Google employs AI for predicting death | Big opportunity for palliative care startups

AI for predicting death

Google’s AI for predicting death

There is a new system powered by Google’s AI. It can accurately predict if a person will die within 24 hours starting from the time he or she was admitted into the hospital. One exciting thing about this unique technology is the fact that it can accurately predict the death of such a person by up to 95%. The AI can achieve this by directly analyzing data which was previously not within reach; this includes notes which were lost in PDFs as well as those which were written on charts in addition to any other relevant information. What it then does is to absorb this information, analyze them and then make the most reasonable prediction. The outcome is quite interesting because it gives information that is quite faster and of greater accuracy when compared with what is obtainable with traditional methods.

This unique technology will find great use in hospitals and clinics, this form of artificial intelligence will be quite useful because it can tell the outcomes of patients such as how long the patients will remain in the hospital, the possible chances of them being readmitted.

Without a doubt, this innovation will go a long way in impacting the health market as well as the life of the people. First and foremost, it ought to help in driving personalized medicine while also helping improve the quality of healthcare. Google’s AI technology ought to be able to come up with scalable and accurate predictions for different clinical situations which should assist doctors in their quest for better healthcare delivery.

If Google can smoothen the process of getting data entered and worked on the way through which data is being utilized, it should go a long way in reducing human error in the field of medicine.

The notes and information retrieved by the AI could go a long way in helping Doctors come up with a better diagnosis at a faster and more accurate rate. Research has shown that doctors tend to fail in their duty to send patients in for palliative care because they sometimes make a prognosis that may be too optimistic.

Recommended for you
Google’s DeepMind detects eye diseases with 94% accuracy
Stanford researchers develop a wearable device to measure stress
Google Glass helping kids fight autism, a new study reveals

How are Google and Stanford University transforming palliative care?

Stanford University has also come up with a unique innovation which seeks to improve palliative care. This technology makes use of an artificial intelligence algorithm which has been trained using the patient record of about 2 million patients. The developed algorithm can detect specific patterns which just one doctor may not be able to identify, and making use of these patterns; it can make an educated guess as regards when a sick patient may die. What the algorithm does is to take a detailed and calculated look at the data and after that make a decision based on the data.

Stanford’s algorithm depends on deep learning, which is the standard machine learning technique which utilizes neural networks to sieve and learns from a large number of data.

Without a doubt, this new AI technology will go a long way in helping to transform the lives of people who may be suffering from one form of chronic disease or terminal illness or the other. This is possible because the data it collects concerning such individuals such as drugs taken as well as scans and procedures have undergone etc. this data is then used for the purpose of helping to ascertain the possible outcomes of such persons.

This, in turn, helps to mitigate the issue of uncertainty that may exist in the lives of such individuals, which means they will be able to tell how long they would live.

This would mean that people who are critically ill would be able to choose whether they want to go home and spend their last days with the ones they love or they want to spend an extended period in the hospital for care.

It, therefore, suffices to say that it is indeed a good thing because it is statistically proven that below half of the eight percent of admissions into hospitals who need palliative care get it. This can have grave effects if health status of the patient nosedives quicker than expected which may mean that such persons would be forced to spend the last part of the lifetime receiving treatments that may only extend their lives for a few more weeks.

This means that instead of spending their last days with loved ones, they would spend it on the hospital bed. This new technology should help ensure that the majority of the extremely ill patients get as comfortable as they can in the final months of their lives while also getting the care they prefer.


In summary, the exploits of Google and Stanford as far as making use of Artificial intelligence to predict when a patient will die is one effort that should help in revolutionizing the healthcare sector especially in the area of palliative healthcare delivery. The fact that the outcomes are a product of the analysis and deductions from the data made available to the AI makes this development a somewhat laudable one because the algorithm brings about results that are about 95% accurate.

This should in no small way cause a change in the way doctors and caregivers handle people who are terminally ill or in critical health situations thus helping them when it comes to making the right decisions as regards their patients per time.

It will help reduce the number of people who die unexpectedly on their hospital beds while also cutting the need to extend the lives of patients in hospital beds using unpleasant methods.

This new technology also helps in creating new opportunities for the AI industry at the end of life care market because it has helped in stirring up interest in further research into this healthcare sector. It also catalyzes to encourage health startups to make inroads into this new and exciting technology while it is still fresh in the market.

Image credit:


Leave a reply

Your email address will not be published. Required fields are marked *


© 2018 Dr. Hempel Digital Health Network

Dr. Hempel Digital Health Network is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to


Log in with your credentials

Forgot your details?