23/07/2018

Artificial Intelligence, our new best friend

Inteligencia Artificial, nuestra nueva mejor amiga - Investigación, Sociedad

When you think about artificial intelligence, the performance of Arnold Schwarzenegger as the cyborg killer in the movie The Terminator (1984) could come to mind. However, today, artificial intelligence or AI, has jumped from the big screen to the street. One of its functions is to help diagnose, control and treat patients or assist in domestic assistance.

Agents interested in Artificial Intelligence technology in the field of medicine and health are trying to take advantage of the existing opportunity in the market through different mergers and acquisitions of several small companies. The most famous names today are Intel Corporation, IBM, Koninklijke Philips N.V, Siemens Healthineers GmbH, Google, BabyLabs Inc., Oncora Medical Inc., Johnson & Johnson, Stryker Corporation and Microsoft Corporation. However, many wonder, what exactly is artificial intelligence used in the sanitary branch?

AI is making a difference

It is not difficult to find examples of how Artificial Intelligence is already improving the results of medical care. At the New York Presbyterian Hospital, for example, a new AI command center helps nurses in a remote control room monitor patients and processes with artificial intelligence tools to reduce fatigue and free doctors to be able to spend more time with patients.

Intel, on the other hand, worked with Sharp HealthCare using AI to identify patients who were at risk of sudden worsening, which could require a rapid response team. Analytical work has demonstrated predictability with 80% accuracy, allowing rapid response teams to be optimized and positioned proactively at key points in hospitals, and even intervening before the situation becomes more dangerous to life of the patient.

At the World Forum of Medical Innovation 2018 (WMIF) on artificial intelligence presented by Partners Healthcare, a group of researchers and clinical teachers highlighted some of the technologies and areas of the health industry that will probably have a greater impact of artificial intelligence in the next decade.

Unify the mind and the machine through the interfaces of the brain-computer

Using computers to communicate is not a new idea, but creating direct interfaces between technology and the human mind without the need for keyboards, mice and monitors is a cutting-edge research area that has significant applications for many patients.

Neurological diseases and traumas in the nervous system can affect the ability of some patients to talk, move and interact meaningfully with people and their environments. Brain-computer interfaces (BCI) supported by artificial intelligence could restore those fundamental capabilities to those who feared they were lost forever.

These interfaces can dramatically improve the quality of life of patients with amyotrophic lateral sclerosis or stroke as well as the 500,000 people worldwide who experience spinal cord injuries each year.

Provide access to care in regions without development or in development

The shortage of trained health service providers, such as ultrasound technologists and radiologists, can significantly limit access to health care in developing countries around the world. Artificial intelligence could help mitigate the impacts of this serious shortage of qualified clinical personnel by taking over some of the diagnostic duties typically assigned to humans.

For example, AI imaging tools can detect signs of tuberculosis on chest radiographs, often reaching a level of accuracy comparable to that of humans. This capacity could be implemented through an application available to providers in low-resource areas, reducing the need for an on-site diagnostic radiologist.

"The potential of this technology to increase access to medical care is tremendous," said Jayashree Kalpathy-Cramer, assistant neuroscience at Massachusetts General Hospital and associate professor of Radiology at HMS.

Mitigate the risks of antibiotic resistance

Resistance to antibiotics is a growing threat to populations all over the world, since their excessive use encourages the evolution of superbugs that no longer respond to treatments. Multidrug-resistant organisms can wreak havoc in the hospital environment, claiming thousands of lives each year.

Electronic health records can help identify patterns of infection and diagnose at-risk patients before they begin to show symptoms. Leveraging machine learning and artificial intelligence tools to drive these analyzes can improve their accuracy and create faster and more accurate alerts for healthcare providers.

"AI tools can meet the expectations of infection control and antibiotic resistance," said Erica Shenoy, associate head of the Infection Control Unit at Massachusetts General Hospital.

Health monitoring through wearables and personal devices

Almost all consumers have access to devices with sensors that can collect valuable data about their health. From smartphones with step trackers to portable devices that can track a heartbeat throughout the day, an increasing proportion of health-related data is generated on the fly.

Collecting and analyzing these data and complementing them with information provided by the patient through applications and other household monitoring devices can offer a unique perspective on individual and population health.

Turn the smartphone into an authentic diagnostic tool

Continuing the theme of harnessing the power of portable devices, experts believe that images taken from smartphones and other similar sources will be an important complement to clinical-quality images, especially in marginalized populations or developing nations.

The quality of mobile phone cameras increases every year and can produce images that are viable for analysis using artificial intelligence algorithms. Dermatology and ophthalmology are the first beneficiaries of this trend.

Researchers in the United Kingdom have developed a tool that identifies developmental diseases by analyzing images of a child's face. The algorithm can detect discrete features, such as the line of a child's jaw, eye and nose placement, and other attributes that may indicate a craniofacial abnormality. Currently, the tool can match ordinary images with more than 90 disorders to provide support in clinical decision making.

"The majority of the population is equipped with powerful pocket devices that have many integrated sensors," said Hadi Shafiee, director of the Micro / Nanomedicine and Digital Health Laboratory at Brigham Hospital.

"This is a great opportunity for us." Almost all the major players in the industry have started to develop software and AI hardware on their devices. That is not a coincidence. Every day in our digital world, we generate more than 2.5 million terabytes of data. On mobile phones, manufacturers believe they can use that information with AI to provide much more personalized and faster and smarter services. "

The use of smart phones to collect images of eyes, skin lesions, wounds, infections, medications or other issues can help unattended areas cope with the shortage of specialists, while reducing the time of diagnosis for certain conditions.

And now, do you still think that AI is just a science fiction thing?