Piotr Orzechowski remembers exactly the moment he realised he had to create a more effective symptom checker. “I read that a lot of people were checking their medical conditions or symptoms online,” he says. “However, for something very simple like a headache, individuals were ending up with scary conclusions, anything from a migraine to a brain tumor.”
Realising this would only increase uncertainty and unnecessary fear, in 2012 he decided to team up with a group of physicians to build a reliable source of medical information. They developed a B2C symptom checker, available online and in a mobile application for people in Poland. Within just a few months, it became hugely popular in the country.
“We then wanted to scale our business up,” Piotr adds. “We thought a better approach would be to provide the technology to existing healthcare organizations, which are clinically credible and have better access to public and other services. That was when Infermedica transformed into a B2B vendor, providing AI-driven solutions for enterprise clients.”
Currently, Infermedica has created three widely-used products: Symptom Checker, Call Center Triage and Medical API. All of them shares the same underlying intelligence known as the Diagnostic Engine. It is a software developed on past research evidence and anonymized patient information including symptoms, risk factors, chief complaints and lab results and is able to estimate the likelihood of different medical conditions using a comprehensive statistical knowledge base.
Since its inception in 2012, the company put together a group of engineers, data scientists and medical professionals to trawl through past literature and more than 18,000 hours of physicians’ work to upkeep the accuracy and efficacy of all its products. “We want our products to be user-centric,” Piotr explains. “For example, whenever someone is felling discomfort, the Symptom Checker will ask them a series of questions. These questions were selected based on algorithmic reasoning of the Diagnostic Engine. That means, the patient will receive an answer and recommendations that are closest to what may be the most likely conditions.”
Currently, Infermedica is piloting a new product specially designed for emergency rooms as an answer to the never-ending queue of patients. Piotr aims to build a tool that will better prioritize patients’ needs and facilitate intake to ease the workload of physicians.
Piotr insists, Infermedica is not a typical startup driven by finances, and regards setting up a company dealing with healthcare as a huge challenge because the industry is demanding and has a much higher barrier of entry. Nonetheless, the increasing cost of medicine and healthcare and misdiagnoses mean there are ample opportunities for improvement and change.
“We are constantly demonstrating the possibilities of our solutions to fellow physicians and investors, especially those who are still unconvinced about the idea of matching AI and healthcare together,” he explains. “While we try to keep ourselves competitive, we must remember the most important thing is always the patient. That’s why sometimes, I think it’s better to slow down and think everything through, observe the results, and evaluate what could be improved before we move on. Ultimately, patients’ health is in our hands and if we are to succeed in our mission to improve healthcare for everyone, the patient has to feel confident and safe.”