A Brazilian health technology company and a public university have launched a research partnership aimed at reducing missed medical appointments through predictive artificial intelligence. Wellon and the State University of Feira de Santana, known as UEFS, are collaborating during this semester to validate machine learning models that anticipate patient nonattendance. The initiative is designed to recover lost clinical capacity, strengthen financial performance, and improve access without expanding physical infrastructure or increasing staff levels.
Tackling a Costly Healthcare Problem
Missed appointments consume between 20 percent and 30 percent of scheduled medical agendas in Brazil, according to data shared by the partners. This idle capacity undermines revenue, limits access to care for patients who need timely attention, and places pressure on clinical teams. The study applies Wellon Pulse, a predictive system designed to map patient behavior and anticipate the probability of no-shows before they occur.
How the Predictive System Operates
Wellon Pulse uses active machine learning to analyze the patient journey and identify patterns associated with the likelihood of absence. The platform then recommends preventive scheduling adjustments by specialty and shift, helping clinics fill open slots strategically before missed appointments disrupt daily operations. Wellon reports that the system serves more than 1,000 healthcare facilities and connects approximately 2 million patients per month, with a daily confirmation rate above 80 percent.
Expected Operational and Financial Gains
The partnership estimates that the technology can reduce unexcused absences by up to 50 percent and increase healthcare revenue by as much as 25 percent. These improvements are intended to optimize existing schedules rather than require additional hiring, new equipment, or physical expansion. According to company data, the solution already supports a high volume of confirmed appointments, allowing institutions to use their installed capacity more efficiently.
Academic Validation for Real-World Impact
For Wellon CEO Eduardo Nunes, working with a university brings scientific method and advanced researchers into the product development process. He said predictive artificial intelligence in healthcare requires formulating hypotheses, conducting experiments, and measuring results before market deployment. The UEFS collaboration is designed to test the algorithm’s effectiveness in real operational environments and provide independent academic validation for the technology.
Conversational AI and Safer Data Handling
Beyond predictive scheduling, Wellon has developed Letícia, a conversational AI assistant that interprets unstructured data such as photos of medical orders sent by patients. The tool converts these images into structured digital information, reducing manual typing and the transcription errors that can compromise care. Nunes said the focus is to automate processes safely while keeping human oversight when necessary, improving accuracy at reception points.
A Growing Digital Health Platform
Wellon Digital is a health technology company founded in 2020 in Bahia, specializing in conversational intelligence and automation for the patient journey. Its platform integrates with the official WhatsApp API and automates scheduling, clinical triage, active confirmations, and satisfaction surveys. The company works with hospitals, laboratories, and health insurers under compliance with Brazilian data protection law, serving a broad base of healthcare clients.
The Wellon and UEFS partnership shows how academic research and private innovation can combine to address operational inefficiencies in healthcare. By validating predictive and conversational AI in real-world settings, the collaboration may provide a model for reducing absenteeism and improving financial sustainability. If the results confirm the projected impact, the initiative could support more accessible and dependable care for patients across Brazil while offering clinics a practical path to greater efficiency.