Alphabet studies how deep learning could predict heart disease risk | Healthcare Dive

Researchers at Google parent Alphabet’s research arm used an algorithm to speed assessments of patient cardiovascular risk factors using eye scans.

  • Researchers at Alphabet and its research arm Verily Life Sciences have found a way to predict a person’s cardiovascular risk factors using eye scans and deep learning, according to a study published this week in Nature Biomedical Engineering.
  • By analyzing scans of the retinal fundus, the tissue at the back of a patient’s eye, the company’s software can tease out data such as blood pressure, age and whether an individual smokes — all potential risk factors for having a major cardiac event.
  • The algorithm could speed assessments of patients’ cardiovascular risk, but more testing is required before it can be used in a clinical setting.

Dive Insight:

“Using deep learning algorithms trained on data from 284,335 patients, we were able to predict CV risk factors from retinal images with surprisingly high accuracy for patients from two independent data sets of 12,026 and 999 patients,” Dr. Lily Peng, product manager of the Google Brain Team and a co-author of the study, wrote in the Google Research Blog.

“For example, our algorithm could distinguish the retinal images of a smoker from that of a non-smoker 71% of the time, compared to a ~50% (i.e. random) accuracy by human experts,” she said.

The results also show “strong gender differences” in the fundus images that could help guide research on the differences in male and female eyes, as well as how cardiovascular disease or risk factors affect retinal health, according to the study.

Last year, Verily began recruiting 10,000 volunteers to help build a comprehensive database of biometric data. The company is also working with French drugmaker Sanofi to develop tools for diabetes management and with 3M on solutions for population health.

This latest study could have implications for treating patients. Cardiology has been trending toward population health and tools such as this new algorithm could help identify and inform care paths for patients.

Recommended Reading:

Source: Alphabet studies how deep learning could predict heart disease risk | Healthcare Dive

Benefits and Possible Dangers or side effects of Artificial Intelligence to Mankind

botThe rise of the artificial intelligence brings technological breakthrough benefits as well and a negative drawbacks and possibly dangerous side effects to how mankind lives our lives today and how we might exist and operate in the world of tomorrow.

Technology advances have driven the world to continually adapt or change more and more, this has changed the way humans live our lives. The example of the simplest communication tools such as the fixed line telephone has now become fully mobile computing device that also doubles as a telephone that we can use anywhere. (Rodriguez 2011)

Artificial Intelligence or (AI) can be described as a field of science making computing devices do activities that require understanding and intelligence when otherwise actioned by a human.  (Copeland 2000) There are many different branches of AI some physical such as robotic, some representing logic or programming tasks. (McCarthy 2007) Most however are trying to mimic or create human traits in computing systems such as knowledge, problem solving, learning, and planning. (Brookshear, Smith and Brylow 2012)

Some benefits of Artificial Intelligent Devices:

  • Robotic devices with intelligence are able to go and execute adaptive activities in places where it might not be safe for human beings to go. Such locations could be deep space, deep sea, military warzones, or contaminated areas. The Defense Advanced Research Projects Agency (DARPA 2014) are currently running a robotic challenge to find devices to go in to such areas as disaster zones or military war zones as support platform. (DARPA 2014)
  • AI devices in theory don’t need to sleep or take a lunch and coffee break they can work until the activity is completed. (There are acceptations to this for maintenance, recharge if battery operated) Hyundai Motors has reduced delivery times on vehicle production by 20% by using robots. ( 2014)
  • There is an ability to mimic or reproduce copies of AI devices to do the same tasks which is not possible with humans without cost. When devices have become efficient and correct at executing a task the device can be duplicated or copied, therefore in the human sense forgoing the need of training & development and loss when the employee leaves which is equal to higher cost.
  • AI devices work on logic and are not prone to human errors or misunderstanding this would be good in tasks such as operating a vehicle.  Such jobs as pilotless cargo aircraft that travel from point A to point B and could run 24 hours a day. (Platform Unmanned Cargo Aircraft 2014) (FAA 2014)

Some possible drawbacks or dangers of Artificial Intelligent Devices:

  • Ethical dangers or drawbacks from trying to create AI devices into human like devices. If humans give intelligence and understanding to an artificial device then will the machine want to work for human benefit and can the device be switched off if needed. Who gets the right to choose to switch off the device and what happens if the device becomes more intelligent than humans by creating new algorithms or adapting it’s programing through learning. (McCarthy 2007) (Barber 2014)
  • An effect of the continued replacement of activities in companies or industries by some level of AI machine or device could be on the need for a work force. AI devices taking over jobs could create unemployment for both skilled and unskilled workers as companies want more efficiency and great profits.   (Turner 2013)
  • Lack of emotional understand or judgement leaves the AI machinery not able judge all the possible outcomes as a human might be able for such jobs as, nursing, security or military devices. The reason being the AI device answers are based on logic questions and not emotional values.  (Berkeley 1997)

Summary / Conclusions

There are many reasons to use and look at artificial intelligent devices or machines as well as there are many reason we should not. These reasons however mainly focus and play on the fundamental human conscious and ethical morals of most of us. Over the coming years governments, companies and people will all try to debate the pro’s suggesting life will be better and the con’s that humans are worried about technical armies and the thought the world could be taken over by AI devices.

There is a thought in place at the moment which is we are quite a way already on this journey. The car industry in point has already reduced work forces globally replacing them with robotic devices on efficiency based reasons / gains and selling it’s to us the consumer that it’s better us based on quality, reliability and cost, but to the workers they lost their job possibilities. In the early days on this change there was uproar but today it’s taken for granted but do we really understand the long term ethical or economical global because most companies and governments think for current business or political agendas one thing for certain right now there are people for and against artificial intelligence debate

Please complete the Poll below on your thoughts:

by Charles Aunger 05/06/2014



BARBER, Nigel (2014). Must We Worry About Artificial Intelligence? [online]. Last accessed 23 05 2014 at:

BERKELEY, István S. N. (1997). What is Artificial Intelligence? [online]. Last accessed 23 05 2014 at:

BROOKSHEAR, J Gleen, SMITH, David T and BRYLOW, Dennis (2012). Chapter 11: Artificial Intelligence. In: HORTON, Marcia (ed.). Computer Science – An Overview. 11th ed., Addison-Wesley – Pearson, 462-500.

CBS (2014). “Killer robots” could threaten basic human rights, activists warn. [online]. Last accessed 23 05 2014 at:

CNET (2004). Smart systems will erase jobs, report warns. [online]. Last accessed 23 05 2014 at:,-report-warns/2100-1022_3-5247644.html

COPELAND, Jack (2000). What is Artificial Intelligence? [online]. Last accessed 23 05 2014 at:

DARPA (2014). About DARPA. [online]. Last accessed 23 05 2014 at:

DARPA (2014). Darpa Robotics Challenge. [online]. Last accessed 23 05 2014 at:

HEALTHRESEARCHFUNDING.ORG (2014). Pros and Cons of Artificial Intelligence. [online]. Last accessed 23 05 2014 at:

HOWELL, Elizabeth (2014). Future of US Warfare: Drone Deliveries & Vertical-Flight Planes. [online]. Last accessed 23 05 2014 at:

LINKS999 (2007). Ethical and moral issues regarding AI. [online]. Last accessed 23 05 2014 at:

MCCARTHY, John (2007). Basic Questions on Artificial Intelligence. [online]. Last accessed 23 05 2014 at:

MCCARTHY, John (2007). Branches of AI. [online]. Last accessed 23 05 2014 at:

PLATFORM UNMANNED CARGO AIRCRAFT (2014). Civil unmanned cargo aircraft. [online]. Last accessed 23 05 2014 at:

RODRIGUEZ, Armando (2011). Then and now: How technology has changed our lives. [online]. Last accessed 23 05 2014 at:

TECHOPEDIA (2014). Artificial Intelligence (AI). [online]. Last accessed 23 05 2014 at:

THE ILOVEINDIA (204). ros And Cons Of Artificial Intelligence. [online]. Last accessed 23 05 2014 at:

TURNER, Bambi (2013). ROBOTICS AND ARTIFICIAL INTELLIGENCE. [online]. Last accessed 23 05 2014 at: