Cancer Prediction is the Pathway to Prevention

/in /by

Examining the benefits of using a systems-based approach to predict and control the evolution of cancer through preventative pathways

Oncology research exposes cancer as a dynamic disease that results from complex interactions between multiple scales of genetic, environmental and constitutional characteristics that are unique to an individual. Currently, the full scope of determinants that cause numerous point mutations to accumulate and/or structural alterations to occur in the process of tumor progression are not fully understood. Shifting from a component to a system-level perspective will help add the context and understanding necessary to make sense of and apply current oncology research innovations.

Preliminary use cases have shown that X-ACT® Health is able to add the missing context of complex cancer processes and promote further understanding of how dynamic interactions between a large number of determinants, such as DNA, pathogens, autoimmune responses, metabolism, environment and aging influence the progression of cancer for an individual patient.

Bridging silos of cancer knowledge will be necessary to help researchers and healthcare providers understand the complex factors that provoke cancer in one individual, and not another. This level of understanding is key to predictively diagnose cancer before any symptoms appear, and would give governments and payers the confidence they need to invest in and standardize personalized preventive pathways.

By using sophisticated algorithms that are computationally efficient and cover many-to-many time dependent relationships, X-ACT Health could revolutionize how we diagnose and treat cancer. The goal isn’t to replace doctors with AI. Instead, we need to build new knowledge and predictive capabilities that enable all healthcare stakeholders—including governments, payers, suppliers and providers—to know when and which actions will help an individual patient avoid cancer. This is the Holy Grail of oncology research and provides the path to cut the long-term costs of cancer care and improve standards of patient care.

 

 

Covid-19: Worldwide Viral Infection Model

/in , /by

Examining the benefits of using Universal Dynamic Engine (UDE) to manage Covid-19 patient risk and critical treatment load.

When news broke of an obscure respiratory disease emerging from the Wuhan market in early January, few imagined that in just over 2 months the world would be facing the worst pandemic since Spanish flu in 1918. With more than 12M people taking commercial flights every day, Covid-19 has spread around our hyper-mobile world with an unprecedented combination of speed and virulence. Further, without widespread testing for Covid-19, it is difficult to know how the pandemic is spreading and how to appropriately respond to it. While it is certain that the total number of Covid-19 cases is higher than the number of known confirmed cases, the total number of Covid-19 cases is not known.

As the Coronavirus spreads, many are predicting that the demand for beds, ventilators and other treatment resources in hospital will far exceed capacity for Covid-19 patients in the near future. Still, public policy and health community responses to Covid-19 are complex decisions that require a delicate balance between protecting health, the economy, and people’s well-being and emotional health. In times of crisis, reliance on theoretical models present challenges for decision makers, because they don’t always replicate real life well—especially when the virus that causes the pandemic is novel.

Covid-19 Modeling: Dynamics Shouldn’t be Ignored

Healthcare systems are not closed loop, linear systems. Therefore, they cannot be accurately modeled without accounting for various elements (components) of the system which continually influence one another (directly or indirectly) to maintain the system in line with the goals of the system. In the case of a pandemic, these goals include slowing the spread of the disease and providing the highest quality of care possible for those infected.

If the relationships between various components of a dynamic system are not well understood, changes made to one part of the chain may affect another in unwanted ways. For instance, treating Covid-19 patients without understanding the influence of pre-existing conditions may place more strain on already overwhelmed medical personnel and hospital resources. A universal mathematical formulation is needed to accurately represent all system dynamics in all cases to provide healthcare system participants with a more complete view of reality. This allows stakeholders to make decisions with more confidence in the outcomes than would be possible if using mechanistic models that may provide a wide range of possibilities and miss new risks due to their dependence on historical data.

UDE and NARS Modeling Capabilities

Universal Dynamics Management (UDE) provides a rigorous process that computes the top-down communicating graphs to deal with the direct and indirect, convergent or degenerative solutions necessary to accurately model the complex, adaptive dynamics of non-linear, open systems. Based on the NARS method of modeling, UDE allows users to accurately represent a high level of interdependent components organized in hierarchy of graphs. Scenario analysis helps modelers construct the remedy and mitigation actions, certify the solution pertinence and discover additional variables and interrelationships that can be dependably used to predict the occurrence of previously known as well as unknown risk behaviors.

Through the use of deconstruction theory, it becomes clear that the overall performance of a system (defined in terms of quantity, quality and cost) does not always represent the expected output of the system given the accumulated characteristics of the constituent components, generally due to a loss in energy. Causal deconstruction allows modelers to uncover results that often defy the common wisdom that stops at the wrong level of analysis and usually produces a host of misleading conclusions.

URM Forum Covid-19 Research

URM Forum has published a discussion paper entitled, “Covid-19: Worldwide Viral Infection Model” as part of our ongoing effort to help government, healthcare and business leaders build the UDE capabilities necessary to accurately represent all system dynamics in all cases. The paper includes results for the digital twin of Covid-19 Crisis Management in Germany, referred to as the CV-19 model, to show how UDE can be applied to help provide critical decision insights as needed to guide actions for any country’s or geographic region’s response to the current pandemic as well as monitor and plan activities in order to avoid a future crisis.

The aim of the presented CV-19 model is to show healthcare system participants how they might use UDE to gain a more complete view of reality as needed to make decisions with more confidence in the outcomes. UDE has been successfully applied for a wide range of healthcare, economic and business uses as the preferred way to predictively characterize dynamic systems under specific conditions; identify singularities and metrics, which define the characteristics of the system under different scenarios of increasing volume and structural changes; and validate the applicability of any proposed corrective solutions.

Join Our Efforts to Deliver Improved Risk Modeling Solutions

The dynamics of our mobile and hyperconnected world amplify the risks. Through the participation of a larger number of constituents, the CV-19 model can be expanded to identify thresholds and help support preparedness for a wider variety of parameters and situations as they relate to any dynamic system, which is influenced directly and indirectly by any number of internal and external forces—including travel patterns, weather, government response and/or socio-economics.

If you are interested in joining our efforts to develop scientific-based solutions to society’s biggest risk management challenges, please contact us.

 

Learn more about our Covid-19 Modeling and Healthcare Decision Support research

Read the “Covid-19: Worldwide Viral Infection Model” paper to learn how UDE is being applied to rovide critical decision insights as needed to guide actions for any country’s or geographic region’s response to the current pandemic as well as monitor and plan activities in order to avoid a future crisis.

Paper

 

 

 

 

Identifying Cost Saving Opportunities in Healthcare

/in , /by

The goal of many governments is to continuously improve the public health system efficiency by increasing preventive and proactive intervention, reducing any unnecessarily overhead due to multiple analysis and even diagnosis for the same case, and consolidating patient history to improve preparedness.

In a client case involving a government sponsored healthcare program, each citizen was attached to a binder that included all of his or her health and drug history, time series analyses, medical procedures, medical attributes and some projections that could help with follow-up tracking. This binder was available to medical and pharmaceutical personnel and was synchronized in case multiple doctors were involved. The gigantic national infrastructure necessary to support this level of information sharing formed one of the earliest applications of big data—(even before the popularization of the term).

Once the binders were implemented, the next challenge was to improve the speed of record updates and allow access to patient records 24/7 from anywhere in the medical network grid (around 37 large academic hospitals organized in 12 groups and tens of clinics and nursing homes). The introduction of smart cards allowed the goals to become a reality. With this improvement, the risks associated with late diagnosis, surveillance and patient record management efforts were reduced.

X-Act OBC Platform predictive emulation and risk management were used to evaluate the cost efficiency of a network of public hospitals serving a large metropolitan area and its suburbs. The hospital system offered healthcare to more than 7 million individuals with 5 million external consultations, 1.2 million beds, 1.1 million urgent care visits (1 every 30 seconds), 38,000 new births, and 1,200 organ transplants each year. Ninety thousand professionals, including 22,000 doctors, 51,000 hospital personnel, 16,000 nurses and 15,000 administrative personnel, served the needs of the constituents.  Additionally, an average of 2,700 research projects in biomedicine with strong connection to the academic world were included under the same management structure.

The risk in this environment is predominantly operational. However as the system involves human safety, management of possible pandemic, and professional errors, legal, economic, reputation and administrative risks are present which require strong predictive analytics to control and alert stakeholders of any performance problems. Using our X-Act OBC Platform technologies and optimal business control (OBC) methodologies, we were able to help the government reduce the cost of healthcare by 9% and have plans for an additional reduction of 10% through the smart use of a universal database.

Through this project, we were able to construct a predictive platform that allowed the management to test decision scenarios and explore options to implement right-time control and surveillance. The technologies allowed stakeholders to anticipate risk and enhance mitigation plans as the system dynamics evolve or change. Having proven the solution through this project, it is our ambition to generalize the approach to cover the whole country. The expanded usage would allow a host of studies and research projects to take place in order to understand the origin, evolution, risk factors and correlations to internal and external influences of both rare and more recognizable maladies.

Dynamic Complexity in Healthcare

/in , /by

A healthcare system can be defined as the organization of people, institutions, and resources that deliver healthcare services to meet the health needs of target populations. Worldwide we have a diverse variety of complex, arduous and multifaceted healthcare systems. Nations design and develop healthcare systems in accordance with their needs and resources, but their choices impact social and political dimensions as well as every governmental department, corporation, and individual, which they are built to serve. Currently many governments are struggling to contain the cost of reliable and equitable healthcare systems. The efficiency of the system is necessary to support the wellness of citizens as well as the economic and social progress of the country. Therefore we can consider healthcare as both a cost to taxpayers as well as an investment in the future.

If we consider the risk dimension of healthcare, we can anticipate a spectrum of risk factors, each of which can become preponderant to the others at any point a time. Operational risk, economic risk, pandemic management, and right-time interventions are just a few of the critical risk considerations. But we must also consider public safety, medication shortage, lack of healthcare professionals, as well as inefficient management of health environments and associated research.

Over the last decades, several government sponsored healthcare mega-projects have been undertaken to add more automation to healthcare management systems. The scope of these projects has varied based on the country’s willingness to invest in the effort, but in each case the main objectives have been the containment of healthcare costs and improvements in the quality of healthcare services. So far, the results have been mixed. Any measurable program success is often tempered with considerable financial burdens and less than expected efficiency gains. From the management of patients, care infrastructure, medical records, and medical research to preventative and palliative care, the spectrum of contributing risk factors is wide and hampered by both static complexity (number of items and attributes) and dynamic complexity (dependencies, time series, case evolution, historical changes).

There is no doubt that the impact of dynamic complexity causes a great number of healthcare transformation project failures. Project outcomes are typically marred by costs that are several times higher than originally planned and significant project delays, which then further inflate the overall costs of the change program. In general, these problems are created when dynamic complexity is ignored during the business analysis phase that precedes information technology system transformation plans. The inability to express dynamics using natural language, difficulties in gaining an end-to-end picture of system dynamics, variations in healthcare procedures and practices, and finally the lack of clarity in required care, prevention and speed of treatments versus the expected results, are major roadblocks in automating healthcare systems.