Ethical, inclusive, and open digital ID

Over one billion people globally lack any type of formal identification, which prevents them from accessing health services and impacts the quality of primary health care performance. In vaccine delivery, this translates to inaccurate records and gaps in vaccine coverage data, while in maternal health and infectious diseases, it means lack of continuity of care, leading to suboptimal health outcomes. For instance, across the developing world, there are often huge gaps between administrative and WHO survey vaccine coverage data: the average gap in Bangladesh is over 17%, and in Nigeria the discrepancy in numbers rises to almost 34%. In infectious diseases like HIV, patient retention rates are gravely inaccurate, with some rates underreported by more than 50%. The lack of accurate data in these circumstances contributes to issues with measuring health care performance, leading to less-than-ideal allocation of resources.

While digital solutions exist to address some of the issues with data, they are often not catered to last mile contexts. For example, there is an abundance of vendors offering digital ID solutions globally, however, most of them are only catered to developed countries, and often prove to be too expensive to be deployed in frontline contexts. Further, they often lack interoperability, which makes it difficult to leverage and integrate them with existing systems. Finally, these solutions are often closed-sourced, leading to questions about their  long-term sustainability.

Our solution is an end-to-end, open, and ethical digital ID product, which utilizes biometrics to identify people. While alternative digital ID solutions exist, Simprintis designed explicitly for low-resource and ‘last mile’ settings. Our technology runs offline, with powerfully compact algorithms, low-processing footprints, and UI designed by community health workers.

Simprints was designed from the ground up with frontline healthcare workers, for frontline healthcare workers: it seamlessly plugs into existing data collection platforms, becoming a part of existing workflows, making the process effortless for end users. A few key features that make our solution innovative and unique are:

• Ethical and inclusive for last mile populations. Simprints is built off diverse and representative datasets. For example, our studies show that Simprints’ fingerprint biometric device, Vero, is 228% more accurate with scarred, worn, and damaged fingerprints common in low resource settings.

• Designed for public health systems. Simprints focuses on folding into the existing digital infrastructure of partners. Furthermore, our technology adheres to open standards (ISO 19794-2), making it interoperable with closed-source and legacy biometric ID systems. In the next few years, we will be working to make Simprints open source, thus increasing our long-term sustainability.

• Innovative and robust privacy practices. Simprints has led innovation in privacy-by-design for low resource settings (e.g. Mission Billion which the World Bank ID4D group requested in order to adopt in their ID programmes), published research in leading ethics journals, and advocated widely across the sector.

Our solution serves impact partners (governments, NGOs, international donors), frontline healthcare workers, as well as beneficiaries. Our impact on each of these is described below:

1. Impact partners: Program implementers rely on accurate data to model needs and allocate resources efficiently. Simprints serves impact partners by ensuring that the data used by them is accurate: each vaccine, each public good, each healthcare intervention is tied to an individual beneficiary, ensuring that there are no duplicates and the delivery of these is verified. Given that our solution is interoperable, in the medium- to long-run, this enables our partners to streamline data collection from across different platforms, and use these to optimize primary health care performance.

2. Frontline healthcare workers: Simprints’ impact on health workers is two-fold: one, it replaces paper-based health cards and registers which can be difficult to maintain, making it easier for health workers to keep accurate data records. And two, Simprints can reduce the amount of time it takes for healthworkers to pull up the right records: in Ethiopia, the deployment of Simprints has improved the efficiency of health extension workers enabling them to pull up records accurately 10x times faster and by reducing data discrepancies from 39% to less than 5%.

3. Beneficiaries: For beneficiaries, lost health cards often mean missing out on life-saving doses of essential medicines and vaccines. By tying beneficiary health records to biometrics, Simprints can help capture those who fall through the cracks between household, health post, and centre. Simprints leverages biometrics to create an integrated digitized healthcare system that links everyone to a single beneficiary record across the entire primary health care system, enabling continuity of care.

Simprints’ CEO, Toby Norman, was working as a researcher in Bangladesh when he experienced the need for better ways to identify beneficiaries. Post this, our co-founders spent a few months in Bangladesh and Benin, testing the accuracy and suitability of existing biometric solutions in front-line contexts. They uncovered that none of the  five most popular biometric solutions worked well with scarred and burnt fingerprints that were quite common in communities where livelihoods involved manual labour. Thus, Simprints was born. 

Our team is extremely diverse, with employees from over 15 countries. Our work in LMICs is driven by a team with grassroots experience: Our project delivery is led by Stephen Taylor, who has lived and worked in Syria and Pakistan for over a decade. In South East Asia. and Ethiopia, our work is driven by Nile Rahman, Country Director, Bangladesh, and Alfenur Kufa, Country Director, Ethiopia respectively. Both Nile and Alfenur bring a strong understanding of regional communities and stakeholders, and their expertise helps us adapt our projects and offerings to match local needs.

We have spent over 2000 hours with frontline workers co-designing a biometric system that allows them to quickly access the right record at the right time. Before a project is implemented, Simprints works in collaboration with program teams and local communities to ensure the program design best suits the needs of users. We also partner up with local leaders and volunteers to create more awareness about the intervention being delivered. Finally, we maintain feedback channels with project partners on-the-ground to better address their needs and iterate our product accordingly.

Our commitment to strengthening health systems through accurate and actionable data directly aligns with the goals of the Challenge. Often, interventions are short-sighted, and only focus on fixing the issue at hand, rather than strengthening the health system as a whole. This leads to technology being introduced unsustainably, and a growing backlog of dysfunctional digital systems. Simprints wants to contribute to sustainable changes in how data is managed in frontline projects, and we believe that MIT Solve and Gates can prove to be the ideal partners in realizing this vision.

Further, MIT is a hub and a frontrunner in all things tech. As we look forward to open-source our solution, Solve can help us access a network of renowned experts at MIT to refine our product and expand our reach.

Finally, an ‘MIT Solve’ accreditation adds a lot of credibility to a product like ours, helping us build partnerships and expand our reach, having an outsized impact on our finances way beyond the prize funding itself. 

Four features make Simprints innovative. First, Simprints is designed explicitly for low-resource and ‘last mile’ settings. The technology runs offline, with powerfully compact algorithms, low-processing footprints, and UI designed by community health workers. We work alongside our partners on-ground to support them with trainings, deployments, and round-the-clock troubleshooting, ensuring our solution is sustainable in the long-run. Simprints has been deployed in over 16 countries, from mountainous and remote regions in Nepal to crowded slums in Kenya—some of the most challenging environments.

Second, with concerning reports about racial bias in for-profit biometric technologies, Simprints instead harnesses ethical and inclusive AI approaches to ensure the system works for every patient. For example, scarred, worn fingerprints are common in developing countries with high manual labor populations, yet rare in the West where most biometric technologies are developed. To combat this, Simprints built a custom dataset of 135,000 fingerprints from representative populations (currently the world’s largest academic dataset) to develop a system that’s 228% more accurate with damaged fingerprints.

Third, Simprints does not create “yet another” digital healthcare application. Instead, the technology is designed to plug into widely adopted and proven digital health and identity platforms that have scaled to millions of patients in developing countries, such as DHIS2, CommCare, and MOSIP. For example, with funding from Gavi, Simprints has been integrated into DHIS2 digital vaccine registries to link children to routine immunisation records in Bangladesh. Our public developer documentation includes examples of multiple integrations. By integrating into existing tools and harmonizing disparate patient dataset through biometrics, Simprints reduces data fragmentation that threatens effective pandemic response. 

Fourth, Simprints is designed from the ground up to be a privacy-first technology. Working with human rights lawyers, we developed “layered” privacy notices that are simple, clear, and understandable even with limited education. Our technology is compliant with GDPR and employs best-in-class data siloing, security, and anonymization. Furthermore, as a team we advocate constantly for our whole sector to step up when it comes to patient privacy, including in Privacy, Law, and Ethics Journal, Reuters, and CNBC Africa. In 2019 our privacy solutions won the World Bank’s Mission Billion Challenge for privacy-by-design approaches. Putting privacy at the heart of technology will be essential to securing trust, safety, and the ethical use of data in fighting pandemics. 

Our impact goals for the next few years are below:

1. Become a Digital Public Good: reach 10x beneficiaries. We are open-sourcing our solution over the next couple of years, which will enable us to scale our impact exponentially. 

2. Vaccine Delivery: support INGO and Ministries’ goal to verify vaccine coverage [GAVI: 60% of the global birth cohort live in Gavi-supported countries. This means that as a global alliance, we help protect nearly half the world’s children through vaccination]. We are already partnering with Gavi to enable verified vaccine delivery in Ghana, and are in conversations with existing partners to expand our footprint.

3. Strengthen Health Systems: work with 10 Ministries of Health to integrate biometrics into their health systems and accurately impact the delivery of services. Increase the efficiency of community healthcare workers. We are already working with 4 Ministries of Health, and in exploratory conversations with a handful more.

We measure success against several progress indicators, broadly broken down into 1) technical KPIs and 2) project impact KPIs.

First, in order to help patients, Simprints must be technically accurate, fast, secure, and lightweight (i.e. have a low size and processing footprint for low-end mobile devices common at the last mile). As of our latest 2020.4.0 release, Simprints has >98% average accuracy rate, can search >3,000 records per second, with end-to-end encryption, and an average 261 byte per template size.

Our understanding of project impact is made up of two components:  how many people are reached by that intervention (breadth), and the effectiveness of an intervention (depth).

The first of these is key for Simprints' mission: we help organisations to verify how many unique beneficiaries their programs are reaching. Some core metrics for any project are therefore:

• the number of biometric enrolments

• the number of times biometrics are used to identify or verify a beneficiary

We track these KPIs at a project level, and also an organisational level: as of the end of July 2022, we have reached over 2.2M cumulative beneficiaries.

In terms of the depth of an intervention, the precise metrics differ depending on the project, and these metrics are typically tracked by our partners rather than directly by Simprints. Some examples are:

• In Bangladesh, we worked with BRAC, the largest NGO in the world, to ensure continuity of pre-and ante-natal care for pregnant women and mothers. A study showed the use of biometrics to uniquely identify mothers during the risky perinatal period resulted in 38% more visits, leading to 19% more newborns receiving all essential newborn care.

• In Malawi, we worked with the University of North Carolina, Chapel Hill  and the Malawi Ministry of Health to accurately and reliably monitor HIV patients. Our biometrics system recorded 29% more visits than the electronic medical record (EMR) system alone, ensuring continuity of care.

• In Ethiopia, we are working with the Ministry of Health on an electronic community health information system (eCHIS) project in Ethiopia. The most externally visible impact of the pilot on the health system has been the efficiency gains that biometric identification has offered. Respondents unanimously agreed that biometrics have improved the efficiency of HEW by saving time in locating clients records. The review observed that HEW were able to pull up records 10x times faster through biometric authentication compared to manual name search.

  • Community members and users indicated that biometrics has played a role in enabling higher levels of client satisfaction, due to reduced queueing times.

  • All key informants reported that biometrics have improved data quality through the reduction of false reporting by verifying service recipients. One FMoH official reported that annual Routine Data Quality Audits showed data discrepancies in the woreda reduced from 39% to less than 5% within the last 3 years and attributes this result to the use of a biometrically enabled eCHIS.

Simprints directly addresses challenges with existing biometric tools that are not built for ‘last mile’ projects. For fingerprint recognition, most technologies are not adapted for scarred, worn, damaged fingerprints that are far more common among long-income populations involved in agriculture or manual labor. Similarly for face recognition, racial and gender bias leads to a less accurate identification rate in sub-Saharan and South Asian populations. To counter this Simprints has developed a biometric digital ID system that’s inclusive for diverse populations, for example developing fingerprint recognition that’s 228% more accurate with scarred, damaged fingerprints. The system works fully-offline on low-end mobile devices, with a user interface designed by frontline health workers. 

Simprints has developed biometric digital ID specifically for the ‘last mile’

Underpinning the user interface is a full biometric digital ID platform that gives partners everything they need to successfully deploy biometrics in their programme. The “front end” mobile application runs on Android smartphones, which have become the primary tool for data collection by frontline workers. It functions as a plugin or “companion” app for widely used data collection platforms via a client API, allowing them to enroll and later biometrically match beneficiaries. The Android app works offline, then syncs to a backend when data is available. This backend can be either Simprints cloud-backend, or a locally hosted server using the primary data collection platform’s syncing infrastructure (Co-Sync). The backend provides secure storage, data analytics, and more biometric algorithms that enables biometric deduplication across large datasets.  

Simprints is a full biometric digital ID platform that gives partners everything they need

Our product is designed to scale to an unlimited number of users as we are using an elastic cloud infrastructure for our back-end and Android for front-end (downloadable and updatable through Google Playstore). Our Android app is entirely written in Kotlin and adopts the most recent architecture and tools (e.g. co-routines, Koin, etc.). We use Google Cloud Platform (GCP) to host encrypted biometric data, capture and analyze event data and run the deduplication engine.

Our data is collected by healthcare workers operating on the frontlines. Depending on the program, we build in various non-monetary incentives to encourage the use of our technology to ensure good-quality data. For instance, in a project in Ethiopia, we incentivised health workers by offering t-shirts that helped with community sensitisation, while in other programs we provide health workers with unrestricted airtime. 

From a project delivery perspective, using Simprints makes health workers' workflows a lot more streamlined and easier - they can pull up records a lot quicker, and it cuts out the bulk of working with paper-based alternatives: in Ethiopia, the deployment of Simprints has improved the efficiency of health extension workers enabling them to pull up records accurately 10x times faster and by reducing data discrepancies from 39% to less than 5%.

Our biometric technology is built to be inclusive - our fingerprint solution is rugged, wireless, long-lasting, and highly accurate with scarred and worn fingerprints that are typical in frontline contexts. While other top facial recognition systems suffer from crippling racial and gender biases in their algorithms, our face recognition solution is designed from a diverse dataset.

Outside of our technology, we design our programs to promote inclusivity and equality. For example, the use of digital ID, in place of identity cards, enables patients seeking care for HIV and TB to access treatment anonymously without the stigma associated with a health card. In a study published in AIDS Care, Simprints emerged as the preferred strategy to monitor engagement in HIV care, as patients and healthcare workers reported few concerns about patient privacy or HIV status disclosure. 

From a gender perspective, 70% of the global health and social workforce are women, with the majority of these roles being frontline community health workers (CHWs). Simprints is designed to ease the burden on these CHWs by increasing the efficiency of a portion of their work.

Internally, Simprints is a strong champion of DEI in its team. We have a very active DEI taskforce that oversees the improvement of DEI across the organisation. Our team comprises of members from over 15 countries. Our board of directors holds us accountable for these initiatives, with the taskforce presenting an annual report to them.   

Simprints focuses on building tech that is optimised for the development context, with frontline workers in mind. Our platform is rapidly deployable, with minimal set-up required, making it easily scalable for larger projects. In addition to program accountability, we can provide deep insights into program activity through customised M&E reports. 

Additionally, each project is allocated a dedicated project manager, who provides advice on program design, implementation, and workflows. They also conduct customised training for users to be onboarded on our platform, and are available round the clock if any assistance is required.

To sustain our business, we are committed to being an earned revenues-driven organisation. We partner with governments and private philanthropy organisations to introduce verified impact in a range of programs, including MDA for neglected tropical diseases and continuity of care in MNCH services. We also raise innovation funding to improve our product and keep up with the latest developments in technology.  As we look to go open-source, we intend to drive revenues through affiliated services.

From the beginning, we fully committed to the social enterprise model of putting impact over profit, while ensuring that we build a sustainable tech company that can achieve scale through B2B and B2G sales. Simprints became cashflow positive in 2019, four years after our founding, having already won over £11M of funding at a 65% compound annual growth rate by focusing exclusively on lean product development, earned revenues, and financial discipline. Today, Simprints has closed over £20M in funding. In keeping with this drive toward financial independence, we aim to concurrently scale up our business development with increasingly large-scale project contracts. We expect to continue to fund our work through a mix of innovation grants, government projects, and catalytic funding. As we look to go open-source, we intend to drive revenues through affiliated services.

To date, Simprints has raised over £20M in funding from various funders and partners across the globe. These revenues can be split into project revenues and innovation funding. A majority of our project funding comes from projects with Ministries of Health, including Ghana, Ethiopia, and Bangladesh, funded by Gavi, CIFF, and Grand Challenges Canada. We have raised innovation funding from a variety of sources, including InnovateUK, Cisco, ARM, and Elevate Foundation, to name a few.