Nav: Home

Study examines use of deep machine learning for detection of diabetic retinopathy

November 29, 2016

In an evaluation of retinal photographs from adults with diabetes, an algorithm based on deep machine learning had high sensitivity and specificity for detecting referable diabetic retinopathy, according to a study published online by JAMA.

Among individuals with diabetes, the prevalence of diabetic retinopathy is approximately 29 percent in the United States. Most guidelines recommend annual screening for those with no retinopathy or mild diabetic retinopathy and repeat examination in 6 months for moderate diabetic retinopathy. Retinal photography with manual interpretation is a widely accepted screening tool for diabetic retinopathy.

Automated grading of diabetic retinopathy has potential benefits such as increasing efficiency and coverage of screening programs; reducing barriers to access; and improving patient outcomes by providing early detection and treatment. To maximize the clinical utility of automated grading, an algorithm to detect referable diabetic retinopathy is needed. Machine learning (a discipline within computer science that focuses on teaching machines to detect patterns in data) has been leveraged for a variety of classification tasks including automated classification of diabetic retinopathy. However, much of the work has focused on "feature-engineering," which involves computing explicit features specified by experts, resulting in algorithms designed to detect specific lesions or predicting the presence of any level of diabetic retinopathy. Deep learning is a machine learning technique that avoids such engineering and allows an algorithm to program itself by learning the most predictive features directly from the images given a large data set of labeled examples, removing the need to specify rules explicitly. Application of these methods to medical imaging requires further assessment and validation.

In this study, Lily Peng, M.D., Ph.D., of Google Inc., Mountain View, Calif., and colleagues applied deep learning to create an algorithm for automated detection of diabetic retinopathy and diabetic macular edema in retinal fundus (the interior lining of the eyeball, including the retina, optic disc, and the macula) photographs. A specific type of network optimized for image classification was trained using a data set of 128,175 retinal images, which were graded 3 to 7 times for diabetic retinopathy, diabetic macular edema, and image gradability by a panel of 54 U.S. licensed ophthalmologists and ophthalmology senior residents between May and December 2015. The resultant algorithm was validated using 2 separate data sets (EyePACS-1, Messidor-2), both graded by at least 7 U.S. board-certified ophthalmologists.

The EyePACS-1 data set consisted of 9,963 images from 4,997 patients (prevalence of referable diabetic retinopathy [RDR; defined as moderate and worse diabetic retinopathy, referable diabetic macular edema, or both], 8 percent of fully gradable images; the Messidor-2 data set had 1,748 images from 874 patients (prevalence of RDR, 15 percent of fully gradable images). Use of the algorithm achieved high sensitivities (97.5 percent [EyePACS-1] and 96 percent [Messidor-2]) and specificities (93 percent and 94 percent, respectively) for detecting referable diabetic retinopathy.

"These results demonstrate that deep neural networks can be trained, using large data sets and without having to specify lesion-based features, to identify diabetic retinopathy or diabetic macular edema in retinal fundus images with high sensitivity and high specificity. This automated system for the detection of diabetic retinopathy offers several advantages, including consistency of interpretation (because a machine will make the same prediction on a specific image every time), high sensitivity and specificity, and near instantaneous reporting of results," the authors write.

"Further research is necessary to determine the feasibility of applying this algorithm in the clinical setting and to determine whether use of the algorithm could lead to improved care and outcomes compared with current ophthalmologic assessment."
-end-
(doi:10.1001/jama.2016.17216; the study is available pre-embargo at the For the Media website)

Editor's Note: Please see the article for additional information, including other authors, author contributions and affiliations, financial disclosures, funding and support, etc.

The JAMA Network Journals

Related Learning Articles:

Learning with light: New system allows optical 'deep learning'
A team of researchers at MIT and elsewhere has come up with a new approach to complex computations, using light instead of electricity.
Mount Sinai study reveals how learning in the present shapes future learning
The prefrontal cortex shapes memory formation by modulating hippocampal encoding.
Better learning through zinc?
Zinc is a vital micronutrient involved in many cellular processes: For example, in learning and memory processes, it plays a role that is not yet understood.
Deep learning and stock trading
A study undertaken by researchers at the School of Business and Economics at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) has shown that computer programs that algorithms based on artificial intelligence are able to make profitable investment decisions.
Learning makes animals intelligent
The fact that animals can use tools, have self-control and certain expectations of life can be explained with the help of a new learning model for animal behavior.
Learning Morse code without trying
Researchers at the Georgia Institute of Technology have developed a system that teaches people Morse code within four hours using a series of vibrations felt near the ear.
The adolescent brain is adapted to learning
Teenagers are often portrayed as seeking immediate gratification, but new work suggests that their sensitivity to reward could be part of an evolutionary adaptation to learn from their environment.
The brain watched during language learning
Researchers from Nijmegen, the Netherlands, have for the first time captured images of the brain during the initial hours and days of learning a new language.
Learning in the absence of external feedback
Rewards act as external factors that influence and reinforce learning processes.
New learning procedure for neural networks
Neural networks learn to link temporally dispersed stimuli.

Related Learning Reading:

Make It Stick: The Science of Successful Learning
by Peter C. Brown (Author), Henry L. Roediger III (Author), Mark A. McDaniel (Author)

To most of us, learning something "the hard way" implies wasted time and effort. Good teaching, we believe, should be creatively tailored to the different learning styles of students and should use strategies that make learning easier. Make It Stick turns fashionable ideas like these on their head. Drawing on recent discoveries in cognitive psychology and other disciplines, the authors offer concrete techniques for becoming more productive learners.

Memory plays a central role in our ability to carry out complex cognitive tasks, such as applying knowledge to problems never... View Details


The Art of Learning: An Inner Journey to Optimal Performance
by Josh Waitzkin (Author)

In his riveting new book, The Art of Learning, Waitzkin tells his remarkable story of personal achievement and shares the principles of learning and performance that have propelled him to the top—twice.

Josh Waitzkin knows what it means to be at the top of his game. A public figure since winning his first National Chess Championship at the age of nine, Waitzkin was catapulted into a media whirlwind as a teenager when his father’s book Searching for Bobby Fischer was made into a major motion picture. After dominating the scholastic chess world for ten years, Waitzkin... View Details


Learning Web Design: A Beginner's Guide to HTML, CSS, JavaScript, and Web Graphics
by Jennifer Robbins (Author)

Do you want to build web pages but have no prior experience? This friendly guide is the perfect place to start. You’ll begin at square one, learning how the web and web pages work, and then steadily build from there. By the end of the book, you’ll have the skills to create a simple site with multicolumn pages that adapt for mobile devices.

Each chapter provides exercises to help you learn various techniques and short quizzes to make sure you understand key concepts.

This thoroughly revised edition is ideal for students and professionals of all backgrounds and skill levels.... View Details


The Science of Accelerated Learning: Advanced Strategies for Quicker Comprehensi
by Peter Hollins (Author)

Make learning: painless, exciting, habitual, and self-motivating. Absorb info like a human sponge. We’ve never been taught how to learn, and that’s a shame. This book is the key to reversing all the misconceptions you have and making learning fun again. Scientifically-proven, step-by-step methods for effective learning. The Science of Accelerated Learning is not a textbook - it’s a guidebook for your journeys in learning. It will show you the most effective methods, the pitfalls we must avoid, and the habits we must cultivate. This book is highly organized and addresses all phases of... View Details


How Learning Works: Seven Research-Based Principles for Smart Teaching
by Susan A. Ambrose (Author), Michael W. Bridges (Author), Michele DiPietro (Author), Marsha C. Lovett (Author), Marie K. Norman (Author), Richard E. Mayer (Foreword)

Praise for How Learning Works

"How Learning Works is the perfect title for this excellent book. Drawing upon new research in psychology, education, and cognitive science, the authors have demystified a complex topic into clear explanations of seven powerful learning principles. Full of great ideas and practical suggestions, all based on solid research evidence, this book is essential reading for instructors at all levels who wish to improve their students' learning."
Barbara Gross Davis, assistant vice chancellor for educational development, University... View Details


10 Mindframes for Visible Learning: Teaching for Success
by John Hattie (Author), Klaus Zierer (Author)

The original Visible Learning research concluded that one of the most important influencers of student achievement is how teachers think about learning and their own role. In Ten Mindframes for Visible Learning, John Hattie and Klaus Zierer define the ten behaviors or mindframes that teachers need to adopt in order to maximize student success. These include:

thinking of and evaluating your impact on students’ learning;

the importance of assessment and feedback for teachers;

working collaboratively and the sense of community;

the... View Details


The Fifth Discipline: The Art & Practice of The Learning Organization
by Peter M. Senge (Author)

Completely Updated and Revised

This revised edition of Peter Senge’s bestselling classic, The Fifth Discipline, is based on fifteen years of experience in putting the book’s ideas into practice. As Senge makes clear, in the long run the only sustainable competitive advantage is your organization’s ability to learn faster than the competition. The leadership stories in the book demonstrate the many ways that the core ideas in The Fifth Discipline, many of which seemed radical when first published in 1990, have become deeply integrated into people’s ways of... View Details


Unlimited Memory: How to Use Advanced Learning Strategies to Learn Faster, Remember More and be More Productive
by Kevin Horsley (Author)

Kevin Horsley Broke a World Memory Record in 2013...

And You're About to Learn How to Use His Memory Strategies to Learn Faster, Be More Productive and Achieve More Success

Most people never tap into 10% of their potential for memory. In this book, you're about to learn:

How the World's Top Memory Experts Concentrate and Remember Any Information at Will, and How You Can Too

Do you ever feel like you're too busy, too stressed or just too distracted to concentrate and get work done? In Unlimited Memory, you'll learn how the world's best memory masters... View Details


Learning Radiology: Recognizing the Basics, 3e
by William Herring MD FACR (Author)

A must-have for anyone who will be required to read and interpret common radiologic images, Learning Radiology: Recognizing the Basics is an image-filled, practical, and easy-to-read introduction to key imaging modalities. Skilled radiology teacher William Herring, MD, masterfully covers exactly what you need to know to effectively interpret medical images of all modalities. Learn the latest on ultrasound, MRI, CT, patient safety, dose reduction, radiation protection, and more, in a time-friendly format with brief, bulleted text and... View Details


Empower: What Happens When Students Own Their Learning
by John Spencer (Author), A.J. Juliani (Author)

Kids begin their learning journey as curious problem solvers who ask questions and create solutions. As they go through school, something happens to many of our students, and they begin to play the game of school, eager to be compliant and follow a path instead of making their own. 

As teachers, leaders, and parents, we have the opportunity to be the guide in our kids' education and unleash the creative potential of each and every student. In a world that is ever changing, our job is not to prepare students for something; instead, our role is to help students prepare themselves... View Details

Best Science Podcasts 2018

We have hand picked the best science podcasts for 2018. Sit back and enjoy new science podcasts updated daily from your favorite science news services and scientists.
Now Playing: TED Radio Hour

Inspire To Action
What motivates us to take up a cause, follow a leader, or create change? This hour, TED speakers explore stories of inspirational leadership, and what makes some movements more successful than others. Guests include high school history teacher Diane Wolk-Rogers, writer and behavioral researcher Simon Sinek, 2016 Icelandic presidential candidate Halla Tómasdóttir, professor of leadership Jochen Menges, and writer and activist Naomi Klein.
Now Playing: Science for the People

#474 Appearance Matters
This week we talk about appearance, bodies, and body image. Why does what we look like affect our headspace so much? And how do we even begin to research a topic as personal and subjective as body image? To try and find out, we speak with some of the researchers at the Centre for Appearance Research (CAR) at the University of the West of England in Bristol. Psychology Professor Phillippa Diedrichs walks us through body image research, what we know so far, and how we know what we know. Professor of Appearance and Health Psychology Diana Harcourt talks about visible...