Earlier this year, Forbes reported that artificial intelligence (AI) algorithms are showing promise in performing medical work which was known only capable to be conducted by human physicians.
The results of a recent study, which was aimed at diagnosing the presence or absence of tuberculosis (TB) in chest x-ray images with the assistance of AI, revealed remarkable accuracy. Researchers first “trained” the AI models with hundreds of x-ray images of patients without and with tuberculosis. Next, they tested the AIs with 150 new x-rays. The systems achieved an impressive 96% accuracy rate – way supersedes that attained by human radiologists.
An innovation in the medical worldA newer discovery by a team at University of Adelaide, Australia who has been working to create an AI system that can predict a person's lifespan simply by studying images of their organs gathered early results indicating AI’s predictions are just as good as a human doctor.
The team used AI to examine the medical images of 48 patients' chests. Result showed that AI predicted which patients would die within five years – with an accuracy rate of 69%, comparable to manual predictions by doctors.
"Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual," says lead author Dr Luke Oakden-Rayner. "Instead of focusing on diagnosing diseases, the automated systems can predict medical outcomes in a way that doctors are not trained to do, by incorporating large volumes of data and detecting subtle patterns," he adds.
Although using AI in medicine is still in its early stage, Isaac Kohane, a doctor and the chair of Harvard Medical School’s department of biomedical informatics, predicts that AI will start having a noticeable effect on medical imaging within the next three years. He opines that physicians might want to start thinking about adapting to the future changes.
AI vs. modern healthcare: Healthcare professional share concernsIn spite of studies revealing substantial evidence, Venture capital arm of Singapore Infocomm Development Authority, SGInnovate, however, revealed that convincing doctors and paediatricians might be “an issue”.
"Doctors and clinicians are not so excited sometimes about new models or new processes – and I don't mean that disrespectfully. It's just that's a reality," expressed Steve Leonard, chief executive of SGInnovate. He also added that regulation would also be a key factor to consider.
Singapore recently increased investments into AI by committing more than SGD100 million over the next five years. The country plans to use AI to address major challenges affecting Singapore society, including health care.
Nonetheless, several doctors believed that despite what AI can do, it could not take over human relationship.
"Health and health care is too human a notion for AI alone to cure it," says Rasu Shrestha, chief innovation officer at University of Pittsburgh Medical Centre and chief of the division of radiology informatics.
Alex Harding, a medical doctor conducting his residency at Massachusetts General Hospital gave a real-life example of example of a patient with a chronic disease, who refuses to have her laboratory tests checked. “I think for her its fear of having bad news,” Harding says. “No computer in the world is going to convince her to go in.” Harding believes that a good doctor is the only one who could do such “convincing”.
Hand-in-hand, complimenting one and another – rather than taking over
Bryan Vartabedian, MD and assistant professor of paediatrics at Baylor College of Medicine said that doctors should see themselves as participants in shaping the future rather than victimising themselves.
“No matter what machines can and cannot do, the progression of technology calls upon us to undertake a critical rethinking of what we are doing now,” he wrote.
Dr Kohane predicted that AI will only cover 5% – 10% of a doctor’s daily routine and it will take a longer time to come-up with a “general-purpose AI” – where a machine can give a “full diagnosis and treatment plan like a competent and well-educated primary care doctor.”
“Most of what we do in healthcare is not about making a diagnosis,” says Dr Harding. “It’s about working with patients, who are humans, to find an appropriate treatment. That requires a human relationship… which develops over time and can’t be replaced by a computer.” MIMS
The power of AI: 3 systems to assist doctors and patients
First implant surgery carried out by a robot marks significant dental breakthrough
Could technology be a bane to healthcare?
Watson: Transforming healthcare with Artificial Intelligence