Affiliate Disclaimer
Some links in this article are affiliate links. We may earn a small commission if you make a purchase through these links, at no extra cost to you. We only recommend products we find useful to our readers
Artificial intelligence is taking the world over by a storm. This doesn’t necessarily require a separate introduction, now, does it? The impacts of these are believed to have proper impacts in emerging fields of medical science too.
A new study (R) conducted by the researchers from the University of Nottingham has found that the computers which do possess the capability of teaching themselves to predict the premature death using artificial intelligence could have an amazing impact in improving the preventative healthcare in the future.
The experience team of researchers, data scientists as well as doctors have not just developed but also tested out a system of computer based machine learning algorithms to predict the risk of early death due to chronic diseases in the population mainly comprising of people in their middle ages.
Upon conducting a few test runs with this specific Artificial Intelligence system, the doctors and the data scientists did find that this specific AI system was very accurate in its predictions and assumptions and was found to perform quite better in comparison to the current approach to the prediction developed by the human experts.
The study was conducted on an extensive scale with the collection of medical data of over half a million people between the ages of 40 to 69 who were recruited to the UK Biobank between 2006 and 2010 and then followed up till 2016.
Dr Stephen Weng, Assistant Professor of Epidemiology and Data Science, who is also the lead author of the study stated that the importance of preventative healthcare is a growing priority in the fight against serious diseases which is why they have been working dedicatedly throughout the years to improve the overall accuracy of the computerized health risks assessment in the general population.
He further stated that while majority of the current available applications do focus on just one chronic disease, it is not enough to detect the possibility of premature death. But, predicting the process of early death owing to the varying number of diseases altogether is a highly complex process, especially with the combination of the environmental and individual factors influencing the condition altogether.
According to Dr Weng, they have actually taken quite a major step in this process in the field of developing a holistic approach to detecting the premature death with the help of machine learning. The process does use the computers help build new risk models dependent that is not just focused around just one model but focuses on varying models like demographics, clinical as well as lifestyle factors along with the kind of diet that one adheres to on a daily basis.
The researchers then went on to map the resulting predictions to the mortality data from the cohort combining the data recorded from the Office of National Statistics death records as well as the UK Cancer registry along with the hospital episodes statistics. Upon assessment, they found that the algorithms did work better and more efficiently to produce with better and much more accurate results in comparison to the standard prediction models developed by the human expert altogether.
The AI machine learning models that were used in the study have been labeled as “random forest” and “deep learning”. The principles and the objectives of this model was what were pitched against the traditionally used Cox model which conducted its prediction based on just the age and gender.
Addressing it all, Professor Joe Kai, who is also a contributing author to this study did state that with the overwhelming growth and need of the intense interest in the usage of artificial intelligence and machine learning to help predict the health outcomes; this is definitely a breakthrough in this. In some instances, these predictions may work and in some others, it might not. This particular study has shown that these subjectives can easily be changed with careful tuning when it comes to the use of the algorithms.
According to Kai, it is important to indulge in better transparency in the process is what can prove out to be beneficial and helpful in aiding the scientific verification and impose better future development in this field of health care. The process of inclusion of artificial intelligence in this field of work might be and sound a little hectic but the process is most definitely one of the best when it comes down to assessing the risks of premature death.
