Ditto AI
Blog

How people are using artificial intelligence in medicine

Posted by Ditto on Dec 11, 2019 8:15:00 AM

By 2021, artificial intelligence in medicine is expected to be worth US$6 billion, according to a Frost and Sullivan report.

AI has informed the direction of many areas in the healthcare sector. For example, AI is powering mobile apps to diagnose patients and ease up hospital resources. It’s also increasing healthcare access in underdeveloped parts of the world.

Here are just a few more ways people are using artificial intelligence in medicine.

 

At Moorfields Eye Hospital, a new way to diagnose patients

At Moorfields Eye Hospital in London, eye doctors developed an algorithm to recommend the correct treatment approach for more than 50 eye diseases, with a 94 percent accuracy score.

The technology – built by Google’s DeepMind – identifies diseases like age-related macular degeneration (AMD), which causes loss of sight and affects more than 170 million people worldwide.

AI can increase bank's revenues by as much as 30 percent text-only CTA

The next step, according to DeepMind, is to ‘help predict severe eye diseases in advance, that could help clinicians prevent sight loss before it even occurs’.

 

Using AI in imaging and radiology

Radiology is an area of medicine where AI is making a difference. In a study published in the Journal of the National Cancer Institute, an AI system matched the accuracy of more than 28,000 interpretations of breast cancer screening mammograms, which were conducted by 101 radiologists.

Imaging AI also means that departments like Pathology can now go digital, instead of having to hold film up to light boxes. The benefits of this include image sharing, remote consultations and tele-pathology for 24/7 servicing.

 

Electronic health records and patient access

Before artificial intelligence, a patient would attend the GPs office, explain what’s wrong, and be diagnosed by a doctor. And while medical history was an important factor in that diagnosis, it was a tricky thing for a human doctor to get access to and sieve through to find patterns or anomalies.

AI solves this concern, and consequently, it is beginning to help personalise treatment. If a patient was diagnosed with a heart condition a decade ago, an AI could quickly see this in their records and offer a treatment program that catered to specific requirements.

 

Using AI to identify cancer

According to the Skin Cancer Foundation, one in five Americans will develop skin cancer by the age of 70, and more than two people die of skin cancer per hour. As scary as this sounds, Triage is trying to reduce these statistics by  ‘detecting diseases in a snap’.

The company is using artificial intelligence to detect a range of skin diseases using a smartphone. By snapping a photo of your skin ailment, Triage runs it through their machine learning algorithm, which can detect thousands of skin conditions and symptoms.

 

AI in medicine is a work-in-progress

The concern with a number of these applications is: Can these technologies be trusted?

If, for example, someone snaps a photo of a mole on their back and Triage says it is benign, when in fact it is cancerous, who’s accountable? 

At Ditto, we’re using AI to automate human knowledge and expertise. We’re finding new ways to provide accountable, explainable audit trails for decisions made by technologies so that humans can retain complete autonomy and oversight.

While the potential of AI in medicine is massive, the technology is only as good as the data we feed it.

AI can increase banks' revenues by as much as 30 percent wide CTA

Topics Future of AI AI at work Machine Learning