New research could help AI identify rare conditions in medical images by augmenting existing datasets
Artificial intelligence (AI) holds real potential for improving both the speed and accuracy of medical diagnostics. But before clinicians can harness the power of AI to identify conditions in images such as X-rays, they have to 'teach' the algorithms what to look for.
Identifying rare pathologies in medical images has presented a persistent challenge for researchers, because of the scarcity of images that can be used to train AI systems in a supervised learning setting.
Professor Shahrokh Valaee and his team have designed a new approach: using machine learning to create computer generated X-rays to augment AI training sets.
"In a sense, we are using machine learning to do machine learning," says Valaee, a professor in The Edward S. Rogers Sr. Department of Electrical & Computer Engineering (ECE) at the University of Toronto. "We are creating simulated X-rays that reflect certain rare conditions so that we can combine them with real X-rays to have a sufficiently large database to train the neural networks to identify these conditions in other X-rays."
Valaee is a member of the Machine Intelligence in Medicine Lab (MIMLab), a group of physicians, scientists and engineering researchers who are combining their expertise in image processing, artificial intelligence and medicine to solve medical challenges. "AI has the potential to help in a myriad of ways in the field of medicine," says Valaee. "But to do this we need a lot of data -- the thousands of labelled images we need to make these systems work just don't exist for some rare conditions."
To create these artificial X-rays, the team uses an AI technique called a deep convolutional generative adversarial network (DCGAN) to generate and continually improve the simulated images. GANs are a type of algorithm made up of two networks: one that generates the images and the other that tries to discriminate synthetic images from real images. The two networks are trained to the point that the discriminator cannot differentiate real images from synthesized ones. Once a sufficient number of artificial X-rays are created, they are combined with real X-rays to train a deep convolutional neural network, which then classifies the images as either normal or identifies a number of conditions.
"We've been able to show that artificial data generated by a deep convolutional GANs can be used to augment real datasets," says Valaee. "This provides a greater quantity of data for training and improves the performance of these systems in identifying rare conditions."
The MIMLab compared the accuracy of their augmented dataset to the original dataset when fed through their AI system and found that classification accuracy improved by 20 per cent for common conditions. For some rare conditions, accuracy improved up to about 40 per cent -- and because the synthesized X-rays are not from real individuals the dataset can be readily available to researchers outside the hospital premises without violating privacy concerns.
"It's exciting because we've been able to overcome a hurdle in applying artificial intelligence to medicine by showing that these augmented datasets help to improve classification accuracy," says Valaee. "Deep learning only works if the volume of training data is large enough and this is one way to ensure we have neural networks that can classify images with high precision."
Marit Mitchell | EurekAlert!
The holy grail of nanowire production
20.02.2019 | Ecole Polytechnique Fédérale de Lausanne
Combining infrared radiation and air management to reduce energy use
19.02.2019 | Heraeus Noblelight GmbH
Up to now, OLEDs have been used exclusively as a novel lighting technology for use in luminaires and lamps. However, flexible organic technology can offer much more: as an active lighting surface, it can be combined with a wide variety of materials, not just to modify but to revolutionize the functionality and design of countless existing products. To exemplify this, the Fraunhofer FEP together with the company EMDE development of light GmbH will be presenting hybrid flexible OLEDs integrated into textile designs within the EU-funded project PI-SCALE for the first time at LOPEC (March 19-21, 2019 in Munich, Germany) as examples of some of the many possible applications.
The Fraunhofer FEP, a provider of research and development services in the field of organic electronics, has long been involved in the development of...
For the first time, an international team of scientists based in Regensburg, Germany, has recorded the orbitals of single molecules in different charge states in a novel type of microscopy. The research findings are published under the title “Mapping orbital changes upon electron transfer with tunneling microscopy on insulators” in the prestigious journal “Nature”.
The building blocks of matter surrounding us are atoms and molecules. The properties of that matter, however, are often not set by these building blocks...
Scientists at the University of Konstanz identify fierce competition between the human immune system and bacterial pathogens
Cell biologists from the University of Konstanz shed light on a recent evolutionary process in the human immune system and publish their findings in the...
Laser physicists have taken snapshots of carbon molecules C₆₀ showing how they transform in intense infrared light
When carbon molecules C₆₀ are exposed to an intense infrared light, they change their ball-like structure to a more elongated version. This has now been...
The so-called Abelian sandpile model has been studied by scientists for more than 30 years to better understand a physical phenomenon called self-organized...
11.02.2019 | Event News
30.01.2019 | Event News
16.01.2019 | Event News
21.02.2019 | Earth Sciences
21.02.2019 | Trade Fair News
21.02.2019 | Life Sciences