Elsa: AI-powered Pediatric Health Assistant Tool
Inspired Ideas is part of UNICEF’s Innovation Fund Investments in Skills and Connectivity
Inspired Ideas is building the future of healthcare by leveraging advances in technology and data science to build smart tools that augment the capacity of lower-cadre healthcare providers in rural and urban Tanzania.
We’re developing the Elsa Health Assistant, a clinical decision support tool that empowers healthcare providers with data-driven algorithms for consistent and optimal decision-making. We are passionate about building tools and services that improve health outcomes and ensure that children have access to high-quality healthcare.
The Elsa Health Assistant is a digital tool that offers automatic symptom assessment, decision-making support, and next step recommendations. Offered as a mobile application in both English and Swahili, the Elsa Health Assistant brings specialist-level decision-making to all health providers, standardizes data collection, and increases the quality of care provided to pediatric patients ages 0-14 years.
Health providers input patient demographics, vitals, symptoms, and test results (when available), and receive insights and next steps recommendations about their patients’ health. Elsa supports common pediatric illnesses (rash, gastroenteritis, etc.), nutrition-related illnesses, high-mortality illnesses such as malaria, sexual and reproductive conditions (including HIV/AIDS), and supports healthcare providers in identifying early childhood development milestones.
Inspired Ideas has been at the forefront of developing machine learning solutions for healthcare because we believe that poor health outcomes for children creates both an imperative to act and an opportunity to learn.
We’re creating the Elsa Health Assistant because we saw a gap in access and delivery of healthcare services in Tanzania. The country of Tanzania faces significant health challenges and its Ministry of Health has recognized the need for improving health delivery and outcomes, particularly related to child and maternal health. The country has one of the lowest doctor-to-patient ratios in the world (1:20,000), with almost 75% of those doctors working in urban areas. This inequity means that children are significantly limited by their health outcomes and die unnecessarily of preventable health conditions.
In addition, providers at the first point of care often have limited resources to diagnose, treat, and support their patients. A study by the World Bank in 2016 found that nurses in rural parts of the country correctly diagnosed a patient only 37% of the time. Since specialist pediatricians are centralized in large cities and often inaccessible by both providers and patients, it is critical that we strengthen the capacity of lower-cadre healthcare providers and provide them with tools for more accurate decision-making.
The Elsa Health Assistant utilizes cutting edge technology, including artificial intelligence and machine learning, to learn complex connections and specialist-specific information, greatly increasing the availability of specialist knowledge in rural areas that lack these services. For example, in rural villages - where there is no pediatrician- Elsa Health Assistant can detect high-mortality illnesses to impact the lives of millions of children in Sub-Saharan Africa.
Artificial intelligence has been successful in other countries at catalyzing healthcare systems. Although the mHealth space is growing quickly in Tanzania, Elsa is among the few AI-powered health solutions supporting clinical decision-making at this level. We are able to scale more effectively than telemedicine platforms that rely on the current number of available physicians, and we can quickly integrate knowledge from national guidelines and local experts to add contextual relevance to our models.
As emerging technology-based tools become more common in the healthcare delivery system, Elsa will grow to empower even more patients and healthcare providers to better identify diseases and improve treatment outcomes.
The Elsa Health Assistant is integratable with other digital health tools and its development is based on standardized government data collection forms. This facilitates easier monthly and quarterly reporting required by the government. Our tools are built and trained on locally sourced data, adding contextual relevance to our models.
We aim to give healthcare providers the resources, skills, and knowledge they need to make better decisions with confidence. Being open source will allow us to reach a wider audience and leverage the community to further perfect tools for decision support. We will be able to reach and connect with more people who share our vision and passion for improving healthcare.
Our team is diverse; we come from a range of backgrounds and we bring experience from multiple disciplines to our work. We have collective expertise in machine learning, artificial intelligence, software development, health research, medicine, biological sciences, and international community development. We are deeply connected in the healthcare ecosystem within Tanzania, and have knowledge of the challenges faced by health stakeholders across the country.
The diversity of our team, our stakeholders, and the communities we serve is critical to developing a solution that works for everyone. Our various personal and professional experiences have allowed us to create a thriving, engaged team where everyone brings a unique perspective to solve the problems at hand. Regardless of our diverse backgrounds, we all share a passion for creating the future of healthcare.
THE WAY AHEAD
Our vision is to build the future of health and create a world in which all people have access to equitable healthcare. With the investment from the UNICEF Innovation Fund, we are able to focus on advancing and perfecting our technologies, deploying and iterating it within regions across Tanzania, and scaling the reach of our impact. This scale and growth will increase our organization’s value and add to our track record as a capable and forward thinking team. We expect to leverage the development of our solution to improve the lives of children and expand the use of AI-powered technologies for health to new markets.
We are incredibly excited to be a part of a larger cohort of innovators and problem solvers in the health space. We look forward to meeting with, and learning from, other cohort members and mentors who are impacting the health of millions of children around the world.