WO2019089949A1 - Methods and systems for a medical screening system - Google Patents

Abstract

Methods and systems for a Master Medical Product (MMP) that quantifies an individual's risk state and measures the degree of change in the risk state as a result of preventive therapy are disclosed. The MMP includes a medical screening tool designed to catalyze a transition of healthcare delivery from traditional hospital- or office-based visitations, to technology-based encounters that are virtual, on-demand- and individualized. The screening tool unequivocally assures normal and rules-out infrequent treatable abnormal, and addresses all necessary cardiovascular screening criteria and reported obstacles. The MMP introduces a paradigm shift using a tool that assists prediction and predestination screening of a subclinical epidemic of non-communicable disease (e.g., SCD).

Description

METHODS AND SYSTEMS FOR A MEDICAL SCREENING SYSTEM

FIELD

This disclosure relates generally to medical screening systems. More specifically, the disclosure relates to methods and systems for quantifying an individual's medical risk state and measuring the degree of change in the risk state as a result of preventive therapy or environmental influence.

BACKGROUND

The National Heart, Lung and Blood Institute (NHLBI) calls sudden cardiac death (SCD) in the young a 'critical public health issue.' One in every ten children is born with a congenital heart condition, of which 30% will have moderate to severe lifelong heart risk. About 80% of sudden cardiac arrest (SCA) victims do not describe previous symptoms. SCD is the number one cause of death for student athletes. Also SCD is the number two medical cause of death for youth under age 25. The incidence of SCD has not declined over the past thirty years. The inherent tendency in medicine is to diagnose and treat symptoms, often referred to as "sick care," which has had a very limited effect on preventing or reducing the incidence of SCD. Standard physical exams miss 96% of the risks associated with heart conditions. Diagnostic tests are unsuitable for screening asymptomatic youth due to the inherent high incidence of false results, which cause unnecessary anxiety and expense. The American Heart Association (AHA), American College of Cardiology (ACC) and the National Collegiate Athletic Association (NCAA) no longer recommends a screening ECG (because of false results, etc.) or a comprehensive diagnostic echocardiogram (because of cost, etc.). Because there is no effective early detection of risk and the high costs of common diagnostic testing, there currently is no acceptable screening solution to this health care crisis.

SUMMARY

Currently no diagnostic technology is able to quantify and assist prevention of the large health care crisis. The SCD epidemic and numerous others (e.g., heart failure, stroke, atrial fibrillation, diabetes, etc.) remain unresolved and consume billions of dollars each year.

The paradigm shift away from diagnostic and statistical medical technologies introduces simple inputs that address logical complex problems, and generate simple probabilistic outputs that users can easily understand, interpret and apply. The automated transformation of data into knowledge will markedly assist the health care community make more informed decisions.

The embodiments described herein can quantify an individual's medical risk state and measure the degree of change in the risk state as a result of preventive therapy or

environmental influence.

In one embodiment, a medical screening system is disclosed. The system includes a specialized medical computer device, a sensor and/or a machine to measure anatomical and physiological data, and an output device. The specialized medical computer device is configured to receive inputs from the sensor and the machine. The output device is configured to output data from the specialized medical computer device. The sensor and/or the machine are/is configured to measure anatomical and physiological data. The specialized medical computer device is configured to generate a likelihood ratio (LR) based on the measured anatomical and physiological data. The specialized medical computer device is configured to generate a risk state based on the LR and/or the anatomical and physiological data. The output device is configured to output the risk state.

In another embodiment, a specialized medical computer device is disclosed. The device includes a processor, non-transitory computer readable storage mediums, and specialized medical computer program instructions stored in the non-transitory computer readable storage mediums. The processor is configured to execute the specialized medical computer program instructions. The processor is also configured to obtain measured anatomical and physiological data. The processor is further configured to generate a likelihood ratio (LR) based on the measured anatomical and physiological data. Also the processor is configured to generate a risk state based on the LR. Further the processor is configured to output the risk state.

In yet another embodiment, a method of determining a risk state is disclosed. The method includes measuring anatomical and physiological data. The method also includes determining a likelihood ratio (LR) based on the measured anatomical and physiological data. The method further includes determining the risk state based on the LR. Also the method includes outputting the risk state.

Other features and aspects will become apparent by consideration of the following detailed description and accompanying drawings.

The embodiments described herein include a Master Medical Product (MMP). The MMP can refine the deluge of data into knowledge, which can provide a user with increased wisdom that assists in making more knowledge -based decisions. MMP can alleviate the occurrence of false positive information and markedly reduce the cost of instituting evidence based medicine. MMP is a high-quality, low-cost system capable of predicting, managing and preventing the global health care crisis.

MMP is a transformative product that can be easily implemented by the world's health care community. MMP has the ability to measure the degree of individual risk and then quantitatively measure the effect of prevention measures, which is destined to cause a cataclysmic transformation of preventive medicine and save millions of lives. MMP can provide vast medical information at low cost. MMP can enable early detection, measuring and monitoring of health risk. MMP can be a repository and producer of personal and global knowledge. Access to MMP can be by Internet subscription. MMP can store a complete model of individual and world knowledge and continuously update and learn relationships.

MMP is destined to be the most revolutionary medical technologic product to be introduced in the past one hundred years. MMP can address a huge social need and meet a required change in worldwide medicine. MMP has identified and validated a high-quality means of measuring the likelihood of sudden death and a means of determining the effect of management. MMP is capable of automatically quantifying pre-clinical disease and monitor outcome of management. For example, having recognized the immense unmet need for a new means to predict and prevent events such as SCD, MMP can deliver a new innovative approach that unequivocally affirms a heart's risk state and rules-out or rules-in silent life- threatening heart conditions. MMP has the capacity to quantify an individual's risk state and measure the degree of change in the risk state as a result of preventive therapy. MMP can fundamentally change how to predict an individual's risk of dying from life-threatening disease risk. MMP can make a monumental change in preventive medicine and alleviate a global health care crisis.

MMP is capable of preventing the continued growth of a global health care crisis.

MMP represents an enormous social and market opportunity. MMP can improve preventive measures and markedly reduce costs and the necessity of costly diagnostic testing. MMP can measure individualized quantitative risk and monitor responses to risk management. In some embodiments, MMP can be a global SaaS-based (software as a service) platform.

The transformational MMP can address the unresolved global health care crisis by quantifying individualized disease risk (cause) and measuring the degree of change of treatment (effect). For example, a multi-tenant SaaS-based MMP, with universal data integration protocols, can cost-effectively transform data into knowledge about a broad spectrum of individual disease risk states. In some embodiments, the MMP can include five state-of-the-art technological components that together stratify individual risk: logic data, systems biology, causality, Bayesian inference, and designed precision and accuracy. The MMP risk state can be a Bayesian factor ("Likelihood Ratio") complemented by visual narratives that assist users to directly participate in health care decision making. The world's health care community, nonmedical entities (e.g., insurance, health clubs, and/or companies) and personal-device users are potential targets to use the low-cost, high quality MMP.

BRIEF DESCRIPTION OF THE DRAWINGS

References are made to the accompanying drawings that form a part of this disclosure and which illustrate the embodiments in which systems and methods described in this specification can be practiced.

FIG. 1 illustrates a block diagram schematic of a SET system, according to one embodiment.

FIG. 2 illustrates a user interface for collecting data via a participant data form, according to one embodiment.

FIG. 3 illustrates a user interface for outputting/displaying the results from the SET test, according to one embodiment.

FIG. 4 illustrates another user interface for outputting/displaying the results from the SET test, according to one embodiment.

FIG. 5 shows a step-by-step calculation of LRs, according to one embodiment.

FIG. 6A and 6B show a LR Assists Data Interpretation, according to one embodiment. FIGS. 7-9 show analytic causality and serial transformation of the LR, according to one embodiment.

FIGS. 10A and 10B show data validation features associated with SCD, according to one embodiment.

FIG. 11 shows a five-step rule-out echo exam, according to one embodiment.

Fig. 12 is a flow chart illustrating a method for determining a risk state for a disease, according to one embodiment.

Like reference numbers represent like parts throughout. DETAILED DESCRIPTION

Diagnostic innovations (such as genomics, biomarkers, guidelines, or big data) in the near term have not transformed the preventive medicine market. Delivery of diagnostic innovations will take years and foster the use of more expensive sophisticated static diagnostic technology. Diagnostic medicine cannot compute cause and effect. Even strong statistical correlations cannot measure or monitor change in treatment outcome. Diagnostic medicine has relied on population based regression statistics. Validation typically requires huge expensive retrospective trials that define correlations and associations. These studies are extremely difficult to replicate. The medical community knows that the same shoe does not fit each individual in a population. In addition, there are thousands of diagnostic manuscripts suggesting various means of preventing, for example, SCD and heart failure. However, to this date the effect on the health care crises has not appreciably changed. For decades conventional diagnostic and statistical management of health risk has been unacceptably suboptimal in the prevention of SCD and chronic disease.

Most potential heart problems in young people are present at birth and silently become unstable over time. Over 95% of these conditions have early risk features that can be detected by a specialized high quality screening test capable of identifying early structural, functional, and hemodynamic abnormalities. In one embodiment, Echocardiography (Echo) is uniquely capable of assessing all three attributes.

The MMP can include a medical screening system. In one embodiment, the medical screening system includes a Specialized Echo Test (SET) or Focused Echo Screening Test (FEST). The SET can include a risk assessment questionnaire, a sequential blood pressure check, and a sensitive SET rule-out test that unequivocally affirms key heart and blood vessel structures, functions, and hemodynamics are of a normal state. The SET rule-out test includes tests on four heart chambers and mitral and tricuspid valve; abdominal aorta (blood velocity profile); left ventricular multi-data diastolic function and systolic function; aortic valve, mitral valve, tricuspid valve and pulmonary valve; left and right coronary artery origination; pulmonary artery pressure; aorta dimensions in the chest and blood flow velocities; right ventricular outflow size; and/or left atrial volume index.

By affirming the normal state of heart and blood vessel features, the SET can rule-out abnormal features known to be associated with >95% of the disease associated with SCD. Those abnormal features include cardiomyopathies (early emergence of abnormal function with dilated, hypertrophic, restrictive, hypertensive, myocarditis, etc.), congenital and acquired structural deformities (bicuspid aortic valve, aneurysm, congenital heart disease, etc.), hemodynamics features (high risk coarctation, pulmonary hypertension, etc.), and abnormal angiographic features (valve regurgitation, shunts, etc.). The SET includes a high-quality, low-cost, and brief (e.g., about 15 minutes) preventive heart screening for asymptomatic individuals, for example, ages 5 - 25. The test objective of SET includes affirming that the selected (specialized) functional, structural, and hemodynamic heart features known to be associated with SCD are unequivocally normal.

It will be appreciated that 98% of the general population is expected to have normal features. By focusing on affirming normal, the SET can rule-out the risk of >95% of the pathophysiologic features of SCD. The 2% that cannot be affirmed as normal are referred to a physician for further assessment. By seeking normal, the SET can efficiently triage the participant population to determine the minority group requiring further assessment. Rule- out test is different from a typical "rule-in" diagnostic test that seeks to identify a rare disease within a broad healthy asymptomatic population, resulting in false results, additional testing, and unnecessary costs.

Fig. 1 illustrates a block diagram schematic of a SET system 100, according to one embodiment. As shown in Fig. 1, the SET system 100 includes an echocardiographic (or echocardiogram) machine 105 to collect patient information. The SET system 100 can also include at least one sensor 110. The SET system 100 can further include various input device(s) 115 to collect patient information and at least one output device 130. Also the SET system 100 includes a specialized medical computer device 120. The SET system 100 can also include local data storage 125 and a cloud device 135. The local data storage 125 can connect to the specialized medical computer device 120 wirelessly or in wire to store data from the specialized medical computer device 120. The specialized medical computer device 120 can connect to the cloud device 135 wirelessly or in wire. The cloud device 135 can include remote storage space(s) and/or cloud processing device(s). The cloud device 135 can store data from the specialized medical computer device 120 and/or process instructions from the specialized medical computer device 120 and output resultant data to the specialized medical computer device 120. The at least one output device 130 can connect to the specialized medical computer device 120 wirelessly or in wire. The at least one output device 130 can include a display device and/or a printing device.

The specialized medical computer device 120 can include a processor and non- transitory computer readable storage mediums (such as a computer-readable memory). The specialized medical computer device 120 can be configured to receive inputs from the echocardiographic machine 105, the sensor 110, and/or the input device 115.

It will be appreciated that the echocardiographic machine 105, the sensor 110, and/or the input device 115 can connect to the specialized medical computer device 120 wirelessly or in wire. It will also be appreciated that the echocardiographic machine 105, the sensor 110, and/or the input device 115 can output patent information to non-transitory computer readable storage mediums, and the specialized medical computer device 120 can receive the patient information from the non-transitory computer readable storage mediums.

In one embodiment, the at least one sensor 110 includes a device (e.g., blood pressure sensor and/or meter) that measures patient's blood pressure. The input device(s) can include a device that collects/measures/senses patient's information.

In one embodiment, the SET system 100 can utilize a validated gold standard for determining normal. The specialized medical computer device 120 can be configured to take the measured information (from, e.g., the echocardiographic machine 105, the sensor 110, and/or the input device 115) and generate a portfolio of specific data for functional, structural, and hemodynamic features. The generated specific data can be saved to, e.g., the data storage 125. The specialized medical computer device 120 can be configured to affirm the normal status versus abnormal status based on the state of the generated specific data, and to generate an individualized report output (e.g., via the output device 130) on a tangible medium after the SET test.

In one embodiment, in operation, a portfolio of specific data is collected, e.g., from a participant data form (e.g., at registration), from data collected/measured/sensed via the input device 115, from a sequential blood pressure reading (e.g., at check-in) via the sensor 110, and/or from the echo/ultrasound machine 105 (e.g., at screening process). Data measured from the echo machine 105 includes functional, structural, and/or hemodynamic data.

In one embodiment, the specialized medical computer device 120 receives the collected portfolio of specific data (e.g., the participant, blood pressure, functional, structural and hemodynamic data), determines the participant's cardiovascular risk status (e.g., likelihood ratio "LR", which is described in detail in later sections) using evidence-based logic (or definitions) for normal versus abnormal, and outputs the appropriate status on a participant report.

Fig. 2 illustrates a user interface for collecting data via a participant data form 200, according to one embodiment. In one embodiment, the participant data form 200 can be displayed by, e.g., the output device 130 of Fig. 1, and filled by a user. In another embodiment, the participant data form 200 includes information (e.g., height, weight, etc.) that can be measured and/or sensed via the input device 115 and/or the sensor 1 10 of Fig. 1 and outputted to the output device 130 of Fig. 1. As shown in Fig. 2, the participant data form 200 can include a schedule tab, a participant information tab, and/or a confirmation tab. It will be appreciated that the schedule tab, the participant information tab, and/or the confirmation tab can be independent page(s) (e.g., webpage) instead of tabs. Fig. 2 shows a user interface for collecting participant's information.

Figs. 3 and 4 illustrate a user interface for outputting/displaying the results from the SET (FEST) test, according to one embodiment. Fig. 3 shows a user interface 300 for displaying a normal structure, function, and blood flow and pressure. Fig. 4 shows a user interface 400 for displaying a possibly abnormal structure, a normal function, and a normal blood flow and pressure. As shown in Figs. 3 and 4, the participant section shows collected/measured/sensed information (e.g., body mass index "BMI") as well as

informational information (e.g., normal BMI range).

The user interface for outputting/displaying the results from the SET (FEST) test as shown in Figs. 3 and 4 can be generated by, for example, the specialized medical computer device 120 of Fig. 1, and outputted or displayed by the output device 130 of Fig. 1. The status of the individual feature types (e.g., normal or abnormal, etc.) can be defined by, for example, an easy to understand color (e.g., green color for normal, and red color for possible abnormal which would require a physician to further diagnose). In one embodiment, if systolic or diastolic blood pressure is elevated above published standards and the individual's functional features are normal, a note states "associated with normal physiology" or the like can be displayed/printed under the blood pressure score in the output (see Fig. 4).

In operation, the participant can access the SET system, via for example, a website, a mobile application, and/or a local computer application. The participant can schedule the SET test, fill out the participant data form (e.g., 200 of Fig. 2) or gathering data (e.g., via sensors or meters) for the participant data form (e.g., 200 of Fig. 2), and conduct payment process (e.g., including payment, coupons, receipts, etc.). In one embodiment, the participant checks in at the scheduled date/time, pre-screening data (e.g., blood pressure, etc., via the sensor 110 of Fig. 1) are gathered, and screening measurements (e.g., via the echo machine 105 of Fig. 1) and/or observations are taken. The results generated by, for example, the specialized medical computer device 120 of Fig. 1 can be reported to the participant (e.g., via the output device 130 of Fig. 1).

Referring back to Fig. 1, the echocardiographic machine 105 (e.g., an ultrasound device) can measure ultrasound information of a participant's heart and/or cardiovascular physiology (e.g., specialized portfolio of highly selected data, which are associated with SCD, and the selected echocardiography data can be uniquely capable of assessing three attributes, structure, function and hemodynamics.)- The specialized medical computer device 120 can generate specific data from the measured information, and/or save the generated data to the local data storage 125 or the cloud 135. The specialized medical computer device 120 can output the results on a tangible medium (e.g., via the output device 130) based on the generate data, where the output has information regarding each of the participant's functional, structural, and hemodynamic features using, for example, a binary scale, of unequivocal normal versus abnormal. The results from the specialized medical computer device 120 can unequivocally affirm the normal state of selected heart and blood vessel features and rule-out abnormal features known to be associated with SCD (such as pathophysiologic functional, structural, and hemodynamic data-state).

Diagnostic tests (including echo, EKG, physical exams, etc.) used for screening of a low incident cause of sudden death are ineffective and exceedingly expensive. The SET system is a high-quality system that can alleviate false positive information and high cost cardiac screening. The SET/FEST is a high quality and sensitive test that can markedly lower cost while retaining highest quality. The SET tests disclosed herein have high quality (leverages the accuracy and authority of a gold-standard echocardiographic test), have low cost (e.g., less than $100 as opposed to diagnostic echo exams costs more than $1000), have high precision and accuracy (unequivocal results as opposed to the inherent false results from EKG and echo diagnostic tests used to screen asymptomatic participants), and are convenient (brief (e.g., about 15 minutes or less) test with immediate results as opposed to a more than 45 minutes traditional screening and diagnostic testing with delayed results). The SET system can address an unresolved social imperative (i.e., sudden death in young athletes and non-athletes).

The SET (FEST) is one embodiment (e.g., for SCD) of the MMP. MMP can transform medicine from diagnostic testing to quantifiable prevention testing and assisted management. MMP can define the cause and measure the effect of treatment. MMP is a totally new innovation that can save lives and lower mortality, morbidity and cost. The embodiments described herein can transform a deluge of data into knowledge and is anticipated to be one of the greatest technological transformations in medicine.

MMP includes multiple sub-components working in concert to orchestrate a singular adaptive master learning machine or product. MMP can easily spread throughout the world's health care community, and the health care community can easily understand MMPs' attributes. MMP can quantify silent disease and measure the response to preventive therapy. MMP is a cataclysmic change, which can be easy to understand and inexpensive to implement. MMP can measure individualized risk. An individual's unique risk profile can be continuously monitored and measure the success or failure of a selected management.

MMP utilizes small data, computes complex knowledge, and makes the knowledge available to those who can use the knowledge at a time and place that is appropriate to achieve maximum effective and monitored use. MMP can refme data into usable knowledge that assures proper use and validation of effect. MMP is a transformative means of managing the deluge of data and disseminating the resultant knowledge.

MMP knowledge management is a set of new organizational activities that are aimed at improving knowledge, knowledge-related practices, organizational behaviors and decision and organizational performance.

EXAMPLE: in testing MMP, 50,000 human interpretations of heart failure reports have been compared with MMP. About 70% of human interpretations were incorrect. This sobering finding has been reproduced throughout the health care community. By design and continuous simulation testing MMP knowledge is 100% accurate and precise. Incorrect empiric interpretation can be immediately corrected by implementing MMP assisted data management. The difference between diagnostic and MMP medicine is nothing short of cataclysmic. Simulation models are used to validate and verify the MMP throughout the entire MMP design and utilization processes. More than 1,000 simulation prototypes are used during design to affirm absolute reproducibility (precision) and nearness to a true objective (accuracy). MMP factors are much faster and more powerful than the retrospective statistical approach.

In some embodiments, MMP can be used as risk stratification for, e.g., SCD, that involves the development of individualized dynamical multi-feature models based on measured anatomy and physiology. SCA and SCD are defined as an abrupt, unexpected, out of hospital event due to a cardiovascular (CV) cause. SCA is the single most common cause of death in western societies. The accurate determination of cause of sudden death is challenging because of the high number of silent comorbidities and complex clinical scenarios at the time of death. There is discordance among investigators and supposed proven cause of death, with investigators attributing more deaths to cardiac causes. However, more recently sudden death is more often attributed to non-CV disease. These variances depend more on the clinical associations and circumstances of death rather than the actual cause of death and therefore are not useful distinctions. Despite numerous prevention initiatives, the incidence of sudden SCA/SCD has remained a major health care crisis for more than 60 years in the U.S.A. (>400,000 individual SCDs per year). There is little evidence to support the use of diagnostic screening tests and statistical correlations as a means of predicting or preventing SCD. From an evidence-based medicine and policy perspective, recommending screening or prescribing conventional screening tests is considered inappropriate unless multiple obstacles are sufficiently resolved. SCA/SCD should be viewed as end- stage events and not a suitable target for prevention. Like all epidemics, prevention must be the highest priority. MMP is a practical preventive solution that can address unsustainable costs and various degrees of social hysteria and false assumptions. MMP can prioritize the quantitative understanding and prevention of, for example, SCD. MMP can be a cataclysmic quantitative solution to the unresolved epidemic of SCD by quantifying both the existing intensity of the risk state and the ability of sequentially monitor improvement or deterioration of the risk state. MMP can be continually evolving, incorporating the results of new data, data sources and change over time.

MMP can solve the problem of risk associated disease using a totally new, validated means of quantifying an individualized risk state (cause) as well as quantify managed outcome (effect). MMP can have the capacity to create a cataclysmic change in how to predict and prevent the emergence of an unresolved global health care crisis.

MMP learning is the linchpin of the whole enterprise. Without MMP learning, fragmentary knowledge would scatter among thousands of databases and scientific articles, each user (doctor, health care provider, client, and/or patient) would be aware of only a small part of the information deluge. MMP learning can assemble all the knowledge into a coherent whole.

MMP includes multiple (e.g., five or more) technological components that interact to generate a unified conclusion. The seed MMP includes a logic data tool (LDT) component, a systems biology tool (SBT) component, a causality tool (CT) component, a Bayesian tool (BT) component, and a validation and verification tool (VT) component.

In one embodiment, the specialized medical computer device 120 of Fig. 1 can include a processor and non-transitory computer readable storage mediums. Specialized medical computer program instructions can be stored in the non-transitory computer readable storage mediums. The specialized medical computer program instructions can include the LDT, the SBT, the CT, the BT, and/or the VT. It will be appreciated that the LDT, the SBT, the CT, the BT, and/or the VT component of the MMP can be a specialized medical computer sub-device. Component One (Logic Data Tool):

LDT can put validated data features together into a logical coherent module. The foremost objective for LDT is to alleviate errors in the data selection process. Logical learning can only occur with the use of prior knowledge, which validates the interrelationship between data components (a posteriori: reasoning or knowledge that proceeds from observations or experiences to the deduction of probable causes). LDT can identify minimum of 2 and maximum of 4 to 5 highly related data that will be used in subsequent components (SBT, CT, BT, and VT).

A rule-based model can be one of the easiest means to identify a small set highly related (associated) data. · The LDT requisites include avoiding data that has look-alike (pleiotropic) features; using stringent data range cutoffs; defining a data-state as a singular expression of an aggregate of highly related data and avoiding using individual data; avoiding using data-to-disease correlations that are neither a sensitive or adequate marker of a disease state; and classifying data by its causal (pathophysiologic) risk relationship (Causality Tool) and ability to affirm therapeutic benefit; and continuously using simulation checks

(Validation and Verification Tool) throughout the design and build processes that affirm highest data precision (i.e., reproducibility) and validation of intended use (i.e., accuracy).

In some embodiments, when it is impossible to generate independent associated data for LDT, a Markov chain, which is a set of features and corresponding weights which together define a probability distribution, can be used. The number of times a specific data- state is visited is proportional to its usability and how they depend directly on each other. Highly associated data will converge to a stable association, so that eventually it always gives approximately the same answers. Logic networks can be trained to maximize either the likelihood of the whole data or the conditional likelihood of what the users want to predict given what the users know. Data are also weighted by how probable they are and the weight of the individual or cumulative data-state. A logic data network of features and

corresponding weights define a probability distribution. Another option is to repeatedly cycle through the data variables, sampling each one according to its conditional probability given the state of its neighbors. Hierarchical structure helps make inference tractable because subparts of the pathophysiologic order interact mostly with other features that share the same structural features. MMP is a combination of logical and probabilistic inference capable of computing the probability of logical formula. MMP can grow into an immense system of subparts. Disease entities do not come in arbitrary forms. Rather disease entities fall into classes and subclasses, with members of the same class being more alike than members of different ones. If the users know the distinctions relevant to the question at hand, the users can lump together all the entities that have the distinctions and that can save a lot decision making time.

Component Two (Systems Biology Tool):

SBT (also referred to as Network Medicine) can be described as dynamic data networks that follow self-similar small world organizing principals that provide a quantitative approach to multi-data analysis. SBT can define associations among individual data/features that conspire to yield quantitative knowledge about a disease state, can simplify the binary rule-out classification of disease as normal versus abnormal, and can assist therapeutic decisions that influence medial therapy.

Component Three (Causality Tool):

CT can define how to address cause and effect relationships. The MMP can automatically quantify the magnitude of emergent risk {cause) and then sequentially quantifies the change of the risk state {effect). MMP can structure the logical distribution of health care resources (e.g., evidence based preventive medicine). An ideal prognostic screening test must quantify inferred pathophysiologic risk (cause) as early as possible and then monitor or manage the continuum of change (effect). Causal risk is composed of a small number of highly related effectors (highly related data) that define constituent features of cause. The emergence of risk introduces the concept of an ever-changing risk state, which emerges as a continuum of risk intensity and opportunity for early identification and sustainable management. The opportunity to quantify the emergence of causal risk alleviates the limitations of diagnostic associations, independent risk factors, biomarkers, risk profiles, guidelines, epidemiologic scores and clinical genomics.

Transitioning from independent risk factors to measured causal factors is essential to logically solve the SCD epidemic. There is an evolving paradigm shift toward precision medicine (PM), which explicitly prioritizes the individualization of care and focuses on unique characteristics of a particular person. Causal factors must always be present if an effect is to occur. Causal factors are not defined statistically but are defined experimentally in that they are known to affect outcome. Diagnostic factors predict outcome while causal factors affect outcome. Quantifying causal risk adds immense credibility to the concept of precision medicine. Causal risk is a dynamic continuum of causal state (not an independent risk factor), which opens understanding and logical management of an individual's risk-status at any time throughout the life cycle. Ideal casual algorithm should require few resources and quick to execute prediction. Essentials include: measure of risk-intensity; measure risk-status within the temporal continuum; measure change of the risk-status (monitor outcome).

At the individual level, independent risk factors evoke uncertainty when used as part of prediction equation and should not be used as a primary means of measuring or managing an individual's risk. Causal factors are totally different from diagnostic risk factors. A causal factor cannot be measured statistically, and must be inferred logically and

experimentally determined to actually affect change in outcome. Cause is not a specific disease state but a sensitive factor within an individual's pathophysiologic continuum.

Causal factors are always present at the initiation of risk states and when the composite data change state they represent effect of the causal factor. Causality is a natural agent that directly connects one process (cause) with another process (effect), where the first is partly responsible for the second, and the second is partly dependent on the first. Risk rarely has a single cause. The importance of individual casual risk is highlighted by paradoxical risk factor observations (e.g., only 33% of SCD victims have high risk disease; most victims have a low-risk pathophysiologic profile; and commonly die from non-CV disease. These findings mandate the aggressive development of a more holistic screening method that measures an individual's causal risk burden instead of specific diagnostic risk factors.

Causality (also referred to as Cause and Effect Medicine) defines the most effective disease management strategy. Novel quantitative relationships of the physiologic state define a "state," which is then related to outcome (effect). The novel use of circular causality utilizes a domain specific pathophysiologic state as a surrogate cause and the change of the state as the effect. Another novel aspect of circular causality is that it eliminates the need of defining a specific diagnosis. Each causal state change depicts a new altered causal state (cause defines a new cause). A new causal state thus defines a new causal state. Perpetual change of the disease state acts a surrogate and depicts and quantifies the success or failure of management (e.g., ultimate effect is a "normal or stable" causal state). Cause and Effect, based on circular causality, is based on the dynamics of causal state as both cause (initial state) and effect (new state). The computational sophistication of "causal relationships" is further enhanced. MMP will profoundly change medical management of disease. The pharmaceutical industry will be particularly benefitted.

One cannot substantiate causal claims from diagnostic associations alone, even if there is a strong statistical relationship - behind every causal conclusion there must lie some causal assumption that is not testable in correlation studies. It is important to put emphasis on the quantification of risk (i.e., cause) that is amenable to monitored intervention and measured change in outcome (i.e., effect). CT includes 4 rules of causal data relationships: contiguity (cause and effect data must be contiguous in time and space); succession (cause must occur prior to effect); conjunction (there must be a constant union between cause and effect); and counterfactuals (change of a data module (effect) can only occur the presence of an induced cause). In CT, causal risk can be accessed by the relationship between measured risk-state (cause) and temporal change of the causal data-state (effect). Causal questions cannot be easily answered by applying statistical methods of independent risk factors or big data. Most conventional learning methods do not attempt to uncover cause and effect relationships between features and target. The clinical management of causality is a form of knowing through intentional action. Causality is inferred and not an entity; it should not be made more concrete or real. Causation is induced logically, not observed empirically. Therefore, users can never know absolutely that exposure X causes disease Y. There is no final proof of causation: it is merely an inference based on an observed conjunction of two variables (exposure and health status) in time and space. Following the requisites of CT is used by the BT to compute quantitative cause and effect relationships.

Data management is most firmly entrenched in two disciplines: statistics (e.g., big data) and biology (complex small data). Of these two disciplines biology is the most important. Pathobiological cause and effect computer algorithms are about to take a commanding role in assisting the medical community resolve the global epidemic of SCD and chronic disease (CD). Physicians can encode logic based on prior experience but they are not very good at estimating probabilities in the presence of a deluge of incomplete or noisy data. The crucial challenge is how to compute probability as you receive larger and larger amounts of data. The answer is to begin to think in terms of cause and effect (e.g., Bayes' theorem) instead of current diagnostic risk factors and regression statistics that have insufficient explanatory power. By using cause and effect models, the users can dig deeper behind and underneath the data to explore richer relationships missing form statistical prediction models. A high quality, low cost, accessible technologic solution (MMP) is destined to revolutionize preventive medicine and introduce the concept of lifelong sustainable wellness.

Component Four (Bayesian Tool): BT includes Bayes' theorem, which is a machine that turns data into knowledge. BT includes one sub-component that summarizes a data-state (data-status; risk-intensity;

surrogate cause) and another sub-component that represents a belief (i.e., nonlinear data change; outcome; surrogate effect). The core of BT is a Bayes factor, which in its simplest form is called a likelihood ratio (LR). The main advantage of LRs is that clinicians can use them to quickly quantify different strategies and thus refine clinical judgment. Unlike population statistics, the LR has a sound theoretical foundation and interpretation that enhances decision making based on individualized bidirectional reasoning.

A quantitative LR includes taking weighted average of highly related data (i.e., logical data, network systems) and computes the status (cause) and change of a risk state (effect). The bigger the LR the more convincingly the finding suggests the emergence of a risk event; the closer the LR is to 0, the less likely the disease; and sequential LRs depict the temporal continuum of risk. A graphic trajectory of sequential LRs is the simplest means of appreciating the dynamic profile of an individualized dynamic risk state.

Inference in Bayesian networks is not limited to computing probabilities. It also includes finding the most probable explanation of the evidence, such the risk state explaining associated symptoms. In BT, Bayesian factors have multiple additional functions. LR can be used in lieu of population statistics as a measure of the causal strength; can quantify CV function in a forward (positive) and backward (negative) direction so that the clinical outcome (e.g., cause and effect) and economic outcomes (e.g., costs, resource utilization) can be understood, projected and optimized in the absence of expensive randomized controlled clinical trials; sequential LRs can monitor short and long term change (e.g., evidence based cause and effect); plots divergence, which is the relative distance between sequential risk states (e.g., a greater distance between normal and a new LR, the greater the risk of an adverse event, and the shorter the distance the lower the risk of an adverse event); LRs can distinguish between evidence and error when LR (evidence) is a measure of how much the probability of truth is altered by changing the risk- state.

Component Five (Validation and Verification Tool):

In VT, simulation models are used to validate and verify the precision and accuracy of the probabilistic risk model. Simulation modeling is the process of creating and analyzing digital prototypes of the model to predict and assure its performance in the real-world environment. It is most efficient and cost effective to use simulation models throughout the entire design process because it is too costly and time consuming to retrospectively verify and validate a complex reasoning system. In a seed study, for example, SCD study, more than 1,000 simulation prototypes are used to affirm absolute reproducibility (precision) and nearness to a true objective (accuracy). Precision is easiest to verify since all results demonstrate that measurements under unchanged conditions always show the same result. LRs are the simplest, quantitative means of validating accuracy of probabilistic risk continuum. Continuous measure of the causal-risk reinforces accuracy and alleviates uncertainty. LR causal factors are much faster and more powerful than the sensitivity and specificity approach.

The combination of Logic (data) and Probability (approximation of a knowledgeable truth) is often considered to be impossible. However, by using the set of informatics (LTD, SBT, CT, BT, and VT) to build the MMP, both Logic (data) and Probability (CT and BT) can be conjoined into a single MMP solution.

MMP is combination of logical and probabilistic inference. MMP combines the two into a unified inference algorithm, capable of computing the probability of logical formula. MMP also includes data algorithms that add weight and the Bayes factors multiply probabilities. MMP further includes a quantitative causal algorithm for quantitative assessment of risk (QAR). The spiral of multiple sub-algorithms converges to form MMP.

The quantitative causal algorithm is composed of a small number of highly related, routinely collected data features (e.g., anatomical and physiological data). For example, diastolic cardiac function that sits at the top of the pathophysiologic cascade is related to the acuity of left ventricular filling pressure and a principal indicator of dynamic heart failure physiology. The measured status of multiple nonlinear diastolic data acts as a surrogate of cause and change of this data-state is a surrogate of effect. The validity of composite data and relationship to the acuity of filling pressure is found in the medical community's data base (i.e., tissue Doppler early diastolic mitral annular velocity (e') is an average of the septal and lateral e'; Early mitral inflow velocity (E) late mitral inflow velocity (A), E/A ratio; mitral E velocity; Deceleration Time (DT); and E/e' ratio calculated using the average e' medial and lateral velocity). Each data has a validated (a posteriori) data range, which depicts the relationship between observable data and the central but unobservable factor that accounts for (causes) the data state. Each data can be measured/sensed by, for example, the echocardiographic machine 105 of Fig. 1.

Fig. 5 shows that the aggregate weight and serial change of the constituent data depict the intensity and variable status of the data-state. There are three simple measures the QAR: weight of the composite data module; weight of abnormal data; and weight of the computed LR. These 3 simple tasks eliminate the need for more complex tasks.

A LR is the likelihood of a given test result in an individual with a disease compared with the likelihood of this result in individuals without the disease (e.g., a person with an emergent disease is more likely to have an abnormal LR than a healthy individual). The size (weight) of this variance has clinical importance.

LR = Disease Data Weight / Risk Module Composite Weight

In some embodiments, LR risk ranges between zero (0.0; low) to 1.0 (high). Finding a LR between 0 and 0.25 argue against a risk event and as the LR approaches 1.0 the greater the likelihood of an adverse event. LRs between 0.25 and 0.75 encompass a continuum from mild to moderate and then to high risk. The polar extremes (i.e., low and high LR) of probability indicate diagnostic certainty for most clinical problems. Sequential LRs are an easily understood means of tracking risk management and outcome. The main advantage of LRs (over other measures of diagnostic accuracy, such as sensitivity and specificity) is that clinicians can use them to quickly and sequentially compare different management strategies and thus rapidly refine clinical decisions.

When LRs are measured in sequence, the post-test odds of the first test become the pretest odds for the second test, and so on. LRs can compare the individualized continuum of risk for the same risk event and when compared to different clinical settings LR comparison discriminates which individual is at most risk and most responsive to management. LRs can provide the best measure of outcome where clinicians can easily take advantage of LRs and thereby apply the lessons and insights for published studies to their own individual management decisions. Further testing should be only used when they will affect

management. If an individual's pretest probability of disease securely rules out a significant risk state (LR 0 to 0.25), additional testing is unwarranted. Further testing should be considered in the middle zones where early preventive medial intervention may be best applied. The largest payoffs of LRs stem from early forecasting and monitored outcome using inexpensive, easy to interpret analytics, which have important consequences useful for the everyday practice of individualized evidence based PM that is not based on population based knowledge.

Fig. 5 shows a step-by-step calculation 500 of LRs. It will be appreciated that Fig. 5 shows an example of calculating LR risk-state (probabilistic cause) and sequential LR change (effect). As shown in Fig. 5, A, B, C, D, E, and F are case examples representing

participant's multi-feature diastolic function. In Fig. 5, the term "Integer" indicates a number representing the published data range for a specific data feature; "e"' is the tissue Doppler early diastolic velocity; "E/A ratio" is Early mitral velocity ÷ Atrial trans-mitral flow velocity; "DT" is declaration time of the early mitral diastolic inflow velocity; "E/e"' is the surrogate measure of LV filling pressure; "LR" is the Likelihood Ratio; "A, B, C, D, E, F" are serial case examples of LR intensity (i.e., diastolic function) and worsening LR over a 3- year interval.

In Fig. 5, LR Quartiles separate the numerical risk model into a more familiar risk profile of normal, and mild, moderate, high risk. Sequential LR data-states provide a display of the change in risk status (in the example each individualized risk state deteriorated (LR increased)). Sequential LRs depict the trajectory of the individualized risk state (e.g., unchanged, improved, or as shown deteriorated).

The first step is to construct a data module (e.g., Diastolic Function: Echo e' septal and lateral, E/A ratio, Deceleration Time, E/e' ratio). In the second step, each data is assigned integer based on its recorded data range. For example, 0 (zero) for normal and 1- mild, 2-moderate and 3-high risk range; use of absolute data values is unnecessary. In the third step (module weight), the sum of data integers equals the individual risk weight. In the fourth step (causal LR), the module weight is divided the total weight of the module (LR size is a surrogate of probabilistic cause: greater the LR the more likely to observe an adverse event); Causal LR replaces diagnostic risk factors as a measured expression of

pathophysiologic risk and monitored outcome. In the fifth step (effect LR), sequential LR change is a surrogate of effect (i.e., effect is the serial change in the causal LR, which can only occur in the presence of a cause (i.e., counterfactual)).

Using a rule-out echo (e.g., SET), about 90% of high risk disease associations can be effectively ruled-out by a specialized echo exam and monitored management. The separation between normal and abnormal data-states is the simplest and strongest means of appreciating the presence of emergent risk.

The most important advent in individual risk prediction is the measured continuum of functional, anatomic and hemodynamic causal factors as opposed to a population based risk management. Individual casual risk is a multivariable, amenable to monitored outcome, and assisted algorithmic management. Defining the roles of existing and new management tools can only take place in the setting of accurate cause and effect risk stratification and monitored outcome.

MMP has reliable models for preclinical testing and means of monitoring responses to prevention measures (PM). PM includes validating a high quality, cost-effective screening exam that has the capacity to unequivocally rule-out emergent risk in low risk asymptomatic individuals; demonstrating how a SET can assist and greatly improve clinical data interpretation and understanding; introducing the concept of sequential quantitative risk assessment and monitored outcome; and introducing the transition to quantitative causal-risk factors and away from diagnostic-risk factors alone.

A prognostic SET test can be used to predict an individual's likelihood of developing a disease or experiencing a medical event. Some of the most commonly used statistics for evaluating diagnostic tests, such as point estimates of test sensitivity and specificity and receiver operator characteristic curves, are not as informative as prognostic tests. The most pertinent summary statistics for prognostic tests are the time-dependent observed within clearly defined prognostic groups, the closeness of the group's predicted probabilities to the observed outcomes, and how use of a dynamic prognostic test reclassifies individuals into different prognostic groups and improves predictive accuracy and overall individual outcomes. SET likelihood ratios (LR) are a simple, reproducible means of measuring the continuum of causal (cardiomyopathic) risk.

The methods and systems center around one basic operation, node aggregation. In operation, a small cluster of nodes in a highly-related network is replaced with a single node without changing the underlying joint distribution of the network. MMP is a technology for representing and making inferences about beliefs and decisions. A probabilistic Bayesian Network is a directed, acyclic graph in which the LRs are looked upon as variables, and the arcs between sequential LRs represent possible probabilistic dependence between the variable (e.g., evidence based cause and effect). This unique category of linked prognostic tests can be used to predict beneficial or adverse responses to treatment. Ultimately, the cost of not addressing the epidemic of SCA/SCD will far exceed the costs of obtaining the means of advancing the field of risk stratification. The methods cannot be derived from purely mathematical models, and will need major input from clinicians who are familiar with clinical decisions and with the diverse contributions of the technological tests. Improvements can occur if appropriate clinical investigators will collaborate. The computational requirements expand beyond algebraic averaging. Predictive power relies upon machine learning (e.g., by the specialized medical computer device 120 and/or the cloud 135 of Fig. 1). The more data fed into the program or computer, the more it learns, the better the algorithms, and the knowledge it accumulates (smarter machine). Codependence does not use "algebraic averaging" but uses codependent relationships. Continual lifelong accumulation of modules relationships formulates a novel form of "human-like" (artificial) intelligence (expanding trove of inductive and deductive knowledge based modules).

Echo is an accessible, inexpensive test free of direct adverse effects. Echo-Doppler diastolic function sits at the top of the pathophysiologic cascade and occurs ahead of systolic dysfunction, ECG abnormalities, symptoms and death. Echo has been validated to be a first- line means of accessing more than 95% of diseases associated with SCA and SCD. However, specific diagnostic tests, including echocardiographic tests, are known to provide no general benefit and entail unacceptable cost. Conversely, a SET using Echo is a high quality, inexpensive test focused on prognosticating and quantifying the early emergence of pathophysiologic factors. Small numbers of highly specialized data can be aggregated into a single quantitative expression of pathophysiologic risk. The aggregate state of 2 to 4 highly related data are used to compose a high-quality, low-cost (for example, less than $99), about 10-minute SET. The greatest attribute of the SET is the ability precisely and accurately quantify causal risk and monitored outcome.

The most important contribution of the rule-out paradigm shift is quantitative prognostic CV risk stratification and monitored outcome related to therapeutic decisions (rather than searching for a specific diagnosis and institution of empirical therapies). A specialized Echo exam unequivocally assures normal and rules-out infrequent abnormal and prioritizes validated functional, anatomic, and hemodynamic data, that meets all necessary CV screening criteria. The result is a high quality, multi-feature, accessible, low cost, brief screening exam. Unique additional attributes include the ability to automatically compute early individualized risk (i.e., likelihood ratio) and monitor and manage risk based

transformative change (i.e., cause and effect) that numerically defines the continuum causal risk-state between normal and abnormal. Evidence based PM, based on individualized quantitative risk, portends the future of preventive medicine and pursuit of sustainable wellness. The rule-out paradigm shift in CV screening is creative, exciting, and naturally understandable and most of all a transformative solution to a global health care crisis.

MMP functions in multiple medical specialties, diseases, queries, and data. MMP is a universal system, which will evolve as a Master Product (transition of data to knowledge throughout the Health Care spectrum).

The MMP is designed to automatically identify and implant specific SET information based on prior experience and user validation. Computer intelligence will continue to increase and mature with repetitive experience and user acceptance. For example, initially there was single seed algorithm for diastolic function. However, with computer experience and validation multiple alternative diastolic function algorithms were developed. The computer has subsequently been designed to automatically select the most efficient diastolic function data algorithm. The computer now can automatically implement data based on experience, replacement of data, avoidance unapproved data, and question or approve information. MMP is designed to continually add computer (e.g., the specialized medical computer device 120 and/or the cloud 135 of Fig. 1) automation and autonomous computer management of new knowledge and defined functions.

In operation, MMP performs the following steps. The first step is Rule-Out Risk Screening that rules-out and individual with CV risk. The paradigm shift in CV screening replaces specific (rule-in) diagnostic tests with sensitive (rule-out) prognostic tests that establish that the biophysiologic state is normal and an underlying CV disease is unlikely. In an asymptomatic screening population, most subjects are expected be determined to be positive (i.e., normal). A positive response effectively 'rules-out' likelihood of an abnormal state. The rule-out principal is particularly useful when there is an important penalty for missing silent low incidence disease, such as screening for a potentially fatal disease (e.g.,

SCD/SCA). It is also useful in reducing the number of possible diagnoses to be considered in the early stage of a screening workup. This is a cost-effective means of effectively triaging an asymptomatic population into a binary no risk versus risk status. The transition from a specific diagnostic test to sensitive prognostic test solves most of the current screening inadequacies of preventive medicine. Upon finding an abnormal risk state a novel measure of causal risk is then measured and reported. An automated noninvasive multi-data method is used to unequivocally rule out a normal versus abnormal pathbiological, structural, and hemodynamic state. This alone would save millions of dollars by reducing the cost of screening thousands normal people.

What clinicians need is a test that identifies asymptomatic individuals who do not yet have an established disease. The test must function in clinical or public health settings, where the prevalence of SCD is low (e.g., 2 to 6 per 100,000 in young and 1 to 2 per 1 ,000 in adults). In an asymptomatic population, the majority of SETs are expected to be normal. The objective of the first step is to document that each individual is unequivocally normal (i.e., positive finding) and any rare negative would more likely have an emergent disease, which would require further risk assessment.

In testing, using the rule out-sensitive principal, about 1,000 consecutive

asymptomatic young athletes and non-athletes were screened (age 4 through 25 years; mean age 15). 98% of SET functional, anatomic and hemodynamic features were unequivocally normal. An abnormal anatomic finding was identified in 21 subjects (~2%); 18 subjects had a bicuspid and 1 unicuspid aortic valve (9 had mild and 2 moderate aortic dilation, 7 had mild to trivial aortic valve regurgitation, and none had coarctation). Forty eight percent (470 subjects) had an enlarged left atrial volume index (28 to 56 mL/ni2; mean 34 mL/m2) consistent with a benign atrial volume overload (i.e., athletic atrial heart with normal LR). Two subjects with a normal SET had an abnormal AHA questionnaire: both were referred for further expert CV attention.

In testing, the SET was brief (^¾10 minutes), cost less than $99 and was performed outside a medical infrastructure by registered sonographers and computer assisted expert data interpretation and remote physician oversight. Using shared decision, all abnormal findings were discussed, documented and referred for medical attention.

For example, clinical SCD in young individuals is associated with a small portfolio of functional, structural, and hemodynamic effectors. The simplest and most cost-effective means of ruling out low risk in a young asymptomatic population is to confirm that causal factors are unequivocally normal. The most cost-effective objective is to answer a simple binary question: Is QRA unequivocally normal: Yes vs. No? A normal SET takes aboutlO minutes, costs less than $99 and performed outside a costly medical infrastructure by a trained sonographer, computer assisted expert data interpretation and medical back-up. In a young population (<35 years old), more than 95% of SETs will be unequivocally normal, which alleviates cost of further testing. An infrequent abnormal would undergo additional risk stratification and assisted decision making.

Clinical SCD in older individuals (>35 yrs. old) generates a very large absolute number deaths (about 400,000 deaths per year). SDA and SCD should be positioned squarely in the crosshairs of the precision medicine initiative (e.g., prioritize causal risk factors as opposed to diagnostic risk factors). The largest cumulative number of SCAs occurs in the asymptomatic general population. Based on this fact, the most cost-effective priority is to define the individualized causal risk state. Most individuals will have a normal or non- threatening causal risk state, which would preclude the expense of additional testing.

The graphic trajectory of the risk continuum enhances SCA prediction and timely decision making. Treatable CV risk factors, such as hypertension, stroke, atrial fibrillation, Type Two diabetes, hyperlipidemia, obesity and smoking, provide a basis for general risk but have been found to have limited effect on individual risk of SCA/SCD. Over the past few years non-cardiovascular conditions tend to be the most frequent association with SCA. There has been a marked transition toward heart failure with preserved EF (HFpEF) for which there is no specific treatment or means of monitoring outcome at this time.

There is a way to establish that a disease risk is unlikely using a LR. Using the rule out-sensitive principal a positive response affirms normal and negative test affirms the presence of possible disease. The test is clinically most useful when there is an important penalty for missing a disease, such as in screening for the risk of SCD. It also reduces the number of possible diagnoses to be considered. The LR provides quantitative information on the at-risk status. The grater the LR value the greater the risk of having a disease, (e.g., LR as an ordinal test and sequential LR as monitoring test)

MMP using a high-quality, low-cost (<$100), brief (<15 minute) specialized rule-out echocardiographic exam focused on unequivocally affirming a normal physiologic, anatomic and hemodynamic data-state and by doing so, ruling out any emergent abnormal feature. This prognostic exam assesses binary factors and aggregates small numbers of data to arrive at a single integrated pathophysiologic forecast. Probabilistic screening classifies an individual into a positive versus negative prognostic risk-state that simplifies understanding, improves management decisions and measures the accuracy of outcome predictions. The variable intensity of a risk state opens a large range over which prognostic screening can be applied.

In testing, clinical concepts we documented: 1. In an asymptomatic population the majority of young and adult individuals will have an unequivocal normal pathophysiologic profile; 2. An individualized rule out binary-SET focused on normal is an efficient and cost- effective CV screening test where there is an important penalty for missing latent risk such a SCD; 3. Binary rule out LR represents a simple high-quality means of risk stratification. However, dichotomizing individuals into a binary designation does not represent the full clinical situation. After determining the presence of an abnormal risk state, additional causal risk stratification must be obtained before a meaningful impact on the prevention of SCD can be realized.

The second step MMP performs is the Quantitative Risk Assessment. The new paradigm shift introduces simple inputs that address complex problems, and generate simple outputs that users can easily understand, interpret and apply. In pursuit of these objectives, 50,000 clinically indicated echocardiographic exams that were interpreted by clinical echo- cardiologists and then blindly re-interpreted by SET were reviewed. See FIG 4. The aim was to show how LRs can assist decision making based on quantitative computer risk assessment. Fig. 6A shows a LR Assists Data Interpretation 600. The vertical coordinate represent the number of echocardiographic reports. The SET is meticulously designed to have 100% precision (reproducibility) and accuracy (validation of intended use). In 50,000 case examples, the SET algorithm automatically excluded 29%) of studies because of insufficient data. The SET automatically interpreted and validated the remaining 71% of case studies. However, the human (Doctor of Medicine "MD") interpreters elected not to interpret 51% of the case studies, which were interpreted by the SET (1 1% normal, 41% mild, 33% moderate and 15% severe). In the 49% of SET versus human interpretations only 28% of human interpretations were concordant with the SET interpretation. The ineptitude of human interpretations was extremely sobering yet consistent with published literature, which has reported interpretation errors between 20% and 80%. Empirical data interpretation based on subjective decisions is difficult to assess or correct and do not stand up to rigorous scientific scrutiny. The SET is a real-world example of a carefully designed quality improvement tool capable of assisted interpretation of clinical data. Quantitative LRs add considerable validity to the data interpretation process.

In Figs. 6A and 6B, a total of 35,577 echocardiographic reports compared human interpretation versus computer assisted interpretation. Correlation 610 between individual human and computer interpretation of identical data was extremely poor. Of the 51%» with no human interpretation the majority (89%) had some degree of diastolic dysfunction (mild 4P/o, moderate 33%, severe 15%). Of the 49%) with a human interpretation severe diastolic dysfunction agreed only 10% of the time. These sobering findings show the power of a simple system's approach to the interpretation complex pathobiological systems.

The heterogeneity of complex data and inadequacies of existing statistical methods has driven physicians toward empirical decision making, which is fundamentally inadequate and increasingly unacceptable. Conversely, causal LRs create a systematic bias towards simplicity while implementing rigorous risk quantification and genuine decision support for risk management. The next generation of decision makers can use the LR solution, which prognosticates the risk of, for example, SCD by assembling and weighing specific variables and by combining each piece of information, arrive at a single integrated forecast of pathophysiologic risk.

LR inductive reasoning creates a systematic bias towards simplicity while retaining consistency with data and validated logic. An important goal of LR causal modeling is to unravel complex data generating processes and place emphasis on monitoring the

manipulation of cause and effect risk relationships. Monitoring risk instead of statistical CV correlations is essential because most SCD events are actually related a complex of both non- CV and CV effectors.

By collapsing nonlinear data into discrete categories destroys valuable information. LRs define a continuum of individual variances of the pathophysiologic state that can assist implementation of evidence based decision making. Simple graphical presentation of sequential data-states can replace and enhance the understanding of complex probabilities.

LRs can measures the distance or separation between two risk states, which depicts the effect of causal factors. The distance of an effect displays improvement vs. deterioration of an individual's risk-status. Graphic display of LR distance enhances understanding and implementation of cause and effect decision making.

Temporal sequencing of LRs is crucial; a clinical effect cannot precede or occur without simultaneously changing its cause. The post-test state of the first LR becomes the pretest state for the second test, and so on. The dynamic change of risk (i.e., effect) initiates evidence based decision making. At some instant, each LR becomes a cause. The concept of circular causality defines the pattern of self-organizing dynamics. Clinical physiology is a complex system with distributed nonlinear feedback. The most critical question about risk is its degree of stability or resistance of change. The distance of risk displacement defines stability, deterioration or improvement of the risk-state. The trajectory of change defines the success or failure the elected management.

There is a fundamental difference between a statistical correlation and causal inference when assessing the outcome of causal risk. A risk factor is a variable with a significant statistical association with a clinical outcome. Many clinicians mistakenly believe that these statistical relationships of outcome suggest causality and mistakenly make the risk factor more relevant. Statistical models are useful in predicting (estimating the likelihood of) a particular disease outcome but are never useful for ascertaining causal change of the risk- state. Causal factors are not defined statistically but are defined experimentally in that they are proven to affect the outcome and not merely observe the outcome. Documentation of a state change is fundamental to a cause and effect inquiry. A SET can document and quantify the status (e.g., LR) of a dynamic pathophysiologic process. Current statistical outcome studies merely collect data and impute association, which physiologically cannot define cause and effect or quantify the response to therapy.

To demonstrate evidence based cause and effect outcome, individual LRs were retrospectively plotted at yearly intervals. Sequential LRs demonstrate functional deterioration, stability or improvement of an individualized pathophysiologic outcome. The graphic display of causal factors is easy to understand and quickly validates the therapeutic or environmental state change of a complex pathophysiologic process (e.g., sequential status diastolic function).

There is a wide individual variance of risk within a population, which cannot be appreciated or parsed in population based risk factor statistics. No statistical test can tell clinicians the appropriate quantitative response at different ranked levels of test outcome. LRs defined a continuum of individual variances of the pathophysiologic state that could assist implementation of evidence based decision making. Simple graphical presentation of sequential data-states can assist easier appreciation of the dynamic status of complex probabilities. The distance of an effect displays improvement vs. deterioration of an individual's risk-status. See Figs. 5A-5C. The numerical distance or separation between two risk states depicts the change of causal factors. At some instant, each LR became a cause. The continuum of causal risk defined the pattern of self-organizing dynamics. The most critical question about risk is the degree of stability or resistance of change. The distance of risk displacement defined stability, or deterioration or improvement of the risk-state. The trajectory of change defines the success or failure the elected management.

Figs. 7-9 show analytic causality and serial transformation of the LR. LR score changed its risk-state with each successive test. Each effect becomes a new cause and over time changes to a new or preexisting state (effect). The sequential state change defines the magnitude and direction of the state-change. Examples of individual LR analytics (e.g., deterioration 700 (Fig. 7), stabilization 800 (Fig. 8) or improvement 900 (Fig. 9)) should alert the status of clinical management. State changes are bidirectional consistent with the concept of causality. Serial monitoring of a state change can assist appreciation of environmental and treatment effectors. It will be appreciated that in each of Figs. 7-9, six participants' LR analytics are shown.

Figs. 10A and 10B show data validation features 1000 associated with SCD, according to some embodiments. Fig. 11 shows a five-step rule-out echo exam 1100, according to some embodiments. It will be appreciated that Figs. 10A, 10B, and 11 address specific features and should not be construed as the only type of functions uses in this new probabilistic technology.

In Figs. 10A and 10B, "Diastolic function data" include e' mitral septal and lateral tissue velocity, E/A ratio of early and late mitral filling velocity, DT (mitral E deceleration time), and E/e' (surrogate measure of LV filling pressure). AV stands for aortic valve;

ARVC stands for arrhythmogenic right ventricular cardiomyopathy; BAV stands for bicuspid aortic valve; CAD stands for coronary artery disease; CHD stands for congenital heart disease; ECG stands for electrocardiogram; Echo stands for echocardiogram; HCM stands for hypertrophic cardiomyopathy; HFpEF stands for Heart Failure with preserved ejection fraction; HTN stands for hypertension; LVH stands for left ventricular hypertrophy; PAT stands for Pul. Acceleration Time; Pul. stands for pulmonary; R/O stands for rule out; SCD stands for sudden cardiac death; and UAV stands for unicuspid aortic valve.

In Fig. 11, AHA stands for American Heart Association; ARVC stands for

Arrhythmogenic Right Ventricular Dysplasia/Cardiomyopathy; BAV stands for bicuspid aortic valve; CV stands for cardiovascular; LVO stands for left ventricular outflow; LR stands for likelihood ratio; Pul stands for pulmonary; RVO stands for right ventricular outflow; R/O stands for rule out; and UAV stands for unicuspid aortic valve.

In an embodiment, the MMP can be an interactive device (e.g., the specialized medical computer device 120 of Fig. 1). The device can have a processor, a memory, and/or storage for data store. The device can include wire or wireless network module that connects to a local area network or the Internet. The device can also include a GUI (graphic user interface) or another interface that allows a user to interact with the device. The device can also have sensors that can sense a user's biological signals (for example, heart rate, respiration intervals, etc.). The device can further include or connect to an echo machine or other data acquisition devise to detect the user's other biological signals (for example, cardiac ultrasound). The device can also include or connect to other equipment such as scale to detect the individual's other biological parameters (for example, weight, body mass index, etc.).

In operation, the processor can detect individual's biological signals and/or biological parameters via the sensors, the echo machine, or other acquisition equipment. The user or computer can seed the processor related types of data (for example, SCD) via the user or embedded modular interface. The acute and chronic disease data can also be predetermined and pre-stored in the memory or storage of the device. The processor can perform machine learning to identify highly related data (for example, tissue Doppler early diastolic velocity) that correspond to the disease state via for example, the Internet. In some embodiments, the processor can have the user or automated input function input any missing information (for example, biological signals, biological parameters, age, etc.) via the user or intelligent interface or obtain such missing information via sensors or equipment the device connects to (for example, the echo machine for echo exam described above). The processor can calculate the LR risk-state and sequential LR change based on the embodiments described above. In some embodiments, the device can include a display (for example, screen or graph) or a speaker or connects to a printer to display or print the LR information and/or any other related information (for example, referring for further expert CV or medical attention).

Fig. 12 is a flow chart illustrating a method 1200 for determining a risk state for a disease, according to one embodiment. The method 1200 includes 1205 measuring anatomical and physiological data. The anatomical and physiological data can include, for example, data collected/measured/sensed via the input device 115 of Fig. 1, data

measured/sensed via the sensor 110 of Fig. 1, and/or data measured/sensed from the echocardiographic machine 105 of Fig. 1. The method 1200 also includes 1210 determining a LR based on the measured anatomical and physiological data (see Fig. 5). The LR can be determined by, for example, the specialized medical computer device 120 of Fig. 1. The method 1200 further includes 1215 determining a risk state based on the LR. The risk state can be determined by, for example, the specialized medical computer device 120 of Fig. 1. Also the method 1200 includes 1220 outputting the risk state by, for example, the specialized medical computer device 120 of Fig. 1, to for example, the output device 130 of Fig. 1. The anatomical and physiological data, the LR, and/or the risk state, and other participant information can be stored in, for example, the local storage 125 and/or the cloud 135 of Fig. 1.

In sum, MMP collects large amounts of data and then refines small amounts of logical data into related knowledge, which assist and accelerates medical decision making. MMP is one of the most revolutionary medical technological products to be introduced in the past one hundred years. Multiple sub-algorithms work in concert to orchestrate a singular adaptive master learning algorithm. A mature MMP is destined to become one of the greatest data management transformations in medicine. Laboratories have affirmed MMP's high precision and accuracy, which is embedded during the design processes. When logical data is provided, the MMP can automatically compute the corresponding knowledge. The new paradigm shift introduces simple inputs that address logical complex problems, and generate simple probabilistic outputs that users can easily understand, interpret and apply. The automated transformation of data into knowledge will markedly improve the health care community to make more informed decisions. It will illuminate the understanding of pathophysiology and assist the prevention of the emerging health care crisis. Medical literature strongly supports the need for a cataclysmic transition to electronic algorithms that unify logic and probability and rely less on static diagnostic studies and regression statistics. Pilot studies show that MMP data interpretation is superior to empiric human data interpretation, that embedded test precision (reproducibility) and accuracy (truthfulness) in MMP reduces human error, that quantitative wellness is a dynamic continuum but diagnostic risk is a static observation, that status of risk is more clinically relevant than static diagnostic assessment, that serial LRs compute change in outcome but serial diagnostic interpretation is inconsistent, that probabilistic risk fosters precision medicine but diagnostic risk is population based, that rule-out test is brief, inexpensive, high quality and able to predict but diagnostic tests are not; and that screening tool continuously learns and prioritizes wellness but diagnostic tests prioritize illness.

Pilot studies affirm that MMP affirms implementation of a superior screening exam.

MMP can eliminate a number of obstacles (see below) that make current screening technologies impossible to alleviate an active health care crisis, which is attributed to the lack of a true preventive medicine solution.

Obstacle 1 : The medical community is overly influenced by anecdotal evidence. Screening tests currently focus on static diagnostic associations, which can be considered unethical. MMP: Science advances mostly via induction from facts that support a theory. The MMP quantifies a continuum of knowledge-based risk. An infrequent negative response 'rules out' the target normal and quantifies the status of preclinical risk.

Obstacle 2: A strong statistical association is commonly mistakenly assumed to provide prognosis or imply causality. MMP: Causality is induced logically and is not a statistic or empirical observation. The MMP applies Bayesian methodology to quantify the risk continuum and the effect of change (i.e., cause and effect).

Obstacle 3: Over-diagnosis and unnecessary treatment is major medical problem. MMP: The MMP determines how knowledge is modified by data; eliminates false positive and false negative information; markedly lowers cost; individualizes a continuum of risk; and assists the implementation of knowledge-based decisions. These attributes alleviate overdiagnosis and assists knowledge-based decision making.

Obstacle 4: Current screening tools are unacceptable because of unsustainable expense, false positive data and inability to monitor outcome. MMP: The MMP monitors, markedly lowers cost, increases quality and assist the individual management risk. The MMP focuses on sustainable individual wellness and away from a failed search for illness.

Obstacle 5: Screening options are not standardized. Priorities and obstacles vary significantly across the socioeconomic strata. MMP: The MMP defines a continuum self- similar risk-classification. Risk is viewed as a combinatorial problem in which many different defects result from a universal model. The overarching focus of the MMP is early prediction and prevention using a universal classification of self-similar risk.

Obstacle 6: High precision (reproducibility) and accuracy (truthfulness) of screening has been impossible to achieve. Even with high diagnostic specificity and statistical association, false positive results and negative consequences too frequent. MMP: The MMP has embedded simulation technology that continuously assures highest precision and accuracy. The result is a quantitative probabilistic risk continuum that alleviates uncertainty and reduces decision errors.

Obstacle 7: The prevention of non-communicable disease (e.g., SCD) has remained an unremitting crisis for decades. Nothing has worked satisfactorily. MMP: The MMP draws causal conclusions and not correlations. Simple graphical 'nomograms' of the risk-state replace complexity. The MMP, which is based on probabilistic causality, is less subjective, less irrational and more accurate than other technologies.

Obstacle 8: The search for end-stage events and relative risk do not improve prognosis or sustain wellness and are subject to marked overestimation of expected benefit. MMP: The MMP defines absolute risk reduction where the event incidence is compared to screening participants who do not have the risk state (i.e., likelihood ratio). Absolute risk surveillance is more relevant and cost-effective than verification trials.

Obstacle 9: Treatment bias, high cost and lack of benefit are pervasive in current screening tools. MMP: Probabilistic causality does not have the costly side effects.

Attention is given to measured causal risk and monitored effect of risk management.

Obstacle 10: Prevention bias occurs when a feature is treated in the absence of proven cause and there is no certainty of benefit or treatment effect. Incremental modifications in the prevailing screening has not solved the crisis or substantially benefited outcome. MMP: Causal probabilistic models provide rigorous risk quantification and knowledge-based decision making. Benefit is measured by a change in individual risk. Cost is markedly reduced by treating a defined risk- state, which is most amenable to change. The objective of the MMP is sustained wellness and not a failed search for illness.

MMP is a systematic approach to identify individuals at 'risk' and determine who responds to specific interventions. The critical principles in decision making link the benefits of measuring individual risk with those of treatment, which has historically been obfuscated by the inability to quantify prior risk and posterior outcome. The benefits and costs of screening chronic non-communicable disease are more dependent upon the status of a risk- continuum than upon individual diagnostic morphology. When people use causal decision aids, they improve their knowledge of the options (high-quality evidence) and become better informed and clearer about what matters most (high-quality, low cost evidence). MMP enables the assisted decision making.

Selective screening tests tend to be thought of as a cross-sectional, short-term operation, is not a suitable means of addressing the global incidence of latent sudden life- threatening events. Beneficial surveillance must be based on mitigation of quantifiable risk using individualized cause and effect decision making (i.e., precision medicine). Based this objective, there is sufficient justification to institute high quality, low cost quantitative risk surveillance beginning in childhood (e.g., age 5 years) and at specified intervals throughout the life-cycle. To make a socioeconomic medical care difference to millions of individual people, MMP can gain better insight into probabilistic causality that drives and changes an individual's underlying 'physiologic risk state'.

Glossary

A wave - late mitral inflow velocity

a posteriori - reasoning or knowledge that proceeds from observations or experiences to the deduction of probable causes (posterior, known knowledge).

a priori - reasoning or knowledge that proceeds from theoretical deduction rather than from observation or experience (prior, prognostic knowledge).

BT - Bayesian Tool

Causality - refers to the relationship between events where one set of events (the effects) is a direct consequence of another set of events (the causes)

CD - Chronic Disease

Counterfactual - effect is the serial change in the causal LR, which can only occur in the presence of a cause

CT - Causality Tool

CV - Cardiovascular

DT - Deceleration time

Echo machine - echocardiogram/ultrasound/echocardiographic machine

E wave - Early mitral inflow velocity

e' - tissue Doppler early diastolic mitral annular velocity

E/A ratio - Ratio of E and A waves

E/e' ratio - Ratio of E and e' waves

GUI - Graphic User Interface LDT - Logic Data Tool

LR - Likelihood Ratio

MMP - Master Medical Product

PM - Prevention Measures

QAR - Quantitative Assessment of Risk

SaaS - Software as a Service

SBT - Systems Biology Tool; Network Medicine

SCA - Sudden Cardiac Arrest

SCD - Sudden Cardiac Death

SET - Specialized Echocardiographic Test

VT - Validation and Verification Tool

Aspects:

It is appreciated that any of aspects 1-6, 7-15 and 16-23 can be combined.

Aspect 1. A medical screening system, comprising:

a specialized medical computer device;

a sensor;

an echocardiogram machine; and

an output device,

wherein the specialized medical computer device is configured to receive inputs from the sensor and the echocardiogram machine,

the output device is configured to output data from the specialized medical computer device,

the sensor is configured to measure a first data,

the echocardiogram machine is configured to measure a second data,

the specialized medical computer device is configured to generate a likelihood ratio

(LR) based on the second data,

the specialized medical computer device is configured to generate a risk state based on the LR and the first data, and

the output device is configured to output the risk state. Aspect 2. The medical screening system according to aspect 1, wherein the second data include at least two data, the at least two data are associated with each other for a disease, each of the at least two data is a marker of the disease, and the marker of the disease includes a biomarker.

Aspect 3. The medical screening system according to aspect 1 or aspect 2, wherein the second data include data for a diastolic function.

Aspect 4. The medical screening system according to any one of aspects 1-3, wherein the second data include tissue Doppler early diastolic velocity (e'), early mitral velocity and atrial trans-mitral flow velocity ratio (E/A ratio), declaration time of the early mitral diastolic inflow velocity (DT), and surrogate measure of LV filling pressure(E/e' ratio).

Aspect 5. The medical screening system according to any one of aspects 1-4, wherein the sensor is a blood pressure sensor.

Aspect 6. The medical screening system according to any one of aspects 1-5, wherein the risk state indicates a low risk, a mild risk, a moderate risk, or a high risk for a disease associated with the second data.

Aspect 7. A specialized medical computer device, comprising:

a processor;

non-transitory computer readable storage mediums; and

specialized medical computer program instructions stored in the non-transitory computer readable storage mediums,

wherein the processor is configured to execute the specialized medical computer program instructions,

the processor is configured to obtain measured anatomical and physiological data, the processor is configured to generate a likelihood ratio (LR) based on the measured anatomical and physiological data,

the processor is configured to generate a risk state based on the LR, and

the processor is configured to output the risk state. Aspect 8. The device according to aspect 7, wherein the measured anatomical and physiological data include at least two data, the at least two data are associated with each other for a disease, each of the at least two data is a marker of the disease, and the marker of the disease includes a biomarker.

Aspect 9. The device according to aspect 7 or aspect 8, wherein the measured anatomical and physiological data include data for a diastolic function.

Aspect 10. The device according to any one of aspects 7-9, wherein the measured anatomical and physiological data include tissue Doppler early diastolic velocity (e'), early mitral velocity and atrial trans-mitral flow velocity ratio (E/A ratio), declaration time of the early mitral diastolic inflow velocity (DT), and surrogate measure of LV filling pressure(E/e' ratio). Aspect 11. The device according to any one of aspects 7-10, wherein the measured anatomical and physiological data include data measured via an echocardiogram machine.

Aspect 12. The device according to any one of aspects 7-11, wherein the measured anatomical and physiological data include data measured via a sensor.

Aspect 13. The device according to aspect 12, wherein the sensor is a blood pressure sensor.

Aspect 14. The device according to any one of aspects 7-13, wherein the specialized medical computer program instructions include a logic data tool, a system biology tool, a causality tool, a Bayesian tool, and a validation tool.

Aspect 15. The device according to any one of aspects 7-14, wherein the risk state indicates a low risk, a mild risk, a moderate risk, or a high risk for a disease associated with the measured anatomical and physiological data.

Aspect 16. A method of determining a risk state, the method comprising:

measuring anatomical and physiological data; determining a likelihood ratio (LR) based on the measured anatomical and physiological data;

determining the risk state based on the LR; and

outputting the risk state.

Aspect 17. The method according to aspect 16, wherein the measured anatomical and physiological data include at least two data, the at least two data are associated with each other for a disease, each of the at least two data is a marker of the disease, and the marker of the disease includes a biomarker.

Aspect 18. The method according to aspect 16 or aspect 17, wherein the measured anatomical and physiological data include data for a diastolic function.

Aspect 19. The method according to any one of aspects 16-18, wherein the measured anatomical and physiological data include tissue Doppler early diastolic velocity (e'), early mitral velocity and atrial trans-mitral flow velocity ratio (E/A ratio), declaration time of the early mitral diastolic inflow velocity (DT), and surrogate measure of LV filling pressure(E/e' ratio). Aspect 20. The method according to any one of aspects 16-19, wherein the measured anatomical and physiological data include data measured via an echocardiogram machine.

Aspect 21. The method according to any one of aspects 16-20, wherein the measured anatomical and physiological data include data measured via a sensor.

Aspect 22. The method according to aspect 21 , wherein the sensor is a blood pressure sensor.

Aspect 23. The method according to any one of aspects 16-22, wherein the risk state indicates a low risk, a mild risk, a moderate risk, or a high risk for a disease associated with the measured anatomical and physiological data.

The terminology used in this specification is intended to describe particular embodiments and is not intended to be limiting. The terms "a," "an," and "the" include the plural forms as well, unless clearly indicated otherwise. The terms "comprises" and/or "comprising," when used in this specification, specify the presence of the stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, and/or components.

With regard to the preceding description, it is to be understood that changes may be made in detail, especially in matters of the construction materials employed and the shape, size, and arrangement of parts, without departing from the scope of the present disclosure. The word "embodiment" as used within this specification may, but does not necessarily, refer to the same embodiment. This specification and the embodiments described are examples only. Other and further embodiments may be devised without departing from the basic scope thereof, with the true scope and spirit of the disclosure being indicated by the claims that follow. The MMP uses data for all sorts of input devices and then with structured commands, automate computer instructions to transform the data in to usable knowledge. The knowledge is stored as a user and general health care resource of knowledge, which can be used for diverse reasoning processes.

Claims

CLAIMS What is claimed is:

1. A medical screening system, comprising:

a specialized medical computer device;

a sensor;

an echocardiogram machine; and

an output device,

wherein the specialized medical computer device is configured to receive inputs from the sensor and the echocardiogram machine,

the output device is configured to output data from the specialized medical computer device,

the sensor is configured to measure a first data,

the echocardiogram machine is configured to measure a second data,

the specialized medical computer device is configured to generate a likelihood ratio

(LR) based on the second data,

the specialized medical computer device is configured to generate a risk state based on the LR and the first data, and

the output device is configured to output the risk state.

2. The medical screening system according to claim 1, wherein the second data include at least two data, the at least two data are associated with each other for a disease, each of the at least two data is a marker of the disease, and the marker of the disease includes a biomarker.

3. The medical screening system according to claim 1, wherein the second data include data for a diastolic function.

4. The medical screening system according to claim 1, wherein the second data include tissue Doppler early diastolic velocity (e'), early mitral velocity and atrial trans-mitral flow velocity ratio (E/A ratio), declaration time of the early mitral diastolic inflow velocity (DT), and surrogate measure of LV filling pressure(E/e' ratio).

5. The medical screening system according to claim 1, wherein the sensor is a blood pressure sensor.

6. The medical screening system according to claim 1, wherein the risk state indicates a low risk, a mild risk, a moderate risk, or a high risk for a disease associated with the second data.

7. A method of determining a risk state, the method comprising:

measuring anatomical and physiological data;

determining a likelihood ratio (LR) based on the measured anatomical and

physiological data;

determining the risk state based on the LR; and

outputting the risk state.

8. The method according to claim 7, wherein the measured anatomical and physiological data include at least two data, the at least two data are associated with each other for a disease, each of the at least two data is a marker of the disease, and the marker of the disease includes a biomarker.

9. The method according to claim 7, wherein the measured anatomical and physiological data include data for a diastolic function.

10. The method according to claim 7, wherein the measured anatomical and physiological data include tissue Doppler early diastolic velocity (e'), early mitral velocity and atrial trans- mitral flow velocity ratio (E/A ratio), declaration time of the early mitral diastolic inflow velocity (DT), and surrogate measure of LV filling pressure(E/e' ratio).

11. The method according to claim 7, wherein the measured anatomical and physiological data include data measured via an echocardiogram machine.

12. The method according to claim 7, wherein the measured anatomical and physiological data include data measured via a sensor.

13. The method according to claim 12, wherein the sensor is a blood pressure sensor.

14. The method according to claim 7, wherein the risk state indicates a low risk, a mild risk, a moderate risk, or a high risk for a disease associated with the measured anatomical and physiological data.