What is the name of your organization?
CIDACS - Oswaldo Cruz Foundation - Center for Data and Knowledge Integration for Health
What is the name of your solution?
iACS - AI for Community Health
Provide a one-line summary or tagline for your solution.
A conversational AI agent tailored to help the community health workers
In what city, town, or region is your solution team headquartered?
Salvador, State of Bahia, Brazil
In what country is your solution team headquartered?
BRA
What type of organization is your solution team?
Other, including part of a larger organization (please explain below)
If you selected Other, please explain here.
CIDACS is a part of Oswaldo Cruz Foundation, an public health institute in Brazil
Film your elevator pitch.
What specific problem are you solving?
Global interest in the role of community health workers (CHWs) in Primary Health Care (PHC) is driven by substantial evidence of their success in providing a broad range of preventive, promotive, and curative services in reproductive: maternal, newborn, and child health; infectious diseases control; assisting actions to address non-communicable; supervising malaria, tuberculosis and HIV treatment.
To better promote equitable universal coverage, PHC requires that information is managed steadily to be able to deliver care where, when and to whom it is most needed. Yet, in most places, essential care still relies on paper forms or information systems that require re-entering identification data and patient history repeatedly. We deploy the AI agents to the community health workers to solve that challenge.
Every geographic location uses different forms, types of information systems and local rules. However, the CHW work process has basic comonalities: family house visits, health problems assessment, taking notes, talking to local doctors and nurses, making sure people vaccinate, making appointments and delivering medicines.
We designed a conversational AI Agent that connects to the local health information system (LHIS) and helps CHW with those tasks, leveraging information from previous visits and family records.
What is your solution?
We are developing a conversational artificial intelligence (AI) agent that connects to the local health information system (LHIS) and assists community health workers (CHW) in their daily tasks. It is named iACS, which means artificial intelligence for the Agente Comunitário de Saúde (the Brazilian term for community health workers - CHW).
With iACS, instead of filling long forms or complicated apps, one could just speak or text: “I visited John at street A number 45 for his hypertension. He is fine” and IACS would fill the forms and send them to the LHIS for billing and disease notification follow-up. By the end of the day the CHW would just check if all the forms are filled correctly and dispatch them.
As a next step, the now structured data is converted to brazilian health interoperability standards to feed the LHIS. Fiocruz Primary Healthcare Dashboard (https://solve.mit.edu/solutions/63515) backend enables us to connect to the LHIS of any Brazilian city effortlessly.
Who does your solution serve, and in what ways will the solution impact their lives?
Community Health Workers (CHW), in several health systems in different countries and continents, deliver health care as part of PHC teams in strategic health care situations.
With iACS, we estimate up to 20-30% timesaving. iACS also helps them manage their time and priorities better, reminding them of priority families and patients to be visited on the planned route. The impact on community health is potentially enormous, as there is substantial evidence of CHW impact on reducing infant mortality and the burden of chronic and neglected diseases.
Our testing ground and scaling up scenario is in Brazil, a global reference in PHC with more than 400 thousand CHW. We are ready to reach the majority of Brazilian territories within a year, and we are convinced that the solution is adaptable to most of the world regions, potentially reaching up to 5 million CHW.
Every local health manager should be able to analyse their own data, empowered by data science training, privacy best practices and locally run artificial intelligence foundation models. Such information can help national managers track health trends, identify disease threats and improve community health worker programmes. All of these actions depend on the quality of the data collected.
