You are lying in a hospital bed about to undergo surgery. Catch? It is not a doctor but a robot that performs the surgery.
Although this is not the reality we live in yet, artificial intelligence is already playing an important role in the medical field and has been so since the 1970s. Developments in medical AI have continued to grow since then, now providing doctors with numerous medical benefits, including the ability to detect cancer.
However, as with all technology, there is always room for error.
According to the National Library of Medicine, the first AI medical computer system was developed at Stanford University in 1973. It is currently used in the medical field in a variety of ways, from drug development and disease diagnosis to surgical treatment.
For example, in the Netherlands, AI is helping radiologists improve lung cancer treatment diagnosis, and in Singapore, “bot assistants” are handling clinical inquiries, converting voice case notes into written format, and automatically organizing images and documents. is being used.
Bot assistant and x-ray reader
One of the most common areas of AI in the healthcare industry is radiology, a field of medicine that uses imaging technology to diagnose and treat diseases.
“Once you have all the images that the radiologist can read, you can get a preliminary pre-examination report done by AI,” said Megan Bybee, assistant professor of clinical radiology and imaging sciences at Indiana University. She said, “AI can highlight things like lesions in medical images or X-rays to bring them to the radiologist's attention and say, 'Is this normal?'
AI has proven to be a useful tool for radiologists who need a faster way to sift through large numbers of medical images and X-rays. Using advanced technologies in AI, certain abnormalities can be pointed out to radiologists in a shorter period of time.
Another medical field where AI is being utilized is mammography, which uses X-rays to examine the breasts for diagnosis and examination.
“Studies have shown that AI readings are very similar to humans in their ability to detect early breast cancer in mammography screening,” Bybee said.
The readings performed by AI allow medical professionals to analyze the given results and decide on a course of action.
Still not as smart as a doctor?
But AI isn't perfect. What happens to patients if the technology misidentifies an abnormality or misreads an X-ray?
Tess Barg, a student at Penn State University, is wary of relying on AI in healthcare.
“AI is entering every industry and is an important part of future development, but there is always room for error,” Barg said. “Health care professionals, especially in the field of medicine that deals with humans, should not rely solely on AI. “There should always be someone checking the results.”
A 2023 study from the Society of Radiology of North America found that radiologists actually outperformed AI in noting lung disease on chest X-rays.
More development may be needed for medical AI to become more reliable before it can rely solely on medical outcomes and diagnoses.
“For AI to be more accurate, we need huge sample sizes,” Bybee said. “No one is saying AI will replace radiologists. What they are saying is that AI can be used for efficiency.”