June 5th According to Wired magazine, researchers at the University of Adelaide have conducted an unprecedented study using artificial intelligence to predict human life, with precision comparable to that of doctors.
Artificial intelligence predicts which patients will die in the next five years by analyzing the patient's chest image with an accuracy of 69%, which is comparable to the accuracy judged by human doctors.
Although predicting death sounds unpleasant, research results can have a major impact on the early diagnosis of serious illnesses, helping doctors to personalize individualized patients.
“As doctors are unable to measure the health of every organ in a patient, accurate assessment of biological age has been limited,†said Dr. Luke Oakden-Rayner of the Adelaide School of Public Health.
Researchers used "deep learning" techniques to let computer systems learn to understand and analyze perspective images. Although the researchers were unable to determine what the computer saw from the chest radiograph, the computer's prediction accuracy for patients with severe chronic diseases such as emphysema and congestive heart failure was high.
Dr. Oakden-Rayner said that automation systems focus on predicting medical outcomes rather than diagnosing disease, opening up new avenues for the application of artificial intelligence in medical image analysis.
Although the subject of the study attracted attention, the sample of 48 patients was relatively small. The research team plans to expand the number of samples to tens of thousands in the next phase of the study, and also envisage new usage scenarios, such as letting AI predict the onset of heart disease.
Mini Centrifuge,Laboratory Mini Centrifuge,Centrifuge Machine For Lab,Samll Centrifuge Machine
Guangdong Widinlsa International Co.Ltd , https://www.gdwidinlsa.com