Sure, artificial intelligence might end up being the downfall of humanity as we know it — that is, if Elon Musk’s fears come to fruition — but for the time being it’s actually quite useful. A new research effort by an international team of scientists reveals that machine-learning algorithms can be a powerful tool for medicine. The group, which published its work in the journal Nature, managed to create and train an AI to successfully identify different types of brain tumors with impressive accuracy.
In order to identify between different types of brain cancer, the team needed some criteria the computer could use to differentiate between them. With over 100 types of brain tumors already in the medical record, the process of identification can be tricky even for human doctors. The researchers used a DNA process called methylation as a sort of biological fingerprint and taught the AI to tell the difference between which of those fingerprints match specific cancer types.
Like all machine-learning algorithms, the AI needed a base of knowledge from which to draw comparisons. The team fed the computer the data of 2,800 cancer patients as a starting point, allowing it to identify an impress 91 different types of tumors. Then, they asked the computer to identify the type of tumor in over 1,000 known samples and found that the AI’s judgement didn’t match up with the human diagnosis in a number of cases.
As it turns out, the computer wasn’t incorrect in its identification; human doctors had misdiagnosed roughly 12 percent of the previously studied samples and the AI was correct.
Image Credit: Shutterstock
News This Week
LAS CRUCES, N.M. — Members of an independent NASA safety panel said they were worried that the Oct. 11 Soyuz launch failure could make safety concerns with the agency's commercial crew program even worse. [...]
Canadian astronaut Col. Chris Hadfield (Ret.) on Friday said that human travel to space will happen sooner than we think, adding that the first destination will likely be a return trip to the moon. [...]
Antibody-based imaging of a particularly aggressive form of breast cancer is undergoing clinical trials worldwide, but the path from trial to application is being hampered by a major obstacle: safety. Concerns stem from inefficient [...]
Following the discovery of graphene in 2003, there has been considerable interest in other types of 2D materials. However, splitting a bulk crystal material into 2D flakes for use in electronics has proven hard [...]
The digital economy is set to unlock tremendous economic value for countries over time. But a common setback for the use of various new technologies is their vulnerability to hackers. That's because companies and [...]
Nanomedical Device and Systems Design: Challenges, Possibilities, Visions now available to rent on Kindle
To accommodate students who wish to read the book at an affordable cost, Nanomedical Device and Systems Design: Challenges, Possibilities, Visions by Frank Boehm (CEO NanoApps Medical Inc.) is available to rent on Kindle. This book benefits [...]
The last five years has seen a surge of attacks on the healthcare industry, with the largest breaches impacting as many as 80 million people. In July this year, it was revealed that 150,000 [...]