[ad_1]
Artificial intelligence can analyze smartphone ‘rash selfies’ to diagnose Lyme disease more quickly
- A rash, called Erythema migrans (EM), usually appears a week after a tick bite
- Scientists ‘trained’ computers to identify EM, helping with early detection
- The AI could tell EM from healthy skin 94 percent of the time and other rashes 74 percent
- Untreated, Lyme disease can cause memory issues, chronic fatigue and arthritis
Artificial intelligence can be used to evaluate smartphone photos of suspicious rashes and detect Lyme disease earlier, according to a new study.
Lyme disease affects roughly 300,000 people in the US every year and is transmitted through the bite of an infected deer tick.
A painless rash, called Erythema migrans (EM), usually appears a week or so later, followed by more serious symptoms including fever, headache, chills, joint pain and swollen lymph glands.
Lyme disease is most effectively treated if caught early.
Untreated, it can cause cognitive impairment, chronic fatigue, heart palpitations and painful swelling that can last from months to years.
A team at Johns Hopkins trained computers to scan images of ‘rash selfies’ and differentiate EM from healthy skin, with a success rate of over 90 percent.
Scroll down for video
The AI can differentiate Erythema migrans (top right), the ring-like rash associated with Lyme disease, from healthy skin and other rashes with more than 90 percent accuracy
According to a report in the journal Computers in Biology and Medicine, scientists at Johns Hopkins Applied Research Laboratory used images of Erythema migrans, other dermatological conditions and healthy skin to ‘train’ deep-learning computers to pick out skin infected with Lyme disease.
The AI correctly discerned Erythema migrans from healthy skin 94 percent of the time and had 72 percent accuracy choosing between EM and rashes caused by cellulitis, herpes and other conditions.
In patients already diagnosed with Lyme disease, the system positively identified EM 86 percent of the time.
The rash appears as a ring around the bite mark, starting between 2 and 2.5 inches wide and expanding up to six to 12 inches, or even more.
Lyme disease is caused by a bite from a tick infected with the bacteria Borrelia burgdorferi. Untreated, it can cause cognitive impairment, chronic fatigue, heart palpitations and painful joint swelling that can last from months to years
A team at Johns Hopkins trained computers to scan images of ‘rash selfies’ and differentiate Erythema migrans, the ring-like rash associated with Lyme disease, from healthy skin and other dermatological conditions. The top left four pictures all show Lyme disease rashes
‘Being able to analyze the rash images using artificial intelligence and deep learning enables us to more accurately diagnose that a patient has Lyme disease, determine the stage of the illness and establish a more appropriate treatment plan,’ says John Aucott, director of the Johns Hopkins Lyme Disease Clinical Research Center.
Misdiagnosis of Lyme disease, especially early, is common because the rash can often disappear before more serious symptoms send a patient to a doctor.
Even if it hasn’t, EM can be mistaken for other dermatological inflammations to the naked eye.
Blood tests to detect antibodies for Borrelia burgdorferi, the bacteria that cause Lyme disease, are often unreliable, while skin biopsies are time-consuming and not always available to clinicians.
‘In the near future, if you notice a round, red skin rash, you can take a picture of it as soon as possible to share with your physician,’ Aucott said. ‘That way, even if the rash disappears and you still have Lyme disease symptoms, your physician can make the correct diagnosis using our analysis technique.’
[ad_2]
Source link