[ad_1]
An artificial intelligence (AI) program developed in China that badyzes test results, medical records and even handwritten notes, helps diagnose childhood diseases as accurately as doctors, said Monday researchers.
From influenza and asthma to life-threatening pneumonia and meningitis, pediatricians in primary care have always compared or surpbaded the system, they reported in Nature Medicine.
Dozens of studies in recent months have shown in detail how AI is revolutionizing the detection of diseases such as cancer, genetic diseases and Alzheimer's disease.
AI-based technology learns and improves in the same way as humans do, but its processing and data storage capability is virtually limitless.
"I think he'll be able to perform most of the tasks of a doctor," said lead author Kang Zhang, a researcher at the University of California.
"But AI will never replace a doctor," he added, comparing this relationship to an autonomous car that remained under the supervision of a human driver.
"It will simply allow doctors to do a better job in less time and at a lower cost."
According to Zhang, the new technology is the first in which AI absorbs unstructured data and a "natural language" to mimic the process by which a doctor discovers what is wrong with a patient.
"It is possible for a human pediatrician to interpret and integrate all types of medical data – patient complaints, medical history, blood tests and imaging – to make a diagnosis," he said.
The system can easily be transferred to other languages and settings, he added.
By comparing hundreds of information on the same patient and on a vast knowledge base, technology can uncover links that previous statistical methods – and sometimes real-life doctors – neglect.
Just in time
To form the proof of concept system, Zhang and a team of 70 scientists injected more than 100 million data points on 1.3 million pediatric patient visits to a large referral center in Guangzhou, China.
The AI program has diagnosed respiratory infections and sinusitis – a common sinus infection – with an accuracy of 95%.
More surprisingly, said Kang, the same goes for less common diseases: acute asthma (97%), bacterial meningitis and chickenpox (93%) and mononucleosis (90%).
Such technologies may well come to the fore.
"The range of diseases, diagnostic tests and treatment options have increased exponentially in recent years, which has complicated the decision-making process of physicians," Nature said in a press release.
Experts not involved in the research said the study was further evidence of Amnesty International's growing role in medicine.
"The job has the potential to improve health care by helping the clinician make quick and accurate diagnoses," said Duc Pham, an engineering professor at the University of Birmingham.
"The results show that on average, the system worked better than younger doctors."
"But that will not replace the clinicians," he added.
The machine learning – which forms general rules derived from specific training examples – "can not guarantee 100% correct results, no matter how many training examples that". they use".
AI-based diagnostic tools abound, especially for interpreting machine-generated images such as MRIs and CT scans.
In the United States, a method unveiled last month in the United States to detect lesions that can cause cervical cancer has revealed precancerous cells with an accuracy of 91%, against 69% for physical examinations by doctors and 71% for conventional laboratory tests.
Likewise, a cell phone application based on AI technology has allowed experienced dermatologists to distinguish potentially cancerous skin lesions from benign lesions, according to a study by the Annals of Oncology.
[ad_2]
Source link
