How would you prove you’re really who you say you are? Most of us would reach for our driving license or passport – perhaps a birth certificate, national ID or social security number.
Yet around the world, approximately 1.1 billion people in the world have no formal proof of identity, preventing access to essential services such as education, finance and healthcare. For many of these people, their birth was never registered. They simply don’t exist – officially, at least.
In healthcare, this lack of identity can mean the difference between life and death. Those of us in high-income countries are accustomed to having our medical requirements and the care we have received stored safely and centrally. Yet this is a luxury that many cannot access; without a permanent, institutionalized record of illness and medication, the burden of medical history is moved to the patient and whatever notes they may have kept – on paper or in their head. Information vital to effective care can easily be forgotten and the pre-empting of hereditary conditions between family generations becomes all but impossible.
This has an impact on all areas of society, but women and children are typically the most affected: every day, approximately 830 women die from preventable causes related to pregnancy and childbirth. Ninety-nine per cent of all maternal deaths occur in developing countries, in low-resource settings, yet the majority could have been prevented with a consistent program of care.
In September 2015, more than 190 countries signed up to the United Nations’ Sustainable Development Goals – otherwise known as the Global Goals – which aim to create a better world by 2030. Guided by a vision to end poverty, protect the planet and ensure prosperity for all, the goals strive to enable all people to live their lives to their full potential.
The 2030Vision program, of which Arm is a founding partner, is designed to support the delivery of the Global Goals through technology and collaborative innovation. One of the companies embodying this ethos is Simprints, a not-for-profit tech company that’s building low-cost biometric identification systems to enable citizen identification in developing countries.
Biometrics for work-worn hands
Biometric parameters, such as fingerprints, are considered sufficiently unique to allow identification with a high rate of accuracy. However, when the Simprints team tried to select a fingerprint scanner for their solution, they discovered that most available models are designed for people in high-income countries – people whose hands are rarely used for hard, manual labour.
“We found that most biometric systems on the market today have neither sensing hardware nor matching software that can cope with the scarring and wear common in the developing world,” says Dan Storisteanu, co-founder of Simprints and Research Fellow at Darwin College, University of Cambridge.
“Many people have been doing manual labour for their whole lives and display an above-average level of burns, scars and worn fingerprints; in field tests with Zambian manual farmers, we found that 84% had some damage to their fingerprints. We tested six commercial sensors that reported high accuracy levels by collecting over 125,000 fingerprint images from low-resource populations in Zambia, Benin, Nepal and Bangladesh. None of the systems tested reached the accuracy rates published in industry reports.”
This lead Simprints to abandon the idea of using a commercially available scanner and develop their own robust, ergonomic, and mobile fingerprint scanner – powered by an Arm Cortex-M processor – that’s dustproof, water resistant and lightweight, with a long battery life. In short, it’s suitable for the harshest environments on earth.
High accuracy in challenging conditions
Simprints’ fingerprint scanner creates a unique ID for each person, enabling health workers in the field to identify patients accurately and quickly access their record of care. The patient simply presents their finger for scanning; the scanner then securely transmits the biometric data – via Bluetooth – to the worker’s Android smartphone, seamlessly obtaining the patient’s records. The data is stored with time stamps and GPS coordinates to record when and where the patient was assessed. When internet connectivity is poor, the device can be placed in offline mode to access a previously downloaded database, and new patient data is uploaded when connectivity is restored.
Crucially, the Simprints fingerprint scanner is compatible with numerous mobile data gathering platforms used by healthcare workers, governments, and global development practitioners. Developed in conjunction with frontline workers, thorough testing in six countries across three continents has shown that, after attending training workshops tailored to their requirements, workers can learn to use the scanner quickly and easily.
“Ultimately, we’ve shown that fingerprinting can achieve high accuracy in difficult development contexts,” continues Dan, “but the sensor, extractor and matching software need to be tailored to the application.”
While the use case for healthcare is a compelling one, Simprints has identified many more uses for its technology – some of which are already being deployed.
Tracking school attendance
Education is often seen as the key to prosperity, yet according to the UN, in 2015, 57 million children of primary school age were out of school – and over half of them were in sub-Saharan Africa. Working with Impact Network – an organization that uses innovative e-learning technology to deliver quality education to schools in rural Zambia – Simprints tracks the attendance of teachers and students, recognizing lapses in attendance and mitigating against absenteeism. Having accurate data in real time, rather than on a monthly or yearly basis, allows swift action to be taken and ensures that each child has a consistent education program.
Effective distribution of aid
Impact organisations, development agencies, and governments lose billions of dollars to corruption and the misdistribution of aid. According to the World Bank, in 2011 only 41% of grain handed out through Indian food programmes reached the intended targets. The loss of other subsidies is estimated to be at 20%, costing the government $10b a year. By requiring a fingerprint for every claim, corruption could be reduced, and more aid could reach its intended targets.
Putting a finger on accountability
Preventing fraud in cash transfer programmes remains a key industry challenge and can result in lack of payment for the intended beneficiaries. Early impact results from one project show that the deployment of Simprints reduced fraud to <1%, with the potential for the project partner gain to at least $65m/year in organisational savings.
In each deployment scenario, Simprints provides data on who’s receiving the service, along with a time-stamp and location information – providing clarity about what’s happening in the field, preventing accidental re-enrolment and highlighting gaps in each project’s reach.
“Access to vital services should be a basic right,” concludes Dan, “and we’re proud to be contributing to a more equitable a distribution of resources.”
“Our goal has always been to improve the lives of the world’s most vulnerable citizens and, as we begin to transition Simprints to large-scale deployments, that goal is starting to become a reality.”
Note: Simprints research conducted in partnership with local NGOs and approved by the University of Cambridge Judge Business School Ethical Review Committee. Informed consent was obtained from all participants.