Bold Business Logo

Google AI Can Read Your Eyes to Predict Heart Attacks

Google, one of the world’s biggest multinational technology companies, revealed an artificial intelligence (AI) tech that claims to predict heart attacks. Their new AI algorithm can evaluate retinal scans, predicting a person’s age and blood pressure, and then analyzes if they have a risk of suffering a heart attack. Their non-invasive device is bound to disrupt the medical industry and create a bold impact in the health and lives of many people.

Google researchers revealed the AI algorithms “look” at a human eye to determine whether the person has high blood pressure, and if they are at risk of a stroke or heart attack. The global health industry is a very profitable one, despite its vastness, and Google is always on the move in creating and managing various health solutions. For example, they created startup health company Calico, as well as their recent bold idea of helping people with depression. They also created the anti-mosquito Debug Project under Verily, as well as acquired Senosis Health a few months ago. Clearly, Google is not afraid of taking a risk on the health industry.

Screening for Heart Health Risks

Google’s AI algorithms are impressive, but as with any new medical tech, there is the need for validation. A study published in the Nature Biomedical Engineering journal revealed that the algorithms did not outperform already existing medical approaches (e.g. blood tests) that find out the same information. Several outside physicians expressed concern that the work needs to be repeated on and validated before people give the predictive AI broader acceptance.

Lead researcher Lily Peng admitted the study is still quite early and that it was a small data set the AI algorithm based itself on. “We think that the accuracy of this prediction will go up a little bit more as we kind of get more comprehensive data,” she explained. “Discovering that we could do this is a good first step. But we need to validate.”

In the published study, the researchers concluded, “The opportunity to one day readily understand the health of a patient’s blood vessels, key to cardiovascular health, with a simple retinal image could lower the barrier to engage in critical conversations on preventive measures to protect against a cardiovascular event.”

“Diagnosis is about to get turbo-charged by technology,” said Yale cardiologist and health care researcher Harlan Krumholz, MD. Although not involved in the aforementioned study, he believes the Google device is a quick way to screen for health risks.

At present, medical professionals can manually check for risk factors, such as their age, weight, gender, and whether or not they smoke. In addition, they can use devices to draw blood or inspect a person’s retina. With Google’s device, however, it takes the guesswork out of it through simply scanning the retina. In their study, Google researchers scanned the retinas of 284,335 patients from both the United States and the United Kingdom. They then fed the scanned images into their AI to detect and understand patterns.

These self-learning algorithms, called artificial neural networks (ANNs), recognize patterns in information, whether it is through text data, voice/speech, and in visual images. As opposed to conventional AI that requires a human to “teach” the algorithm, ANNs learn on their own. In this case, ANNs learn and understand which “signs” from visual images indicate possible long-term health dangers. By reviewing enough data, the AI can associate patterns often found in people at risk.

All in the Eyes

It may seem outlandish for some people how a person’s eyes can reveal signs of a possible underlying disease. However, the eyes really do indicate these because of changes in the retina. In fact, even diabetes and high blood are just a few of the diseases the eye can reveal. It may seem unusual, but the method is actually an established methodology.

The eyes have a back interior wall, called the fundus, and it contains many blood vessels that reflect a person’s overall health. Through studying and analyzing their appearance, doctors can determine many things. Diseases such as arthritis, Alzheimer’s Disease, diabetic retinopathy that leads to blindness, and various brain tumors can be spotted in eye tests. Even certain dark spots at the back of an eye may indicate a melanoma, which may sometimes lead to cancer. Truly, the eyes can reveal plenty about a person’s health if a medical professional or even an AI algorithm knows what to look for.

In the study, Google’s AI was tested with old retinal images of two patients at a time – always with one who suffered a heart event in the following five years while the other person did not. Google’s innovative AI was 70% accurate – a rate that is similar to the 72% accuracy of the SCORE method, which required a person to give a blood sample.

Google is not the only tech giant investing in innovative health solutions in recent years. Much like them, Microsoft also had several people working on cancer solutions, while Apple has several in the works as well including a medical study involving their Apple Watch’s heart rate monitor.

It is the first time Apple sponsored a medical study themselves, despite several others already using their devices and software in separate independent medical studies. Their chief operating office (COO) Jeff Williams said that their involvement will help people regarding certain conditions, as well as to help “advance discoveries in heart science.”

Apple is running the study, submitting their gathered data to the Food and Drug Administration, wherein the heart rate researcher will concentrate on atrial fibrillation (afib) – irregular heart rates. Afib is one of the leading causes of stroke and other heart conditions. According to estimates by the Centers for Disease Control and Prevention, afib takes the lives of around 130,000 annually. Apple admits the study is not perfect yet, lacking the precision of clinical diagnostic tools, but it is promising nonetheless.

Every day, experts and innovators come up with bold ideas in the name of a healthier population. With big names like Google, Apple, and Microsoft investing in health and medicine, the industry will have nothing else but improvement in its future.

 

The Future of Mobility and Transport in Urban Environments

Urban transportation has gone a long way since its beginnings dating back to as early as the 17th century. Today, there many more options including airlines, buses, trains, ferries, cable transits, and of course, taxis and peer-to-peer ride sharing. Cities, especially bigger ones, are open to a number of bold innovations – ones that shape and design the future of transportation and mobility.

According to a 2014 report by the United Nations (UN) titled “World’s population increasingly urban with more than half living in urban areas,” about 54% of the earth’s population belong to urban areas – a number that experts believe will increase to 66% by 2050.

Urban Traffic and Transportation

This massive flocking towards cities not only points to an increase in the population – it also points to an onslaught of transportation issues.

City dwellers all over the United States and elsewhere around the globe experience congested roads, unreliable public transportation, and inefficient traffic patterns – things that if not addressed would lead to increased travel times and monetary loss not only for the people but for countless businesses as well. In fact, the economic costs of things like time lost in traffic and wasted fuel take 2-4% from a city’s gross domestic product (GDP). It also adds to pollution – air and noise included.

As experts identify these issues connected to older transportation systems, in come various solutions that form what urban transportation is today and what else it can be tomorrow. People in the urban environment are more than ready for dramatic change, things that go beyond the disruption caused by ride sharing giants like Uber and Lyft.

The Future of Mobility Today

“Aging transportation systems are failing and solutions are needed to maintain people’s freedom to move,” said Stephen Dyer, Vice President of Business Strategy at Ford Asia Pacific. “These solutions cannot hinge on ideas that are still several decades away from broader adoption, like flying cars.”

 

There are cities all over the world already making moves to address the failing transportation solutions of yesteryear. For example, Singapore has interconnected bus and train systems, while Barcelona, Spain has about over 300 miles of fiber optic cables connecting an ongoing Internet of Things (IoT) solution across the city. Major automotive company Ford has collaborated with cities all over the US and worldwide, including Detroit, New York, and Seattle as well as parts of China and India. Cities like Jacksonville, Florida and Atlanta, Georgia have also taken steps in improving their transportation as well. These are just part of the tip of the iceberg – countless other efforts are ongoing all over the country and around the world.

A collaborative report by Bloomberg New Energy Finance and McKinsey, titled “An integrated perspective on the future of mobility,” explores how technology can work together with socioeconomic trends. In it, they revealed three models of advanced urban mobility – Clean and Shared, Private Autonomy, and Seamless Mobility – all achievable by the year 2030. They realize each city is different, so factors that may work for one might not work for the other. However, there are many common factors, including privately owned vehicles, public transit, greener options like walking and bicycling, and exploring various new mobility services like e-hailing – including ride sharing and autonomous vehicles. In addition, updating policies and regulations as well as updating urban design to fit these changes are of utmost importance.

Improving Urban Transport

In the same Bloomberg/McKinsey report, they proposed several ways to address issues and improve commercial transport. Some of these potential solutions already exist, but could still have a few improvements or need wider deployment.

  1. Urban consolidation centers (UCCS) – Ideally, these centers should be in the cities’ outskirts, where deliveries are bought, organized, and then dispatched. UCCs already exist, and reduce as much as 45% of delivery-related mileage, and cut vehicle maintenance costs. They even reduce nitrogen oxide (NOx), carbon dioxide (CO2), and particulate emissions.
  2. Autonomous ground vehicles (AGVs or AVs) – Driverless cars need little to no human intervention. As with UCCs, these AVs also cut emissions and ease congestion, reducing as much as 50% of the cost per parcel. While already existing, they will most likely become more available by the 2020s.
  3. Night deliveries – In dense and developed cities, night deliveries could cut delivery time in half, delivery costs by 40%, and vehicle emissions by 70%. It even allows suppliers to use bigger trucks – allowing less travel time with more cargo.
  4. Load pooling – The idea of an online platform that matches customers with truckers who have extra space can reduce delivery-related mileage by up to 30%. This “Uber for cargo” can cut delivery costs by as much as 25% while also reducing vehicle emissions by 30%.
  5. Electric vehicles – As with UCCs and AVs, the use of electric vehicles also significantly reduce emissions. In addition, they are much quieter compared to regular cars. As such, it could be the push needed for the public to accept the idea of night deliveries.

These potential solutions, along with countless bold ideas from a number of innovators, contribute to a significant advantage for the cities of tomorrow. This disruption can have many positive effects including reduced pollution and even lessen traffic-related deaths. Having these changes can improve the quality of life for billions of people in cities and urban environments around the world.

How can we help?