DeepMind Masters Retinal Disease Detection
By MedImaging International staff writers Posted on 27 Aug 2018 |
Image: Google DeepMind is exploring new frontiers in medical diagnostics (Photo courtesy of Google).
Artificial intelligence (AI) can accurately detect 53 kinds of sight-threatening retinopathy at least as effectively as experts, claims a new study.
Researchers at Google DeepMind (London, United Kingdom), Moorfields Eye Hospital (London, United Kingdom), and University College London (UCL, United Kingdom), first trained a deep learning AI algorithm to spot ten different features of eye disease, based on 14,884 three-dimensional (3D), high-definition optical coherence tomography (OCT) retinal scans. The researchers then gathered another 997 scans and asked DeepMind and eight consultant ophthalmologists and specialist optometrists to recommend urgent referral, semi-urgent referral, routine referral, or observation for each scan.
In the most crucial category, urgent referral, DeepMind matched the two top retina specialists and had a significantly higher performance rate than the other two specialists and all four optometrists. No cases of urgent referral were missed by the system. When all referral types were taken into consideration, computer error rate was 5.5%, comparable to the error rate for the two best retina specialists (6.7% and 6.8%) and significantly better than the other six experts. When the clinicians had access to the OCT scans, fundus images, and patient summary notes, five had similar accuracy to the AI system, whereas the system still outperformed the other three. The study was published on August 13, 2018, in Nature Medicine.
“We set up DeepMind because we wanted to use AI to help solve some of society's biggest challenges, and diabetic retinopathy is the fastest growing cause of blindness worldwide. There are more than 350 million sufferers across the planet,” said co-senior study author Mustafa Suleyman, co-founder of DeepMind. “Detecting eye diseases as early as possible gives patients the best possible chance of getting the right treatments. I really believe that one day this work will be a great benefit to patients across the NHS.”
“Our research with DeepMind has the potential to revolutionize the way professionals carry out eye tests, and could lead to earlier detection and treatment of common eye diseases such as age-related macular degeneration,” said Professor Sir Peng Tee Khaw, MD, director of the Biomedical Research Centre in Ophthalmology at Moorfields Eye Hospital. “With sight loss predicted to double by the year 2050, it is vital we explore the use of cutting-edge technology to prevent eye disease.”
DeepMind is a British artificial intelligence company founded in September 2010 which created a neural network that learns how to play video games in a fashion similar to that of humans, as well as a neural network that may be able to access an external memory like a conventional Turing machine, resulting in a computer that mimics the short-term memory of the human brain; it was acquired by Google in 2014.
Related Links:
Google DeepMind
Moorfields Eye Hospital
University College London
Researchers at Google DeepMind (London, United Kingdom), Moorfields Eye Hospital (London, United Kingdom), and University College London (UCL, United Kingdom), first trained a deep learning AI algorithm to spot ten different features of eye disease, based on 14,884 three-dimensional (3D), high-definition optical coherence tomography (OCT) retinal scans. The researchers then gathered another 997 scans and asked DeepMind and eight consultant ophthalmologists and specialist optometrists to recommend urgent referral, semi-urgent referral, routine referral, or observation for each scan.
In the most crucial category, urgent referral, DeepMind matched the two top retina specialists and had a significantly higher performance rate than the other two specialists and all four optometrists. No cases of urgent referral were missed by the system. When all referral types were taken into consideration, computer error rate was 5.5%, comparable to the error rate for the two best retina specialists (6.7% and 6.8%) and significantly better than the other six experts. When the clinicians had access to the OCT scans, fundus images, and patient summary notes, five had similar accuracy to the AI system, whereas the system still outperformed the other three. The study was published on August 13, 2018, in Nature Medicine.
“We set up DeepMind because we wanted to use AI to help solve some of society's biggest challenges, and diabetic retinopathy is the fastest growing cause of blindness worldwide. There are more than 350 million sufferers across the planet,” said co-senior study author Mustafa Suleyman, co-founder of DeepMind. “Detecting eye diseases as early as possible gives patients the best possible chance of getting the right treatments. I really believe that one day this work will be a great benefit to patients across the NHS.”
“Our research with DeepMind has the potential to revolutionize the way professionals carry out eye tests, and could lead to earlier detection and treatment of common eye diseases such as age-related macular degeneration,” said Professor Sir Peng Tee Khaw, MD, director of the Biomedical Research Centre in Ophthalmology at Moorfields Eye Hospital. “With sight loss predicted to double by the year 2050, it is vital we explore the use of cutting-edge technology to prevent eye disease.”
DeepMind is a British artificial intelligence company founded in September 2010 which created a neural network that learns how to play video games in a fashion similar to that of humans, as well as a neural network that may be able to access an external memory like a conventional Turing machine, resulting in a computer that mimics the short-term memory of the human brain; it was acquired by Google in 2014.
Related Links:
Google DeepMind
Moorfields Eye Hospital
University College London
Latest General/Advanced Imaging News
- Bone Density Test Uses Existing CT Images to Predict Fractures
- AI Predicts Cardiac Risk and Mortality from Routine Chest CT Scans
- Radiation Therapy Computed Tomography Solution Boosts Imaging Accuracy
- PET Scans Reveal Hidden Inflammation in Multiple Sclerosis Patients
- Artificial Intelligence Evaluates Cardiovascular Risk from CT Scans
- New AI Method Captures Uncertainty in Medical Images
- CT Coronary Angiography Reduces Need for Invasive Tests to Diagnose Coronary Artery Disease
- Novel Blood Test Could Reduce Need for PET Imaging of Patients with Alzheimer’s
- CT-Based Deep Learning Algorithm Accurately Differentiates Benign From Malignant Vertebral Fractures
- Minimally Invasive Procedure Could Help Patients Avoid Thyroid Surgery
- Self-Driving Mobile C-Arm Reduces Imaging Time during Surgery
- AR Application Turns Medical Scans Into Holograms for Assistance in Surgical Planning
- Imaging Technology Provides Ground-Breaking New Approach for Diagnosing and Treating Bowel Cancer
- CT Coronary Calcium Scoring Predicts Heart Attacks and Strokes
- AI Model Detects 90% of Lymphatic Cancer Cases from PET and CT Images
- Breakthrough Technology Revolutionizes Breast Imaging