HEALTH ASAP - Health Analytics Solving Access and Prevention

Predictive analytics to improve access, prevention and affordable healthcare for all mankind.

Knowledge has a key role on healthcare provision, if properly transformed into action. Healthcare data is often large and fragmented. Health analytics can provide predictive models to enhance proactive, preventive and affordable interventions.

Predictive analytics in health can support better decisions, improve
patient care and quality of life, disease management, healthcare
administration, and supply chain efficiencies (eg. avoid readmissions, evaluate transition risk of acute coronary syndrome to myocardial infarction, compare the cost-effectiveness of lung cancer treatments, avoid shortages, etc.).

Nevertheless, there are unsolved issues. For example, not all critical
conditions have reliable prognostic algorithms, most of the clinical predictive
algorithms lack time-related analysis, and predictive analytics is still a
challenge on remote and harsh environments, due to difficulties on data
collection, making humanitarian missions difficult to plan.

HEALTH ASAP aims to solve these challenges and avoid unnecessary
morbidity, mortality, and health system costs, by providing valuable
evidence-based insights and innovative features for health prevention and
affordability.

This project will give frontline healthcare workers and decision makers tools to:

• predict time-related health risks and outcomes of a population or individuals;
• prevent negative health outcomes of demand-lead time on health resources planning (medicines, health products and healthcare workers);
• provide cost-effectiveness and predictive risk analysis;
• generate new models for health predictive algorithms.

HEALTH ASAP is a responsive platform for user’s basic inputs (‘predictors’),
processing large amounts of cross linked data to generate outputs for modelling, and gives accurate outcomes to support decisions.

‘Predictors’ are inputs such as comorbidities (concomitant diseases),
history characteristics (genetic polymorphisms and mutations, previous disease episodes, or risk factors), type and severity of diseases (diagnosis and
preventive characteristics), demographics (age, sex, gender, socio-economic
status), subjective health status and quality of life (psychological, cognitive
and psychological functioning), and physical function status (Karnofsky score,
WHO performance score), among others.

The flux of actions includes:

1. friendly front-end environment for inputs (‘Predictors’) and decision support;
2. data mining center (published evidence from several jurisdictions related with epidemiological data, clinical pathways, care services, costs, transition probabilities, health utilities, patient-related outcome measures, ‘predictors’, etc);
3. back-end algorithms for modelling techniques (eg. Monte Carlo simulations, Markov Models, and ‘smart cohorts’), giving probabilistic simulations and estimations of health outcomes in the short and long-terms.

The modeling techniques for estimation of outputs (back-end) can run fast in available tools based on open source data languages (DAX, M, SQL, MDX and R), which allows input data mashup, modelling, and intuitive visualization with actual analytics tools. Also, on open source languages we provide an environment for crowdsolving and development of new assets.

HEALTH ASAP will contribute for prevention, by providing accurate
predictions on health outcomes and new predictive algorithms; for
sustainability, by providing support to cost-effectiveness decisions; and for

access, by providing recommendations to avoid negative impact of demand-lead time on health resources.

It is time to democratize value-based and predictive care principles in real life. Artificial intelligence behind HEALTH ASAP solution provides increasingly accurate predictions in the long term, giving valuable insights for health prevention and quality of life.

The solution is based on existing methods combined to integrate predictive analytics and value-based healthcare available for all.

We developed ‘smart cohorts’ concept, aiming to collect population data (health records, clinical trials, etc) most related with an individual, from data mining process, to create ‘family-related’ cohorts for accurate estimations across time.

We also provide subgroup predictive analysis for humanitarian mission planning in harsh environments, using techniques to fill censored data, avoiding negative impact of demand-lead time for health resources.

Finally, we promote ‘crowdsolving’, open calls for the community of developers to integrate prediction algorithms with open source languages.

The linkage of data requires artificial intelligence capacities to provide accurate predictions.

Modeling techniques for outputs estimation can run fast in available tools based on data languages like DAX, M, SQL, MDX and R, which allows input data mashup, modelling, and intuitive visualization with actual analytics tools. Also, Monte Carlo Simulations (for random sampling) and other statistical methods can run on R script integrated in SQL.

All these methods, and others of interest, need to be developed by a technological solution compatible with common interfaces.

The platform runs with open source language, which allows calls for solutions among community developers.

1. Strengthen data-mining process (‘predictors’ sources) through data collection and integration (API/SQL integration, for example) for several health conditions and jurisdictions;
2. Identify a suitable online environment to host the HEALTH ASAP platform and transfer the prototype for online environment;
3. Collect evidence-based predictive risk algorithms in strategic fields to kick-off (fertility, breast cancer, myocardial infarction, surgery and infectious diseases) on several jurisdictions, to increase generalizability of health predictions;
4. Raise partners to strengthen financing and technological capacity building;
5. Design a pilot testing (6-12 months), facing challenges related to chronic conditions of short-time transition states and measure the results.

• Transfer the prototype to an online environment and conclude at least one pilot testing for each type of client;
• Implement blockchain technology to store data across peer-to-peer network on country-based clients and servers to increase security and data processing capacity;
• Adapt the platform for device compatibility (app);
• Hire permanent staff, particularly data scientist, programming/statistics, and client support;
• Solve concrete health challenges with HEALTH ASAP in short-medium term (contract agreements), both in healthcare units from developed countries and NGOs missions in poor countries;
• Partnership for integration to complement other tools.

These steps will allow us to achieve scalability and sustainability.

The platform will be available online (responsive, app compatible). Users have free access to several features (outcomes and predictions). Inputs (‘predictors’) and tendency of user’s analysis may strengthen the algorithm (machine-learning).

Health ASAP strategy addresses different needs according to beneficiaries. At this point we prioritize 4 client segments:

  1) Healthcare workers, healthcare providing units and decision-makers;

  2) NGOs (specifically those whom act on harsh geographic areas);

  3) Informal caregivers (parents, families);

  4) Insurance companies.

For each segment we will partner with specific KOL institutions for testing and advocacy to their pairs and followers. 

We plan to run and finish a pilot testing for each type of client in the next 18 months. After each pilot testing is done, we develop a specific targeted marketing strategy to deploy our solution.

The estimation of how many people we expect to be serving will depend on how fast we can deploy the solution.

The team has the following summarized strengths:

1. High knowledge on pharmaceutical sciences, and advanced graduations on Health Technology Assessment (modelling, and economic analysis) and Management (marketing specialization);
2. Work experience on health policy, healthcare administration, health data analytics, health technology assessment, supply chain of health products, and social innovation in the third sector;
3. Skills on health bigdata, open source language for programming and analytics, and project management;
4. Prizes on entrepreneurship and innovation contest, with projects related to health innovation.

The revenue model development will follow the projects maturity. The revenue stream will be the following:

1) For the first stage (0-3 years):

• Free subscriptioners that access several features (mainly for professionals, resource planning, and caregivers, which gives value to the project in terms of diffusion, advocacy and brand awareness);

• Project setup fees (charged to clients with needs of complex platform requirements on algorithms and modelling);

• Subscription fees (charged to users with access to premium features by periodic fees according to requirements (reporting and outcomes) and access (number of users and access levels).

2) For the second stage (3 years and onwards):

• Scaling up the number of customers by segment, maximize and diversify the number of premium subscriptions of individual users (allows a baseline income based on diversified clients, which contributes for long term sustainability – avoids dependence of few clients, such as large institutions, maintaining an unbiased mission and core activity);

• Process users activity-tracking to generate valuable health trends for specific clients.

The project income aims to reinforce and re-invest in the project, adding technological and scientific robustness to Health ASAP as a leading innovative healthcare-tech project to empower our features to all users (mainly features for free subscriptioners).

Mainly to:

1. Improve access to effective and affordable care through a worldwide solution available for all;
2. Acquire an initial investment to finish prototyping and go for the pilot phase;
3. Find partners to help configuring the most effective technology behind our solution;
4. Obtain exposure and reputation to achieve valuable partners who share the same goals and can help us increase the feasibility and scalability of our solution;
5. Save lives together, because everyone deserves a better change.

• Data center strong enough for accurate predictive analytics
• Effective programming to deploy a fast and ergonomic interface
• Competition on health analytic solutions providers

Solve can give us the kickstart by finding the right partners for project development and the initial funds to gather more expertise around these challenges.