

Review of the engagement with the Safety Plan Features & algorithm, (X-axis); and Comparison of doctor treatment behaviours against resultant outcomes i.e., factors leading to doctors with higher positive treatment outcomes versus factors leading to negative consumer outcomes. Consideration of any beneficial actions that deviated from treatment guidelines, (Y-axis).

[0059] The mental health management software structure 72 is also in operative communication with a Recurrent Neural Network-based predictive machine learning model 76 adapted to predict time taken for the development of metabolic syndrome in patients using psychotropic medication. The data inputs to train the model 76 and the output predictions include:

Measures in metabolic monitoring table, (X-axis);

Various Static Factors of patient including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, which could all act as independent variables, (X-axis);

Time, (Y-Axis);

Use of deviations in the measures based on threshold [in the table below], (X-Axis)

SUBSTITUTE SHEET (RULE 26)

Prediction of probability/time to develop metabolic syndrome [is 3 of diabetes, hypertension, hypercholesterolemia, obesity, etc.] from the onset of treatment; and Built-in triggers for prompting interventions by the doctor.

[0060] The mental health management software 70 structure is in operative communication 12 with a Jaccard Similarity-based recommender system algorithm 78 adapted to determine the the preferred medication based on symptom profile, adjusted for comorbidities & previous adverse effects profile of treatments. Efficacy wis determined by the consumer's treatment compliance rate and remission of symptoms. Remission will be measured with standardised clinical measurement tools such as DASS-21 rating scales, for response with 50% reduction, 50-67% is continuation of episode, recovery 67% reduction, remission seen as 50% increase in symptoms. Symptom profile & side effects have a direct correlation with the treatment compliance. The attributes of the recommender system 78 include:

Identifying most appropriate antidepressant based on symptoms profile [selected by 2+ most debilitating symptoms] & known neurotransmitter profile of the medications - including consideration of use of polarised effects profile to positively influence symptoms remission [e.g., use of a medication with known sedating properties to manage insomnia sleep disturbance];

Filter based on contraindications to use [comorbidities e.g., liver function impairment, renal function impairment, heart function impairment, etc.,]; Filter based on risk for unwanted side effects profile [using Allergy/Adverse Reactions profile algorithm];

Result of [5] most suitable medications as options for treatment, ranked by relevance to achieving prospective treatment outcome. The goal is to measure the distance between past consumer's symptom profiles and new patients with a similar profile and recommend treatments used that were successful for past patients.

SUBSTITUTE SHEET (RULE 26) [0061] An orthopedics recommender system software structure 80 is in operative communication 12 with a Collaborative Filtering-based recommender system algorithm 82 adapted to recommend the most effective non-surgical intervention treatment plans for mobility hindering injuries. The algorithm 82 input/output data points include:

Various Static Factors of consumer including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, which could all act as independent variables (X-axis);

Diary of ailments and their intensity (X-axis);

Effectiveness of treatment plans calculated from pain questionnaire recorded during the time interval and Sheehan's Disability Scales (X-axis);

Time (X-axis);

Current treatment plan being implemented (X-axis);

Remission of symptoms over time (X-axis);

Treatment compliance rate (X-axis); and

A newly Recommended treatment plan, Physiotherapy sessions/exercises, Analgesia dosage, etc., (Y-axis).

[0062] The orthopedics recommender system software structure 80 is in operative communication 12 with a Convolutional Neural Network-based classification machine learning model 84 adapted to classify wound-related images for wound management through image diaries and Estimation of healing time. The data inputs to train the model 84 and the output predictions the include primary inputs/outputs:

Images from the diary (X-axis);

Classification results (Y-axis); and

Database of wound infection images to train the classification model (X-axis).

[0063] Secondary inputs/outputs include:

Classification results (X-axis);

Wound Age (X-axis);

Estimated time to heal (Y-axis).

SUBSTITUTE SHEET (RULE 26) [0064] The orthopedics recommender system software structure 80 is in operative communication 12 with a Recurrent Neural Network-based predictive model 86 adapted to classify wound- related images for wound management through image diaries and Estimation of healing time. The data inputs to train the model and the output predictions the include primary inputs/outputs:

Images from the diary (X-axis);

Classification results (Y-axis); and

Database of wound infection images to train the classification model (X-axis).

Secondary inputs/outputs include:

Classification results (X-axis);

Wound Age (X-axis);

Estimated time to heal (Y-axis).

[0065] An anesthetics risk management software structure 88 is in operative communication 12 with a Polynomial Regression Algorithm-based predictive machine learning model 90 adapted to predict the risk of using anaesthetics on each particular patient, and aid doctors in completing risk assessments. The data inputs to train the model 90 and the output predictions include:

Various Static Factors of consumer including, Sex, Height, past medical history and family medical history [ first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis)

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, which could all act as independent variables (X-axis);

The quantity of anaesthetics used in previously recorded operations on the medical database, and its previously recorded after effects on different patients shall act as an independent variable (X-axis);

Pre-Op and Post-Op satisfaction questionnaire comparison (X-axis);

Medical history, as well as previous medication taken, (X-axis);

Anaesthetists need to do a risk assessment before op, and to get feedback, a qualitative survey, thus the risk score from the assessment of using the anaesthetics on a patient shall act as our dependent variable (Y-axis);

Doctors inputs on potential risks that the algorithm missed could also act as another independent data point (X-axis);

SUBSTITUTE SHEET (RULE 26) Lifestyle choices such as, Smoking, diet, weekly number of hours spent exercising, etc.; and

Type of operation and risk associated with it (X-axis).

[0066] The anesthetics risk management software structure 88 is in operative communication 12 with a Jaccard Similarity-based recommender system algorithm 92 adapted to predict the amount of time needed to complete different medical tasks by doctors to aid in time management. The inputs/outputs and/or attributes of the recommender system 92 include:

The type of task being completed, such as a risk assessment, could act as an independent variable (X-axis);

History of tracking the time taken for individual doctors to complete particular types of tasks (X-axis), learning behaviour of individual doctors over time;

Time taken for doctors to complete particular types of tasks in general (X-axis), starting point for time allocation of tasks, the model becoming more accurate in predicting the appropriate amount of time allocation for tasks as it learns the behaviour of the individual doctor it is recommending to;

Experience level of doctor i.e., PGY from medical school (X-axis); and Time allocation for each task (Y-axis).

[0067] The anesthetics risk management software structure 88 is also in operative communication 12 with a K Nearest Neighbor Algorithm-based recommender system 94 algorithm adapted to recommend the most effective postoperative pain management treatment plan. The data inputs to train the model and the output predictions include:

Current treatment plan implemented for particular patient (X-axis);

Consumer's static Risk factors, Age, Weight, height, Sex, etc (X-axis);

Consumer's dynamic lifestyle factors, alcohol use, illicit drug use, Smoking, Diet, Exercise routine, weekly number of hours spent exercising, Sleep Patterns, Work hours, etc. (X-axis);

Operation the consumer has recently undergone, (X-axis);

Consumer medical history, (X-axis);

Monitor patient postoperative vitals (X-axis);

Potential Complications from alarm system from vital checking, (X-axis);

Time increments after each vital update (X-axis);

SUBSTITUTE SHEET (RULE 26) Pre-op and post-op satisfaction and pain questionnaire (X-axis);

Reduction pain seen in post-op questionnaires (X-axis); and Newly Recommended treatment plan (Y-axis).

[0068] The Jaccard similarity coefficient (i.e., "algorithm") 92 is efficient at comparing patients that have undergone similar operations. It can detect what cases are shared and which are distinct. It is a measure of similarity for the two sets of data, with a range from 0% to 100%. The higher the percentage, the more similar the two sets. K Nearest Neighbour 90 is a secondary algorithm used in the risk management software structure. It stores the training dataset and learns from it only at the time of making real time predictions.

[0069] A gynecology recommender system software structure 96 is in operative communication 12 with a Jaccard Similarity-based recommender system algorithm 98 adapted to recommend an effective non-surgical treatment plan for chronic pelvic pains. The inputs/outputs and/or attributes of the recommender system 98 include:

Various Static Factors of consumer including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, which could all act as independent variables (X-axis);

History of this condition for consumer [duration of symptoms & intensity] List of symptoms and their intensity (X-axis);

Effectiveness score of plans calculated from pain questionnaire recorded during time interval and Sheehan's Disability Scales to assess social & occupational functioning (X-axis);

Time (X-axis);

Current treatment plan being implemented (X-axis);

Remission of symptoms over time (X-axis);

Treatment compliance rate (X-axis); and Newly recommended treatment plan (Y-axis).

[0070] The gynecology recommender system software structure 96 is in operative communication with a K Nearest Neighbor Model-based recommender system algorithm 100 adapted to

SUBSTITUTE SHEET (RULE 26) recommend an effective non-surgical treatment plan for chronic pelvic pains. The data inputs to train the model 100 and the output predictions include:

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, which could all act as independent variables (X-axis);

History of this condition for consumer [duration of symptoms & intensity] List of symptoms and their intensity (X-axis);

Effectiveness score of plans calculated from pain questionnaire recorded during time interval and Sheehan's Disability Scales to assess social & occupational functioning (X-axis);

Time (X-axis);

Current treatment plan being implemented (X-axis);

Remission of symptoms over time (X-axis);

Treatment compliance rate (X-axis); and Newly recommended treatment plan (Y-axis).

[0071] The Jaccard similarity (i.e., "algorithm") 98 coefficient is efficient at comparing patients with similar conditions. It can detect what symptoms are shared and which are distinct. It is a measure of similarity for the two sets of data, with a range from 0% to 100%. The higher the percentage, the more similar the two sets. K Nearest Neighbour 100 is a secondary algorithm used in the gynecology recommender system software structure. It stores the training dataset and learns from it only at the time of making real time predictions.

[0072] A gynecology recommender system software structure 96 is in operative communication 12 with a Collaborative Filtering-based recommender system algorithm 102 adapted to recommend effective infertility treatment plans. The inputs/outputs and/or attributes of the recommender system 102 include:

Various Static Factors of consumer including, Sex, Height, past medical history, and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use, weekly number of hours spent exercising, Sleep Patterns, Work hours, which could all act as independent variables (X-axis);

History of this problem, being duration of condition and intensity of symptoms;

SUBSTITUTE SHEET (RULE 26) Effectiveness score of plans calculated from Periodic monitoring of menstrual cycles, Cervical Mucus, Ovulation cycles, and determined number of eggs [denoting low medium & high success rate] (X-axis);

Time (X-axis);

Current treatment plan being implemented (X-axis); and Newly recommended treatment plan (Y-axis).

[0073] The gynecology recommender system software structure 96 is in operative communication 12 with a Convolutional Neural Network-based classification machine learning model 104 adapted to classify wound-related images for wound management through image diaries and Estimation of healing time. The data inputs to train the model 104 and the output predictions the include primary inputs/outputs:

Images from the diary (X-axis);

Classification results (Y-axis); and

Database of wound infection images to train the classification model (X-axis).

Secondary inputs/outputs include:

Classification results (X-axis);

Wound Age (X-axis); and

Estimated time to heal (Y-axis).

[0074] The gynecology recommender system software structure 96 is in operative communication 12 with a Recurrent Neural Network-based classification machine learning model 106 adapted to classify wound-related images for wound management through image diaries and Estimation of healing time. The data inputs to train the model 106 and the output predictions the include primary inputs/outputs:

Images from the diary (X-axis);

Classification results (Y-axis); and

Database of wound infection images to train the classification model (X-axis). Secondary inputs/outputs include:

Classification results (X-axis);

Wound Age (X-axis); and Estimated time to heal (Y-axis).

SUBSTITUTE SHEET (RULE 26) 1

[0075] The Convolutional Neural Network model 104 is trained to recognise the key features of a wound category (not shown). The Recurrent Neural Network model 106 recognizes data characteristics in data and uses patterns to make predictions about future events.

[0076] The gynecology recommender system software structure 96 is in operative communication 12 with a Collaborative Filtering-based recommender system algorithm adapted 102 to recommend effective post-operative treatment plans to prevent adverse effects. The inputs/outputs and/or attributes of the recommender system include:

Medical history of consumer (X-axis);

Consumer physical attributes, weight, age, height etc. (X-axis);

Consumer's Lifestyle habits, smoker, diet, exercise routine etc. (X-axis);

List of Symptoms and their intensity (X-axis);

Effectiveness score of plans calculated from pain questionnaire recorded during time interval and Sheehan's Disability Scales (X-axis);

Time (X-axis);

Current treatment plan being implemented (X-axis);

Remission of symptoms over time (X-axis);

Treatment compliance rate (X-axis);

Wound classification model results (X-axis); and Newly recommended treatment plan (Y-axis).

[0077] An obesity medicine/bariatric surgery recommender system 108 is in operative communication 12 with a Collaborative Filtering-based recommender system algorithm 110 adapted to recommend effective pre-operative weight loss plan. The inputs/outputs and/or attributes of the recommender system 110 include:

Medical history of consumer, (X-axis)

Various Static Factors of patient including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

List of Symptoms metabolic syndrome and their intensity (X-axis);

SUBSTITUTE SHEET (RULE 26) Time (X-axis);

Effectiveness score of plans calculated from weight changes and symptom updates recorded from diary;

Current treatment plan being implemented (X-axis);

Remission of symptoms over time (X-axis);

Weight loss over time (X-axis);

Treatment compliance rate (X-axis); and Newly recommended treatment plan (Y-axis).

[0078] The obesity medicine/bariatric surgery recommender system 108 is in operative communication with a Polynomial Regression Algorithm-based predictive machine learning model 112 adapted to predict chances of consumer suffering from metabolic syndrome. The data inputs to train the model 112 and the output predictions include:

Medical history of consumer (X-axis);

Medical history of consumer parents and family (X-axis);

Consumer physical attributes, weight, age, height, sex and ethnicity, Blood pressure, Blood sugar levels, cholesterol levels etc, (X-axis);

Consumer's Lifestyle habits, smoker, diet, exercise routine, Alcohol intake etc (X- axis);

Database of previous any metabolic syndrome and their details (X-axis); and Likelihood of developing metabolic syndrome (Y-axis).

[0079] The obesity medicine/bariatric surgery recommender system 108 is in operative communication with a Polynomial Regression Algorithm-based predictive machine learning model 108 adapted to predict the likelihood of consumer requiring for Bariatric surgery due to development of metabolic syndrome. The data inputs to train the model 108 and the output predictions include:

Medical history of consumer (X-axis);

Medical history of consumer parents and family (X-axis);

Consumer physical attributes, weight, age, height, sex and ethnicity, Blood pressure, Blood sugar levels, cholesterol levels etc, (X-axis);

Consumer's Lifestyle habits, smoker, diet, exercise routine, Alcohol intake etc (X- axis);

Database of previous any metabolic syndrome and their details (X-axis); and

SUBSTITUTE SHEET (RULE 26) Likelihood of developing metabolic syndrome (Y-axis).

[0080] An obesity medicine/bariatric surgery recommender system 108 is also in operative data 12 communication with a Collaborative Filtering-based recommender system algorithm 110 adapted to recommend an effective risk reduction and remission plan for individual patients to treat metabolic syndrome. The inputs/outputs and/or attributes of the recommender system 110 include:

Medical history of consumer and parents of consumer, Metabolic syndrome history (X-axis);

Various Static Factors of patient including, Sex, Height, past medical history and family medical history [first-degree & second-degree familial history], pharmacogenetics, Allergy/Adverse drug outcomes history which could all act as independent variables [independently or combined] (X-axis);

Dynamic Factors, current medications, blood sugar level, blood pressure and weight, smoking, alcohol use, illicit drug use which could all act as independent variables (X- axis);

List of Symptoms of metabolic syndrome and their intensity, if applicable (X-axis); Time (X-axis);

Risk of developing Metabolic syndrome, from prediction model;

Effectiveness score of plans calculated from weight changes and symptom updates and Risk of developing metabolic syndrome, recorded from diary (X-axis);

Current treatment plan being implemented (X-axis);

Remission of symptoms over time (x-axis);

Weight loss over time (x-axis);

Treatment compliance rate (X-axis); and Newly recommended treatment plan (Y-axis).

[0081] The various Machine Learning Algorithms/Models are implemented in a dedicated Machine Learning sub-system (or as an independent system) (not shown) to support the Machine Learning Algorithms/Models architectures. The sub-system has and online real-time Machine Learning Model feature database for making real-time predictions (or i.e., "inferences"). The subs-system also has an offline Model feature database used for Algorithm training and batch scoring. A model repository (or registry) stores trained Machine Learning Models, implements lineage tracing for Machine Learning models, and can

SUBSTITUTE SHEET (RULE 26) also act as version control for Models. A Machine Learning Model performance feedback loop automates Model update retraining tasks initiated by the deployment of Models to the data processing phase (i.e., prediction/inference). An alarm manager receives alerts from a Model monitoring system. The alarm manager initiates actions by publishing notifications to services that deliver alerts to a target application(s) in operative communication with the sub-system. A machine learning model re-training scheduler initiates model retraining at business-defined intervals. A Machine Learning Model lineage tracker enables reproducible Machine Learning experiences by enabling the recreation of Machine Learning Model behaviour at a specific point in time. The lineage tracker also collects altered reference by means of alternative iterations of Machine Learning Model lifecycle phases. Alternative Models and feature lists are evaluated as experiments for deployment.

[0082] In a preferred embodiment, the intelligent medical services system 2 is implemented on a distributed/cloud system.

[0083] In a preferred embodiment, the intelligent medical services system 2 includes a GUI (not shown) for a doctor and a GUI (not shown) for a consumer for interacting with the software structures of the system and/or an application with which the system 2 is integrated.

[0084] In an aspect to the present invention, there is provided Machine-readable code containing one or more sets of instructions (not shown) for implementing the intelligent medical services system 2.

[0085] In another aspect to the present invention, there is provided a method (not shown) performed by executing a set of steps for implementing an artificially intelligent management system 2.

[0086] A Person Skilled in the Art will appreciate that one or more Methods are disclosed wherein the various Machine Learning and other Intelligent elements can be implemented in one or more Methods including one or more steps.

[0087] A Person Skilled in the Art will appreciate that an application associated with the system may include various GUI's and other software structures directly or indirectly associated with the system and in direct or indirect communication with the system.

SUBSTITUTE SHEET (RULE 26) [0088] A Person Skilled in the Art will implement the system as a secure and safe system, including from a privacy standpoint.

[0089] To use the artificially intelligent management system as a doctor 4A, a doctor (or other appropriate person) user 4A, interacts with the system 2 by accessing one or more of the software structures via an application (not shown) depending on a particular consumer user's 4B health information/record and status (i.e., a "User Story"). Various aspects of a consumer's 4B healthcare can be managed by the doctor 4A in real-time.

[0090] The doctor 4A provides inputs via a computing device 8 (in a preferred embodiment i.e., a desktop or laptop computer) which are communicated to the system 2 via the cloud 10 (i.e., networking) and processed by the system 2 to provide intelligent information outputs (not shown) to assist the doctor to provide healthcare to a consumer 4B. The doctor 4A uses the system 2 (and in particular the outputs from the Machine Learning Models, Recommender Systems, Mathematical formulas, Functions and other elements) to prevent, diagnose, treat, and cure physical and mental ailments in a consumer 4B. The system 2 aggregates and leverages data with continual use to provide increasingly better healthcare outcomes. Inputs and particularly the outputs (not shown) are stored in the database 6 and can be accessed by the system 2 as required.

[0091] The general form of a doctor 4A User Story involves the doctor 4A accessing the system 2 via the cloud 10, interacting with the system via a computing device 8 GUI (not shown) as required. For example, a doctor 4A may access the obesity medicine/bariatric surgery recommender system 108 of the system 2 and receive information regarding the likelihood of developing metabolic syndrome predicted by the Polynomial Regression Algorithm-based predictive machine learning model 112. The doctor 4A may then contact the consumer 4B to discuss the model's output with the consumer. It will be appreciated that a plethora of such doctor User Stories are possible by accessing any one or more of the software structures n related to a consumer 4B user of the system 2.

[0092] To use the artificially intelligent management system 2 as a consumer 4B, the consumer 4B interacts with the system 2 by accessing one or more of the software structures of the system 2 via an application (not shown) depending on their particular health

SUBSTITUTE SHEET (RULE 26) information/record and status (i.e., a "User Story"). Various aspects of a consumer's 4B health are managed by the doctor 4B, including in real-time.

[0093] The consumer 4B provides inputs by means of a computing device 8 (in a preferred embodiment i.e., a tablet) to the system 2 via the cloud 10. The inputs are processed by the system 2 to provide intelligent information outputs (not shown) to assist the consumer's 4A doctor 4B to provide healthcare to the consumer 4B. The consumer 4B uses the system 2 to receive healthcare in the form of advice to prevent, diagnose, treat, and cure physical and mental ailments in a consumer 4B. Inputs and particularly the outputs (not shown) are stored in the database 6 and can be accessed by the system 2 as required.

[0094] The general form of a consumer 4B User Story involves the user accessing the system 2, interacting with the system 2 via a GUI (not shown) as required. For example, a consumer 4B may access the system 2 via and the obesity medicine/bariatric surgery recommender system 108 and receive information regarding the likelihood of developing metabolic syndrome predicted by the Polynomial Regression Algorithm-based predictive machine learning model 112. The doctor 4A may then contact the consumer 4B to discuss the model's 112 output with the consumer 4B. It will be appreciated that a plethora of such User Stories are possible by accessing by accessing any one or more of the software structures n managed by a doctor 4A user of the system 2.

[0095] A doctor 4A/consumer 4B User Story can comprise n system 2 software structures in parallel, sequentially, or a combination of both, n software structures can be used repeatedly and updated by a doctor 4A/consumer 4B User Story or n User Stories.

[0096] It will be appreciated that the presented healthcare management system integrates disparate and niche healthcare management systems into a single, real-time capable, holistic healthcare management system that aggregates and leverages data in a way that is useful to healthcare providers and consumers.

[0097] Throughout this specification, references to a Central Processing Unit-based system ("CPUbased Systems) are to be taken to include references to any one or more computing devices, including Mobile Devices, Personal Computers ("PC"), Embedded Systems, Quantum Computers, Distributed Systems (i.e., Networked CPU-based Systems), Microprocessors,

SUBSTITUTE SHEET (RULE 26) Microprocessor-based systems, Microcontrollers, CPUs, Computer Systems, Multicore Processor-based Systems, BYO Devices, or similar, or any combination thereof, and, any Peripherals to the foregoing, as deemed necessary by a person skilled in the art to perform the embodiments presented in this specification, or other embodiments of the present invention. Further, references to Peripherals, without limitation, are taken to include computer memory, and databases stored on computer memory, and similar.

[0098] References to a Machine-readable Code Containing a Set of Instructions (i.e., Software Program(s), Machine Readable Code, etc.) are to be taken to include references to instructions (including in respect of a method or algorithm) for a CPU-based System that can be implemented in machine code, assembly language, high-level computer languages, simulation and model-based design packages, or similar, or any combination thereof as deemed necessary by a person skilled in the art to perform the embodiments presented in this specification, or other embodiments of the present invention.

[0099] References to a Software Structure are taken to include references to a Software Program having layered architecture, event-driven architecture, microkernel architecture, microservices architecture, space-based architecture, data flow architecture, independent component architecture, call and return architecture, data-centered architecture, virtual machine architecture, real-time architecture, or similar, or any combination thereof as deemed necessary by a person skilled in the art to perform the embodiments presented in this specification, or other embodiments of the present invention.

[0100] References to a Network, Computer Network, Communications Network, Cloud, are to be taken to include references to wired or wireless Networks, any type of Network Topology, and any related protocol to implement such a network where required, or similar, or any combination thereof as deemed necessary by a person skilled in the art to perform the embodiments presented in this specification, or other embodiments of the present invention.

[0101] References to Machine Learning (ML) are to be taken to include references to Machine Learning, Deep Learning, and other types of Supervised, Semi-supervised, and Unsupervised

SUBSTITUTE SHEET (RULE 26) Learning, or similar, or any combination thereof as deemed necessary by a person skilled in the art to perform the embodiments presented in this specification, or other embodiments of the present invention. References to an ML Model are to be taken to include references to an algorithm trained by means of ML.

[0102] References to Artificial Intelligence are to be taken to include references to CPU-based systems that are able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, learning, decision-making, and natural language processing, or similar, or any combination thereof as deemed necessary by a person skilled in the art to perform the embodiments presented in this specification, or other embodiments of the present invention.

[0103] The term "operative communication" refers to the passing of data within a software program (i.e., machine readable code containing a set of instructions), between software programs, calling APIs, calling functions, accessing variables, modules, the implementation of networking principles, or other known computer communication principles within or between computer systems.

[0104] The PSA will appreciate that Software Programs can comprise modules and in a preferred embodiment the various aspects of the present invention can be referred to as "modules".

[0105] The PSA will appreciate that the system 2 can be implemented by applying Information Technology ("IT") principles whereby software is coded, databases created, and networks are built to create the system 2.

[0106] Although the invention has been described with reference to a specific example, it will be appreciated by those skilled in the art that the invention may be embodied in many other forms.

SUBSTITUTE SHEET (RULE 26)