OneHealth - An Initiative to address healthcare inequalities

Black people in particular continue to have the highest infection, hospitalization, and mortality rates while still lagging behind every other ethnic/racial group in vaccination progress

Two years into the COVID pandemic, the health inequities that have medically disenfranchised millions of black and brown patients for decades have seen no remedy despite having been thrust into the spotlight following glaring errors in the management of the emergency response which means that data collection is not accurately tracking nor collecting patient data on long COVID in an unbiased way. This will lead to missing and incomplete data, outcomes, and treatments."

Our goal is to fill the gaps in public health knowledge by crowdsourcing data from citizens to improve healthcare queries and responses. In a post-pandemic and recovery phase, data is one of the most important tools for policymakers, public health officials, and health system agents. Citizens need to have easy access to exposure alerts, symptoms, testing, treatment, and vaccination, but much of this journey is not shared with public health and remains invisible to the health system. These gaps in data gathering, especially regarding activities, contacts, and status of citizens are one of the key challenges for an effective response. The data-scarce planning leads to inefficiencies, lives lost, and socio-economic costs. Technological app-based solutions can be used to fill in gaps, but they currently suffer from two key problems that make citizens uncooperative and disengaged:

1. A lack of incentives for users to adopt and engage with the app

2. Citizen’s privacy concerns

How can we achieve crowdsourcing, privacy, and personalized engagement in a  pandemic while keeping healthcare equity as priority? And how can we simultaneously provide planning tools for public health by making citizen data available in real-time?
Unavailability of information related to Pandemic spread and communicable diseases across the globe due to data silos and non-participatory crowdsourced epidemiology model, helping to identify the next pandemic
According to research published with data evidence from MEDLINE, EMBASE, and ClinicalTrials.gov, for the COVID19 Pandemic, approximately 30% of hospitalization in the United States could be avoided given people were aware of the transmission and geographical areas.
Given the silos of information about infection spread, people are unable to track and trace the spread of the infection, giving a short window to be cautious before the health is affected. Lack of evidence and early warning systems have created a gap in exposure alert systems, leading to the transmission of communicable diseases. This further affects the BIPOC community at large.

A novel open-source computational privacy software from MIT and PathCheck that captures crowdsourced health information, analyzes it for public and precision health and engages users via personalized recommendations. We provide computational techniques to handle the five steps of the pandemic progress - exposure, symptoms onset, test, treatment, and vaccination. 

We are proposing a platform for underserved communities to help them gather information about Post Pandemic Conditions; PCC and other chronic diseases), available treatment, from doctors around the US. The app will be designed for crowdsourced data collection while preserving privacy. We are using No Peek Privacy for maintaining privacy standards. 

We will have three onboarding personas. 

1. Citizens - underserved communities

2. Doctors/ Health Practitioners

3. County / State Health Officials

4. Communities officers - Various communities across the US

These personas would qualify the circle of Capture, Analyze, and Act structure of the product journey. The solution will work with communities mainly to achieve health equity among the people. 

From reporting of symptoms that can help predict the next communicable disease, to grievances from anonymous citizens, we at PathCheck and Hoods Medicine would be working towards an equitable yet crowdsourced health reporting system. Our toolkit provides services and computational methods at each step

1. Exposure - We provide a privacy-preserved solution for exposure notification and contact tracing.

2. Symptom onset - The users can upload their symptoms and keep track of how the disease is progressing. Furthermore, they have access to download the symptom trajectory of other users in the same demographic and comorbidities group.

3. Test - We provide a solution to find the nearest testing sites, schedule testings, and also create verifiable credentials to share testing results. Further, if the user agrees, they can share the data (in a privacy-preserving way) with the health officials who can learn more about the population-level trends of the disease.

4. Treatment - The users can upload their ongoing treatment and what their recovery trajectory looks like. In addition, similar to the symptoms curve, the users can also access the treatment progression curves of the relevant group (based on demographics, comorbidities, etc).

5. Vaccination - We have the solution for vaccine scheduling, second dose reminder, adverse event reporting, etc along with verifiable and tamper-proof credentials. Moreover, we offer this with the help of low-tech solutions - paper credentials as well.

Our software also helps in the availability of information related to Pandemic spread and communicable diseases across the globe through a participatory crowdsourced epidemiology model, through an app interface for people to receive exposure alerts, keeping privacy preserved. Our Evidence-based early warning system helps administer, over-the-counter treatment plans, and the availability of healthcare facilities resources such as hospital beds, oxygen supply, nearest vaccination booths.

Two years into the COVID pandemic, the health inequities that have medically disenfranchised millions of black and brown patients for decades have seen no remedy despite having been thrust into the spotlight following glaring errors in the management of the emergency response.

We deploy citizen-facing apps focused on BIPOC communities to achieve a NoPeek data-rich view of individuals' interface with report, trace, test, isolate, and vaccinate. 

To avoid perpetuating this tragic compounding of inequity, we need to ensure we hear black people and believe symptoms in a world where the face of our broken systems is usually a care provider at the point of care. We need to explore innovative ways to extract bias from the processes not only for the benefit of perennially underserved communities who are most vulnerable to the considerable institutional and public health failures, but also so we can accurately track and assess the true picture of post-COVID and long COVID sequelae, incidence, and other chronic diseases symptoms, and ensure whatever appropriate standards of care are equitably applied to all patient populations

Our solution will help minor communities to express their grievances and queries regarding healthcare without racial discrimination. Without our privacy-preserving approach, every individual is benefited from one care, one health, one platform. That is our motto - OneHealth - an equitable initiative for all communities inclusive, to end the inequalities and bring more equitable healthcare visibility. 

We have 3 audiences:

1. Public health agency (PHA) planners: we create a data-rich stream of citizen activity to fill the gaps and complement other data-gathering efforts to understand case rates, the severity of symptoms, equitable deployment, lower cost of impact evaluation and interviews, and effectiveness for vulnerable populations. Our insights and reports will help these planners to work with BIPOC communities and also representatives from social communities that focus on minorities. This focused attention will help alleviate the inequalities and bring forth more visibility in the chain of equitable Healthcare delivery. 

2. Organizations with large campuses: here the decision-makers usually take the lead in adopting innovation and expensive solutions to ensure safety. We hear a significant frustration among these leaders through our user research as they have to wait for city or state solutions. Hence we improve resilience and confirm effectiveness with A/B trials on different campuses of the same organization. We are focused to highlight any need for equipment or services if needed by any hospitals that are serving primarily theBIPOC communities. The Organizations will also have a direct center of excellence to look into queries and grievances that are concerning for the BIPOC community.

3. Citizens (with Focus on BIPOC Community). We create early alerts of exposure, personalized risk scores for activities based on their health conditions, and visualizing local conditions of the spread is critical. We focus on communities-driven healthcare queries, use the crowdsourced data to provide and allocate necessary resources. These communities will be driven by members of BIPOC and minority communities, which will bring forth an equitable platform for all healthcare needs. 

In terms of Privacy, our solution works for citizens without a smartphone. We developed a paper-based QR code with cryptographically tamper-evident digital signatures for citizens to engage in the test-trace-treat-vaccinate ecosystem without disclosing any personally identifiable information. 

We started with the USA, India, and Cyprus. We are planning to expand to Nigeria and Brazil as well to capture views from a wide variety of audiences and refine our hypothesis at various consumer touchpoints. 

In this regard, our user research surveys helped gauge a few points: 

1. Penetration of mobile phones ( not always smart phones) even in rural areas. 

2. Skewed awareness regarding private information. 

3. People's ability to scan QR scan codes due to Google pay etc. 

4. People's lack of awareness about the data gaps in public health regimes.   

5. Consumers demand technical solutions to be non-obtrusive.  

This helped us understand the importance of developing a robust data capture system. But to create awareness and reach a wide spectrum of people, it is crucial to enable privacy-preserved data capture systems. Also, to increase downstream adoptions, our solution will be enabled through what users already have, like smartphones/normal phones/card-based systems, which ensures its ubiquitousness and human-centered design.

This solution is led by Dr. Shanice Hudson who completed a BS in Biology at MIT, and an interdisciplinary PhD in Bioengineering, Biostatistics & Bioinformatics, and Pharmacology & Toxicology at the University of Louisville. She currently works as a research scientist at Elanco Animal Health. She spearheads the Hood Med Chats web series, directs scientific content development, & manages community partnership efforts. 

Dr. Shanice is leading Hood Medicine Initiative is a nonprofit public health collective made up of scientists, hackers, and other assorted geeks who are dedicated to improving the health of black and brown people, and the healthcare ecosystem for the BIPOC communities. With Pathcheck collaborations on technology and policy initiatives, there is a strong possibility to downsize the problem of healthcare inequalities in various communities. 

Our Technology and Leadership team at PathCheck and advisors have deep expertise in public health, medicine, epidemiology/infectious disease, software development, bioinformatics, human-centered design experts, ethics and behavior scientists, policy experts, and geneticists. 

IP and innovations (see work above on NoPeek)

We have a formal contract with MIT for shared resources and technology transfer.

Our volunteer base is international and they strongly guide the needs, design, messaging, and prioritization of the solutions in those regions.

Prof. Raskar sits on the steering committee of the NSF PREPARE program which is focused on pandemic planning and resilience. Prof. Raskar testified as an expert witness for the US Congressional hearing on Contact Tracing and Exposure Notification. The project received seed grants from NSF and Schmidt Futures Foundation. We participate in all standardization efforts.

With the Solve platform, we can achieve an ecosystem that is diversified in providing resources that can achieve healthcare equity. 

With Solve partnership, our initiative strengthens and we can focus on expanding our avenues through potential partnerships that the platform can provide. Our impact and strategy have been around the focus plan of achieving the following :

1. Social business aspects: Impact Opportunity, Customer Discovery, Theory of Change, and Strategy for Scale. 

2. Achieve an order of magnitude to determine the success of an equitable healthcare platform that focus on BIPOC community

3. Bring more partnership and reach out to a wider audience focusing on similar issues through the SOlver community and identify and address the underlying inequality healthcare service problem.

4. Working with Solve members helps government leaders and regional health IT companies gain confidence in our NoPeek privacy and open-source crowdsourcing software.

5. Solve members can help us create education and outreach programs for government leaders as we explain the legal aspects of privacy, safety, and efficiency. We hope that many Trinity member educational institutes will run workshops about the benefits and NoPeek and crowdsourcing.

6. Getting expert technical support from Google/Facebook/CubeIQ/Palantir and other tech members will ensure even higher quality software.

In a world where information moves as fast as the click of a button, public health interventions lack the understanding of data gaps in their systems and simple test, predict and alert, without analyzing outcomes. Traditional pipelines emphasize the top-down approach that is often intrusive and involves biased and expensive solutions. Governments are unable to adopt other techniques due to the inherent gaps present in public health data, leading to failed response plans. At present, a major segment of the user journey from getting infected to the recovery path is hidden or disjointed from public health. To address this and bridge such gaps in public health data, we adopt a bottom-up approach that enables voluntary crowdsourcing through tools powered by no peek privacy methods to capture useful data from the citizens and empower people with personalized guidance on their phones. Such techniques would prevent data silos at a few major key players and guarantee citizens the control of their data having such high-quality data in a protected way is the key to “pandemic preparedness” and inhibition of mis/disinformation.

Hence, We create open-source crowdsourcing and engagement software that governments and campuses can deploy for their citizens. Thereby, empowering them to reduce the gap in public health knowledge.

We use the “NoPeek” privacy approach, partially developed at MIT. The NoPeek approach is one step ahead of consent-based or anonymized approaches because it removes the need to share any raw data with servers.

Our technology stack uses the “Capture, Analyze, and Engage” loop.

Capture: (i) background (age, health history, ...) (ii) status (symptoms/tested/vaccinated, non-user data: mobility data, previous disease data, vaccination trends, social media data) and (iii) Activity: (sensor trail e.g. GPS/BlueTooth trail, ..)  All using NoPeek protocols to preserve privacy.

Analyze: We use federated learning modules without looking at raw, individual citizen data and help form foresight based on our predictive models. Further, in hindsight, we understand the disease dynamics as well as provide insights into the current snapshot of the pandemic.  

Engage: This module delivers personalized Information  (risk score, cases nearby, precaution, test or vaccine credentials, creates nudges, supports gamification, incentivizes safe actions, and makes recommendations specific to each citizen. With NoPeek, the server never discovers what message was shown to any citizen.

 Our effort is innovative and unique for three reasons. 

1. Tech: Unlike other organizations that use consent and anonymity as a poor substitute for privacy, we use the “NoPeek” approach developed at MIT. The set of NoPeek algorithms is computationally hard to breach and has rigorous mathematical proofs that attest to these cryptographic guarantees. We believe consent and anonymity-based approaches are prone to data breaches. Governments are unwilling to let private for-profit or non-profit players deploy a solution for their citizens with such risks.

2. Trust: Unlike commercial solutions with vendor lock, sensitive health data and guidance require transparent, unbiased, open-source solutions 

3. Open Innovation: Similar to the Mozilla foundation, our open-source software is an open and transparent innovation platform.

Our 5-year goal leverages the progress we made last year as a new non-profit that rapidly pulled together funding, staff, and thousands of volunteers and developed and deployed open-source software.

Year 0 (last year): We started a new nonprofit for contact tracing and vaccination solutions in the last year

• Raising funding, hiring staff, and recruiting thousands of volunteers

• Developed open-source software, build a network of regional health-IT companies

• Deploy production software with 6 governments

• Create a thought leadership program with research, workshops, and papers

• Have reached out to a target audience of approximately 1 Million with the BIPOC community as the primary audience. 

Year 1: Build the NoPeek privacy app, server, and dashboard 

• Build software stack

• Secure additional funding

• Expand team

• Host workshop with stakeholders and early pilot partners

• Initiate standards conversation for pandemic data exchange

• Create a platform for the BIPOC community and focus on inequality issues, providing

• Have reached out to a target audience of approximately 1 Million with the BIPOC community as the primary audience. 

Year 2: Focused on rolling out initial pilots 

• Focus on our 4 confirmed campus partners in four countries

• Confirming the effectiveness via randomized control trials. 

• We will run these A/B on different campuses of the same organization.

• We will hone our approach (paper credentials, etc.) with vulnerable and underserved communities. Our project target audience will be 2 Million in the second year.

Year 3:  Goals center on solidifying and expanding these pilots 

• Ensuring that this set of software, partners, and practices can be scaled and used in dozens of other countries and institutions.

• Written case studies and testimonials

• Opportunistically, deploy in a real outbreak if it emerges

• We will focus on creating a larger audience at a scale of 6 Million from the underserved community. 

• We want to propose a plan of Equitable Community Champions for each of the communities who will create a larger network of communities

Year 4 and Year 5: Build an equitable BIPOC center of the Health Care ecosystem  

• We need to create a center of excellence that focuses on Equitable healthcare for BIPOC communities. 

• With this center of excellence, appx. 20 M members of underserved communities can have their health inequalities addressed. This will be a point of address for future Policies on equitable healthcare across the US and territories where the solution gets expanded to. 

• Work with healthcare leaders to a more sustainable policy focusing on healthcare equity. 

We are focused on improving the equitable healthcare ecosystems for all communities across the globe. Our focus is Health for All, Privacy for All.

We deliver an open-source software toolkit for governments and campuses that is deployed by regional health-IT companies. Here are our goals and success metrics.

• Goal 1: Secure pilots commitments

Metric: 4 government entities agreeing to pilot (partially achieved)

• Goal 2: validate technology (NoPeek and other technology)  

Metric: recommendations and privacy validated by third parties in pilot

• Goal 3: deploy a successful pilot

Metric: success criteria pre-established for each pilot and met after 4 months.  Pilots rolled out to at least 50% of the citizens at that campus with 20% monthly active users

• Goal 4: new clients

Metric: We witnessed 23 Million users on our crowdsourced App, Karuna. Our exposure notification platform has been used by millions of US citizens and has been deployed in the states of Alabama, Louisiana, Guam, Cyprus, Hawaii, and Minnesota.

Our internal metrics are:

• Thought leadership (Monthly mentions in news media, subscribers to our social media channels)

• Creating a worldwide network of health-IT companies: (numbers of companies using our software versus the number of companies attending our webinars)

• Contribute to Public health research publications

What do citizens report? What is getting captured? 

The App will capture - PCC, and other chronic diseases symptoms and provide information, aid, and relevant information keeping privacy preserved. 

What is getting analyzed? 

The Analyze section will focus on data gathered from the crowdsourced model and give necessary analysis on the PCC, a self-reporting tool, a self maintainable tool to help anyone carry on their in-house follow-ups for PCC.

 What actions need to be taken? 

The Act component will be the output to each Analyze and Report section of the platform. The Personas ( health practitioners/state health officials) will be working closely with PathCheck and Hood Medicine foundation to create a healthcare service layer on request by any citizen with a focus on reducing bias. Any doctors once boarded on the app, would be accessing a number of stage 1 health queries ( stage 1 PCC- which can be done through teleconsultation ). The identities of these queries will be withheld since the consultation should be focused on reducing bias. A self symptom checker that can be adopted by any citizen. If a doctor would prefer to help any citizen, they( the citizen requesting, and the doctor) to switch to the PathCheck teleconsultation platform, again with identities withheld. Only after both parties enter the consultation phase, the names will be shown as per the discretion of the patient on the tele platform. 

The state health officials will have access to surveys from citizens to understand how their county/ states are performing in achieving a health equitable system, by not only looking at stats from the Capture component but also from the grievance section introduced in the Capture Section. 

In the Future course - We are also looking at an integration that will also capture any grievance for a certain area under a county where any racial discrimination-related information, bias in medical assistance, would be screened programmatically through the PathCheck NLP suite of health equity model.

The core technology enabling our bottom-up crowdsourcing methodology is No-Peek privacy-preserving techniques developed jointly with MIT.   We leverage the usage of split learning and differential privacy to capture the data from people. To get insights from the data, we Developed machine learning and deep learning models that can predict prevalence, and forecast outbreaks based on self-reported population-level aggregated symptoms data, and NPI prescription models. To provide the best interventions, we developed multi-agent reinforcement learning models for prescribing non-pharmaceutical interventions, given the current state of the disease and economic impact. Traditionally simulating such large population data takes tremendous time, hence we develop a novel DeepABM system that makes use of Graph Neural Networks (GNNs) to scale agent-based models to run simulations with 100,000s agents in less than a few seconds

Out solution incorporates several digital solutions:

• Web interface / Mobile Interface feasible for Low bandwidth internet connectivity zones.

• Collection of data through privacy-preserving methodologies 

• Paper Cards

• The developed Deep ABM leverages the use of GNN which enables simulating the whole population data in a few seconds. 

• Developing novel machine learning models that can forecast COVID cases and interventions which can assist greatly in low resource settings

The leaderships of Hood Medicine come from diverse backgrounds in terms of ethnicity, education, gender, etc. The organization has been focused on advocating for gender and racial equality and promoting initiatives that help reduce the racial and gender gap in the health ecosystem exacerbated by the pandemicHood Medicine Initiative (HMI) team is focused on using science for good to tirelessly enhance equitable healthcare access and outcomes. They run a small courier service and also work to support the anti-racism organization Overcoming Racism. The recent focus on public health programming and policy for implementation in minority communities and pandemic relief to struggling communities adds to the inclusivity. Members from Hood medicine and PathCheck focus on engaging underserved communities to advocate for science and health education, and preventative medicine. We are also committed to empowering families and supporting mental health and well-being, especially adolescents and adults who have experienced trauma, anxiety, and depression. In this direction, there is a targeted mental health program with a  focus on children, young adults, and African American single parent engagement. This enhances diversity, inclusivity and promotes wellness and self-care in BIPOC communities.

1. Based on our previous work and goals, we build on existing relationships with governments and partners established in 2021. 

• We have demonstrated extremely promising results in our first year.  Our team has experience in working with governments and campuses to help them deploy our solutions. 

• As a B2B solution, we play an active role in promoting adoption through our playbook and shared best practices across pilots and earlier deployments.

2. Years 1-3 we will scale with local partners and with network effects.  Having credible validation evaluations of our pilot results is essential for industry confidence.

• Scaling with local partners: Our pilots are ‘light house’ deployments on four continents, and those references will guide campus-specific implementation in other parts of the world. We work with local health-IT companies to deploy, which allows us to dramatically scale while remaining lean at PathCheck. 

• Network Effect: We create a network effect by encouraging employers to suggest the app to their employees and staff on their campuses. These early adopters will in turn spread the word of the app’s usefulness to other citizens in their daily lives.

With the newly announced budget by the Biden administration to support Pandemic planning, we look forward to collaborating with multiple counties/states and even with the federal government to make customizations to the open-source solutions that we built. Additionally, we also plan to work with institutions and other organizations with large campuses to deploy our solutions. In the past, we had raised about 20 Million users at peak pandemic response and around a million downloads for the exposure notification app. The majority of our clients are state governments and local authorities of the US apart from citizens across the world. We launched the Karuna app, crowdsourced epidemiology, and Pandemic response app in India and Indonesia.

In 2020-21, we raised 2.6 million USD in donations and grants Additionally, PathCheck generated 1.2 million USD in revenue for a total of 3.8 million USD. This is based on our work with various state and county level governments for the EN solution we provided. Recently, we partnered with WHO to work on universal health credentials.

Apart from these, our data science team had won the Facebook Data for Good Pandemic challenge and other competitions from NIST, XPRIZE, etc.