When will I die? Artificial intelligence will predict the death time of patients with chronic diseases

When will I die? Artificial intelligence will predict the death time of patients with chronic diseases

March 29, 2019 Source: Sina Technology

Window._bd_share_config={ "common":{ "bdSnsKey":{ },"bdText":"","bdMini":"2","bdMiniList":false,"bdPic":"","bdStyle":" 0","bdSize":"16"},"share":{ }};with(document)0[(getElementsByTagName('head')[0]||body).appendChild(createElement('script')) .src='http://bdimg.share.baidu.com/static/api/js/share.js?v=89860593.js?cdnversion='+~(-new Date()/36e5)];

Sina Technology News Beijing time March 29 news, according to foreign media reports, research shows that artificial intelligence may be able to predict the death time of patients with chronic diseases.

Scientists and doctors have developed an artificial intelligence tool using 500,000 patient data to predict which patients are at higher risk of premature death. The family's family history, salt intake, medication use, and the use of sunscreen were all taken into account.

According to the researchers, the artificial intelligence system's predictions in the test are "very accurate" and the reliability is about 10% higher than the estimates made by existing machine learning systems.

The study was conducted by the University of Nottingham, UK, and Dr. Stephen Weng, assistant professor of epidemiology and data science, led the study. "In the fight against serious diseases, the priority of preventive care is getting higher and higher," Dr Weng said.

“We have been working for many years to improve the accuracy of computer technology that uses computers to assess the health risks of the general population. Most research applications focus on single disease areas, but predicting the probability of death from multiple diseases is extremely complex, especially in Considering the various environmental and individual factors that may affect the impact, we have developed a unique and comprehensive approach to predicting the probability of a person's premature death through machine learning techniques, a major advance in the field."

The artificial intelligence algorithm was generated from 502,648 patient data between the ages of 40 and 69. They participated in the British Biobank study between 2006 and 2010 and have been tracked until 2016. The algorithm considered a total of 60 health predictors, including the subject's body mass index (BIM), blood pressure, vitamins or nutritional supplements. Subjects' intake of fruits, vegetables, meat, cheese, cereals, fish and alcohol was also taken into account.

“We compared the predicted results with the death data of the UK National Statistics Agency's death record, the UK Cancer Registry, etc.” Then they compared the algorithm to two standard machine learning techniques. The results show that the accuracy of the new model is 10.1% higher than the existing technology. "We found that the machine learning algorithm predicts the accuracy of death much higher than the standard prediction model developed by human experts." Dr. Weng pointed out.

Professor Joe Kai, the study's author and director of primary care at the University of Nottingham School of Medicine and Health Sciences, added: "People have a strong interest in using artificial intelligence or machine learning techniques to predict health outcomes. In some cases, this technique may be helpful. Sometimes it is not. In this case, we have shown that through careful adjustment, these algorithms can effectively improve the prediction effect. These techniques may be very fresh and difficult for many researchers in the health field. We believe that As long as these methods are reported in a transparent and clear way, this medical field will be scientifically validated and further developed."

A previous study by the University of Nottingham suggested that there are four artificial intelligence algorithms that predict heart disease with much higher accuracy than the current techniques used in heart disease guidelines.

Scientists predict that artificial intelligence will play a key role in the development of customized healthcare. But they also added that further research is needed to confirm the effectiveness of machine learning in other populations and to better integrate artificial intelligence into everyday care. (leaf)

Transparent Dressing MDK-TD-01

China Transparent Dressing,Transparent Membrane Dressing supplier & manufacturer, offer low price, high quality Transparent Medical Dressing,Clear Surgical Dressing, etc.

Transparent Dressing,Transparent Membrane Dressing,Transparent Medical Dressing,Clear Surgical Dressing

Henan Maidingkang Medical Technology Co.,Ltd , https://www.mdkmedical.com