Storyline
British tech company, Skin Analytics has received regulatory approval for DERM—its autonomous AI skin cancer detection system that uses a smartphone.
Dermatologists are in short supply, with an average of only 30 dermatologists per one million population in Europe.
Skin cancer is one of the most treatable forms of cancer if it’s caught early but in the UK alone, NHS England says urgent skin cancer referrals have skyrocketed 170 percent in the last decade, and 11 percent of urgent cases wait over a month just to be seen.
The simple app operates on most smartphones and with a lens attachment called a dermascope. It allows a technician to photograph a patient’s lesion and receive an almost instantaneous diagnosis.
The AI technology compares a single photograph of the lesion with a library of images before giving a diagnosis. Skin Analytics says DERM achieves a 99.8 percent accuracy rate in ruling out cancer.
DERM was awarded the EU’s Class III CE marked medical device under the European Medical Device Regulation, making it the world’s first legally authorized autonomous AI for detecting cancer.
It has already been deployed at 21 NHS (National Health Service) sites in the UK, racking up 135,000 patients assessed. More impressively still, it has helped to detect more than 13,000 cancers, according to Skin Analytics.
“When you think about health care, it’s a very human discipline. And at the end of the day, diagnostics is just one small part of the process. So by using technologies like DERM to be able to assess more patients, find the right patients to get into the dermatologist, we can really drive up the efficiency of our health care system and be able to see more patients faster and get them better outcomes,” says Neil Daly, founder and CEO of Skin Analytics.
DERM can automatically discharge up to 40 percent of urgent skin cancer referrals. It could become an important tool for early diagnosis of cancer and potentially save many lives.
This article and video were provided by The Associated Press.
Script
[Sunbathers on Santo Amaro beach in Oeiras]
[Woman sunbathing]
[Men on sun loungers]
[Kirsten Vehlow explaining the DERM]
[Woman’s neck with skin lesion]
[Lesion on neck]
[The DERM app showing location of skin lesion]
[Camera/light attachment being snapped onto smartphone]
[Kirsten Vehlow photographing lesion on neck]
[Smartphone analyzing the lesion using AI and reporting it is not harmful]
Kirsten Vehlow (interview): “It’s pretty simple and straightforward. So, as you can see here, I’ve got an app open which has got a camera here. And then this is the dermatoscope that you basically connect to the phone, turn on the light and then you can see if I take a picture of Audrey’s lesion here it gives a nice clear view there of the lesion. And what I can do is then decide to use this image. It will check the quality of the image just to make sure it’s sufficient for the AI assessment. And then I can simply submit for assessment here. And within a few seconds we’ll get the results directly on the phone. And here you can see the results with more detail about what it means. And the image will be saved there against the results as well as the location.”
[Neil Daly discussing paper with a colleague]
Neil Daly (interview): “Technology is incredibly accessible because we can use a range of different smartphones to be able to capture the image. We use it with a dermatoscope which attaches to the smartphone to make sure we get a very high quality image, that taken together, enables us to be able to see a lot of different patients very, very quickly because of the availability of the hardware.”
[Smartphone showing the Skin Analytics app opening]
Neil Daly (interview): “DERM is a tool that is another asset we can bring to bear to try and reduce the number of people who die from cancer. When you think about health care, it’s a very human discipline. And at the end of the day, diagnostics is just one small part of the process. So by using technologies like DERM to be able to assess more patients, find the right patients to get into the dermatologist, we can really drive up the efficiency of our health care system and be able to see more patients faster and get them better outcomes.”
[Kirsten Vehlow using the AI app to analyse lesion]
Neil Daly (interview): “So the way that Derm works is that we’ve trained it with a huge number of images. Now where we have the history, pathology outcomes for the lesions that were biopsied. And the AI is able to extract out the features that can identify these cancerous and pre-cancerous lesions. And then it’s able to learn from those so that when you give it a new image, it can identify, is this something that we need to be concerned about or do we need to take action for or is this something that we can safely discharge and send the patient home?”
[Dermatoscope attached to smartphone]
Cherry Thompson (interview): “I think it’s probably a step forward and probably very beneficial. I know some dermatologists and one is probably the top one in the country, and I have great faith in them. Nevertheless, maybe computer diagnosis might be more super efficient, I don’t know.”
[Men sunbathing in park]
[People on boats (punts) on the river Cam using umbrellas to shield from the sun]
Neil Daly (interview): “Well, I’m very passionate about skin cancer because I grew up in Australia, where we have a lifetime risk of two and three. So almost every Australian is going to get skin cancer in their lifetime. But also because it’s one of the few cancers that is visible on the surface of the skin when you can still do something about it. So in the first stage of cancer, skin cancer, the survival rate’s up almost at 99%, it’s about 97%. And it means that if we can get the cancers, then we can get the number of deaths down to almost zero. That is different from many other cancers. And so I’m very, very passionate about making sure we apply this technology to really make a difference in the survival rate for cancer.”
[Person sunbathing in park]
[People by the lake in central Bucharest at sunset]
This script was provided by The Associated Press.
