[ad_1]
A new study in China found that an AI system can help some doctors diagnose common childhood diseases. The study, published in Nature Medicine yesterday (Feb. 11), was co-authored by a long list of Chinese and American researchers. It is only one of the last in China to look into the use of AI in medical diagnosis.
The study trained an in-depth learning system on 101 million data points generated from electronic records of 1.3 million patient visits to a Guangzhou medical center. The researchers found that the AI system was able to meet or surpbad two groups of junior doctors by accurately diagnosing a range of diseases, ranging from asthma to pneumonia to sinusitis and diseases related to the mouth. The IA has also been able to meet or exceed diagnostic performance with some groups of experienced physicians, for example in the category of upper respiratory problems.
In some cases, the system was able to diagnose conditions with an accuracy of 90 to 95%. In any category, the AI model plunged below a diagnostic accuracy rate of about 79%, higher than a group of younger physicians, but lower than that of l & # 39; other. Experienced physicians have generally been more successful than the AI system.
One of the limitations of the study, however, is that researchers pulled all their data from visits to a single medical facility in China, raising questions about its applicability outside the country and even to Guangzhou facility itself. (As is often the case with artificial intelligence systems, the algorithm tests the same source data on which the model is based, which raises further questions about reproducibility. )
Nevertheless, the study complements the growing body of research demonstrating the benefits of AI in diagnosis. For example, a study published last October, also in China, showed that another algorithm accurately predicts 88% of patients waking up in a vegetative state. This happened a few months after the US Food and Drug Administration approved a diagnostic system for diabetic retinopathy using AI, which can lead to blindness if it is not not treated.
Other work has also shown the utility of AI in image-based diagnostics. Last year, an AI system beat doctors to accurately diagnose brain tumors during a competition in Beijing. It is important to note, however, that these systems are more likely to serve as useful diagnostic tools for physicians than to completely replace their judgment.
It is not a surprise if much of this research comes from China. The country has invested huge sums in AI research. The number of articles published on this topic has risen by 400% since 2007. The use of this technology to improve medical care is of particular importance as the country seeks to standardize care in rural areas. and urban areas, cope with the aging of the population and improve the efficiency of a health system that needs many more doctors and nurses.
Source link
