Register for free and continue reading
Join our growing army of changemakers and get unlimited access to our premium content
Kang Health is an Israel-based startup that is using AI combined with patient information to offer more accurate online diagnoses.
Credible research show high levels of medical mistakes: 10 percent of patients at hospitals are treated in the wrong way, at least 25 percent of US patients get superfluous and unnecessary treatments and unnecessary deaths are the third leading cause of death in the US. The medical profession then, seems like an obvious space for innovations using AI. We recently wrote about a Chinese chatbot that uses AI to support doctors by speeding up the process of diagnosis. Now Israel-based startup, Kang Health is a website and app that uses AI to provide more accurate medical information to people at home.
Kang Health mines the internet to look up medical symptoms online. Simple internet symptom searches often lead to false diagnoses; Kang uses information about patients to provide more reliable results. Users enter their age, gender and the symptoms they are experiencing and the system mines through search results of similar cases from people with the same demographic data and symptoms. It offers them the typical diagnoses, or suggests they require medical assistance. Although the technology is mining predominantly user-generated information, Kang Health employ two full-time doctors to validate the information coming in to ensure its accuracy. The service also allows users to store their medical history and lifestyle for higher quality results. Founder Allon Bloch explains, “Kang will liberate consumers from the shackles of the lowest common denominator of low quality health related online data… We don’t pertain to dispense medical advice; we are simply providing consumers with the ability to be directly involved in making decisions about their own body.”
Kang Health is set for release in April or May 2017. Could this system be used to support accuracy in other sectors?
Please login or Register to leave a comment.