PARIS: An artificial intelligence (AI) program developed in China that is tested by test results, health records and even handwritten notes of diagnosed childhood disorders that are as accurate as doctors, researchers say Monday (Feb. 12).
From the flu and asthma to life-threatening pneumonia and meningitis, the system is equally matched or acquired by the primary care of pediatricians, they reported in Nature Medicine.
Dozens of studies over the past few months have detailed how AI changes the detection of diseases including cancer, genetic disorders and Alzheimer's.
AI-based technology learns and improves the way people are, but is virtually unlimited capacity for processing and storing data.
"I believe it can do most of the work the doctor does," said senior author Kang Zhang, a researcher at the University of California, San Diego, AFP.
"But I will not be replaced by a doctor," he added. d, compared the relationship to an autonomous car that remains under the supervision of a driving man.
"This only allows doctors to do a better job in less time and at lower costs."
The new technology, said Zhang, is the first where AI absorbs unstructured data and "natural language" to imitate the process in which a physician figures out what is wrong with a patient.
"A pediatrician can imitate people to understand and integrate all types of medical data – patient complaints, medical history, blood tests and imaging – to make a diagnosis," he said.
The system is easily transferred to other languages and settings, he added. ] By comparing hundreds of pieces of information about a single patient with extensive storage of acquired knowledge, technology relates to the first statistical methods – and sometimes physicians-and- blood – unseen.
TO NICK OF TIM E
To train the concept-proofing system, Zhang and a group of 70 scientists have taught more than 100 million data points from 1.3 million patient patient visits to a major referral center in Guangzhou, China.
The AI program is defined as respiratory and sinusitis – a common sinus infection – with a precision of 95 percent.
More shockingly, Kang said, it also made less common diseases: severe asthma (97 percent), bacterial meningitis and varicella (93 percent), and mononucleosis (90 percent).
These technologies can only come in time.
"The range of illnesses, diagnostic tests and treatment options has increased exponentially over the years, rendering the decision-making process for physicians that are more complex," according to Nature's a statement.
Experts who are not involved in research say that the study is further proof of the expansion of AI in medical paper
potential health care impunity by assisting the clinic to work fast and accurate diagnosis, "says Duc Pham, an engineering professor at the University of Birmingham.
" The results show that, in general, the system is better than junior doctors. " Machine learning – generating general rules from specific training examples -" can not guarantee 100 percent of the correct results, no matter how many examples of training they use. " AI-based tools for diagnosis are so numerous, especially in interpreting machine-made images such as MRI and CAT scans.
A method launched last months in the United States to see wounds that can lead to cervical cancer that pre-cancerous cells have found 91 percent accuracy, compared to 69 percent for physical tests performed by doctors and 71 percent for in conventional laboratory tests.
Also, a cellphone app based on AI technology made by experienced dermatologists in distin-guising potential cancerous wounds on your skin Among those who are benign, according to a study on the Annals of Oncology.