Google AI can predict when a patient is going to die. Doctors of any hospital can’t exactly predict when a patient is going to die. But here, Google has claimed that its Medical Brain team has developed a latest algorithm which can predict the death chances of patients in hospitals.
According to the early results, accuracy rate of Google’s AI is more than hospital’s own death predicting systems. Well, that’s pretty impressive, isn’t?
Artificial Intelligence uses the electronic health records and other patient’s information to get the chances of death, discharge, and readmission of the patient.
The new tool created by Google’s medical team have capacity to forecast different patient’s results, such as how long patients may stay in hospitals, how many chances of their re-admission and even can notify the risks that they will soon die.
According to Google’s research which is published in Nature, it describes this algorithm:
“These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios”.
In order to try the tool, Google applied its algorithm tools to a patient having metastatic breast cancer. After 24 hours of being admitted to a hospital, Google gave her a 19.9% chance of dying in the hospital, on contrary with the 9.3 estimate with the hospital’s Early Warning Score.
As a result, the patient died in a hospital from her condition as predicted by Google’s AI toolin less than 2 weeks later.According to a case study published in Nature.
Not to know, it’s not the first time Google’s algorithm tools have been applied to predict and announce the results of patients in healthcare. As earlier in May, Google’s AIresearch department made an augmented reality microscope for cancer detection. Thisspecial AR microscopes (ARM) employ machine learning to detect cancerous cells.
Jeff Dean, AI head said;
“Google’s augmented reality microscope (ARM) combines both methods, it blend[s] the expertise of automated machine learning systems with human expertise.”
There is no doubt that this new invention of AI tool by Google’s team is great development in medical industry. However, if Google can smooth the procedure of entering data and increase the means by which that data is used then chances of human error in medical care could be reduced.
