© New Zealand Broadcasting School 2019

NZ produces the world's first A.I. medical diagnosis platform

Luke Hempleman
Glenn and Hong
Glenn Linde, left, and Dr. Hong Chiong, right. Glenn Linde

"MedicMind" will allow doctors to auto-diagnose a large range of diseases based on a single photograph.

Spokesperson for the Dunedin company, Glenn Linde, said there was no other system that was specifically for medical researchers.

MedicMind is a cloud-based platform where researchers can upload images, training the artificial intelligence to recognise the disease through pattern recognition.

Linde said that the more doctors and researchers used the program and provide data for the intelligence to use, the more accurate it would be.

"Say you teach a kid what an apple is and you don't give very good pictures of apples, the kid might not learn what an apple is very accurately. So if you've got really good data, you can give really good results."

MedicMind founder Dr Hong Sheng Chiong, a Dunedin eye doctor, wanted to use AI with image classification for improving medical diagnosis for patients.

Dr Hong said he envisioned a system that would enable researchers and clinicians to develop their own AI without prior knowledge of coding or programming.

“There are many people like myself out there with a vision to develop the next generation AI for improving healthcare, but the technical skills and cost are overwhelming. Research in the field of medical AI is not accessible to the general public at all."

Glenn Linde said in the past, doctors would need to hire a programmer to try and use an AI for anything like this.

He said MedicMind, with a 'drag and drop' interface, changes that, making it simple and easy for researchers and doctors to use.

Linde said doctors could create their network and then let patients or nurses upload their own photos which the AI could recognise and diagnose, almost removing the need for a doctor entirely.

"In the short term you definitely want to have a doctor... but in the long term I think AI is taking over so much you probably wouldn't even need doctors."