Google is developing a system of artificial intelligence to predict health risks and the death of patients

Source: gate Arab news technical

جوجل تطور نظام ذكاء اصطناعي للتنبؤ بالمخاطر الصحية وميعاد وفاة المرضى

Working research team at Google called the Medical Brain Team on the development of the algorithm of artificial intelligence that can predict health and the risk of death among hospital patients show their results early it’s higher accuracy of the warning system, private hospitals, and this confirms the progress made by the research team that will help the company to progress significantly in the field of health care.

According to Bloomberg, the potential of the health care neural networks at Google-which is a form of software artificial intelligence is particularly beneficial in the use of data for development automatically – was able to create a tool that can predict a range of patient outcomes, including duration of survival of persons in the hospitals, and their in treatment, the risk of death.

The team conducted a research study on the situation of women with breast cancer at a late stage, where the flooded fluid in her lungs already. Checked by two doctors and to examine the X-rays and then read the computers in the hospital her vital signs and, accordingly, the capacity risk of death increased by 9.3%.

While it took Google algorithm-which examined 175,639 data points from electronic medical records your patient has provided its assessment of the risk of death increased by 19.9%. A hotel completely different from the assessment of early warning in the hospital (aEWR), which was a rate of 9.3%.

The team published Google on May 8 of this year his results in a paper entitled “deep learning scalable accuracy with electronic health records”, where he said about the algorithm to predict: “I miss this algorithm on predictive models traditionally used in all cases. We believe that this approach can be used to create accurate predictions and scalable to a variety of pathological conditions”.

I was more impressed experts The is the ability of Google to examine the data that was previously out of reach of users, such as: the doctor’s notes recorded in the files PDF, in addition to the handwritten notes also. Where they were feeding the neural network with this information and build on the conclusion of the expectations, and did so faster and more accurately than current techniques. As showed Google’s system of records that led him to the conclusions.

Try hospitals, Doctors and health care providers and others over the years to improve the use of stocks of electronic health records and other patient data, where can for more information that is shared and highlighted in time to save lives -or at least help the medical personnel to spend less time on paperwork and more time taking care of patients – but the current road address data health care is expensive and cumbersome and time-consuming.

The next step from Google is to transfer this system is predictive to the clinic, where he said Jeff Dean, head of Department of artificial intelligence at Google to the newspaper Bloomberg: “the work of a Research Unit, which is referred to as the brain medical Medical Brain– a large collection of tools of artificial intelligence that can predict the symptoms and diseases high level of accuracy”.

Within the company there is a lot of excitement around this initiative. Where said one of the Google staff: “I come finally to the application of the new system of artificial intelligence has promised a commercial”. Ever since Google announced the subsidiary of the. As a company artificial intelligence first “AI-first” in 2016, much of its work in this area has contributed to the improvement of the current services provided by Google. Granted the developments made by the team of brain medical Google Medical Brain Team the opportunity to break into a new market altogether – and that’s what he tried to do his company’s founders Larry Page and Sergey Brin over and over again.

Think software free health care to write the code manually these days. In contrast, the Google approach, which depends on the learning machines to determine the data on its own can agree on nothing else, where he said You bajaj former CEO of Farley, one of the arms of the health care to buy that, and the appropriate Investment Company Forrest my capital: “they are aware of the problems that deserve solutions. They now experiences small enough to know exactly trends and fruitful”.

For optimism about the prospects for Google, it remains to harness the techniques of artificial intelligence to improve healthcare outcomes is a major challenge. Where there were attempts to other systems such as the system of “Watson”, a subsidiary of IBM, IBM Watson, who tried to apply artificial intelligence techniques in the field of medicine, but his success was limited in relation to the provision of water and integrate the technology into payment systems.

I’ve always sought Google to access to medical records digital. For the consideration of the latter company the internet giant deals with the University of California San Francisco University of Chicago compared to 46 million pieces of patient data anonymous. Said system of artificial intelligence of Google to establish predictive models for each hospital. As Google works to secure new partners to gain access to more medical records.

Overcoming the deepest in the press is only going to add huge amounts of information possessed by Google already. Andrew Burt, senior privacy buy Immuta: “I’ll have companies like Google and other giants of the technical ability of the unique -and even proprietary almost – to take advantage of all the data that we produce”. He has written a Bert in conjunction with oncologists in the children of Samuel point an article recently saying: “governments must prevent this data from becoming limited to a small number of companies, as is the case in the online ads are controlled by Google”.

Skip Google with caution when it comes to information of the patient, especially with the increased public scrutiny on the data collected during the recent period. Last year the British set lab deb Maine DeepMind -one of the laboratories of artificial intelligence, a subsidiary of the – When opting for an application that analyzed medical records public without telling the patients that their information will be used as such. However, the latest study insists Google and its partners in the hospitals that their data is anonymous and safe and used with the permission of the patient. Said point the company may face greater difficulty in maintaining the accuracy of the data if expanded to hospitals to smaller networks of health care.

However, the point that these algorithms can save lives and save money. He hoped to choose health records with a sea of other stats. In the end, can include models of artificial intelligence information on local weather and traffic -the other factors that affect patient outcomes – “it’s almost like that the hospital will become a living organism”.

Google in collaboration with the company Verily to develop tracking devices signals more energetic. Even if not used patients devices and health Tracker wearable, The have Google a lot of other data to take advantage of them. Where Android phones from Google many things like how people and valuable information to measure the decline of mental and other diseases. All these in I can add algorithms to the medical.

Medical records are considered only part of the plans of the health care systems of artificial intelligence at Google.Where a team of medical research of the development of systems of artificial intelligence, Radiology, Ophthalmology, cardiology and dermatology also, as well as creating team work applied to skin diseases and malignancies.

Google is also doing a new experience in India using the program its artificial intelligence to examine the image of the eyes in search of the early signs of a condition called diabetic retinopathy.

Over time, you can Google license these systems to clinics or sold as a service through the cloud computing company to help in the processes of medical diagnosis. As Microsoft and its biggest rival Google is also developing services artificial intelligence is predictive, so you’ll need Google in the beginning to get more records, which tend to vary significantly across health service providers. Can you buy Google, so that pages that had been held with the University of California at San Francisco UCSF University of Chicago is not a business.

In the meantime the company says it’s premature to be based on the business model. In its annual conference for developers Google I/O held in the month of May, Lily Peng, a member of the team of Medical Brain presentation research group, which outperformed the humans in the discovery of the risk of heart disease. She said again “I want to emphasize that this is the time of early independence on a specific model to work”.

Link to it from the source: Google’s system development artificial intelligence to predict health risks and the death of patients

Leave a Reply

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