During a recent PW Podcast episode, we spoke with Vlado Bosanac, Chairman & CEO at Advanced Human Imaging Ltd, developers of body scanning technology that utilizes smartphone cameras and artificial intelligence that are embedded into digital apps. Following is a summary of that interview:
How does your technology work?
We use the incredible diagnostic capability of the mobile phone, along with its camera functional- ity. We allow a consumer to utilize their own mo- bile phone to take images of lesions or marks on their skin that they’re concerned about and put it through our machine-learning algorithms that provide a result back to them in regard to what it might be. And skin cancer is just one of those particular elements or conditions we can look at. We have 588 conditions built into the artificial intelligence (AI) engine, across 133 categories, from mild rashes to obviously the more destruc- tive side of the skin cancers for which we’re able to give people an idea of what it is they’re looking at. They can then make a more informed decisions on the next plan of action or treatment that may be required through their dermatologist.
Called DermaScan, the technology has been used globally more than 700,000 times in the last year, including by more than 220,000 clinicians as a second check on their own visual sighting. We’ve also run a number of tests with dermatologists to see how it compares, from the AI perspective, against them with their knowledge and human interaction. All instances have either been a draw or we’ve beaten the dermatologist in the time and outcomes produced through the AI.
The technology has been used in patients with a number of ethnicities, skin types, and conditions, lesions, or skin disorders. We have the same vali- dation across all skin types and tones, with great accuracy whether the user is of Asian, African- American, or Caucasian descent. That’s because the data on which the technology is built is based on millions of impressions that give the AI engine the ability to learn across skin tones and types.
We want to give individuals enough informa- tion to make a thought-out next step using their personal device, on which we list three to five disorders or conditions that might be what they have. They’re able to click on any of those and see a full overview of what having that condition or disorder could mean and to have their dermatol- ogist check it out. We’re not trying to replace the dermatologist but instead triage patients into the system to determine what attention is likely need- ed based on what is detected through their images.
I think the technology helps put people’s minds at ease to some degree, because it’s human nature to wait until there’s an event to take care of some- thing; most of us aren’t proactive enough in our own health. So, we want to support people with good information to help them understand that what they have could be something they need to pay attention to and have looked at by a derma- tologist.
