Google has developed a model based on machine learning that can predict the likelihood of a patient’s death
A test was carried out on about 200,00 adult patients and about 46 billion variable parameters were used
The results were about 95% accurate
Yes, you read that right. Google can now tell how long a patient might be admitted or if there are chances of them getting readmitted again! Google has already taken a step into the world of healthcare with their research on cancer detection. If you haven’t come across it, you can read about it here.
There is an enormous amount of data being generated in the healthcare industry everyday, and where there is data, there is always room for machine learning! And that’s exactly what researchers at Google led by a team of scientists from Stanford, University of Chicago and UC San Francisco have utilized to come with an AI that predict the likelihood of a patient’s death with 95% accuracy.
How this works?
The AI looks at various variables from the patients health records and history like age, gender, previous diagnosis etc. it even manages to incorporate scribbled notes from the doctor and PDF’s into its final predictive model. The AI can then predict the probability of the patient’s death within 24 hours of them being admitted and it can do so with 95% accuracy.
According to a research paper published by Nature , Google tested its new AI on about 216,221 adult patients from two academic medical centers in the US. the number data points used were about 46 billion (46,864,536,945 to be precise). These results were impressively accurate. The researchers used the AUC-ROC statistics to to measure the accuracy of the model.
You can also have a look at how the AI performed against traditional methods here.
So let’s summarize:
Now the next step in this would be to apply this technology in different fields and try to improve and maybe even save lives. You never know what AI is capable of.
As AI makes its entrance into the world of healthcare it gives a huge scope for researchers and developers to come up with new ideas to help make human lives better and easier.