Global childhood vaccination rates have reached unprecedented levels in recent years, with many preventable diseases experiencing significant decline. Despite this progress, the World Health Organization reports that essential vaccine availability has plateaued internationally, creating challenges for healthcare systems worldwide.
Healthcare facilities, especially in resource-limited areas, struggle with accurately forecasting patient attendance for immunization services. This uncertainty results in either critical vaccine shortages that leave children unprotected or costly surpluses that expire before utilization.
Enter macro-eyes, an innovative startup developing an AI-powered vaccine forecasting solution that integrates diverse real-time data sources with valuable insights from healthcare workers on the front lines. Their groundbreaking platform, the Connected Health AI Network (CHAIN), demonstrated remarkable success by reducing vaccine wastage by 96% across three Tanzanian regions. Following this achievement, the company is expanding its implementation throughout Tanzania and Mozambique.
"Healthcare systems operate in complex environments with incomplete data," explains Benjamin Fels, CEO of macro-eyes and co-founder with MIT's Suvrit Sra. "Traditional predictive models requiring comprehensive patient information fail when essential variables are missing. Our technology must function effectively regardless of data availability or setting."
The company's predictive methodology has already produced another successful product, Sibyl, a patient scheduling platform that has analyzed over 6 million hospital appointments and dramatically reduced waiting times by 75% at a major U.S. cardiac facility. Sibyl's analytical capabilities now enhance CHAIN's comprehensive forecasting abilities.
Both solutions represent macro-eyes' broader mission to revolutionize healthcare through artificial intelligence. By developing systems that function effectively in data-scarce environments, they're simultaneously advancing the entire field of AI technology.
"Machine learning breakthroughs emerge from tackling fundamental challenges in the world's most demanding settings," Fels notes. "When we address problems where resources are limited, AI becomes more intelligent, efficient, cost-effective, and resilient."
Developing a Vision
The collaboration between Sra and Fels began approximately a decade ago when Fels worked as an algorithmic trader and Sra served as visiting faculty at UC Berkeley. Their experience applying quantitative methods across different industries revealed significant inefficiencies in healthcare systems.
