DeepFake knee osteoarthritis X-rays from generative adversarial neural networks deceive medical experts and offer augmentation potential to automatic classification

Abstract

Recent developments in deep learning have impacted medical science. However, new privacy issues and regulatory frameworks have hindered medical data sharing and collection. Deep learning is a very data-intensive process for which such regulatory limitations limit the potential for new breakthroughs and collaborations. However, generating medically accurate synthetic data can alleviate privacy issues and potentially augment deep learning pipelines. This study presents generative adversarial neural networks capable of generating realistic images of knee joint X-rays with varying osteoarthritis severity. We offer 320,000 synthetic (DeepFake) X-ray images from training with 5,556 real images. We validated our models regarding medical accuracy with 15 medical experts and for augmentation effects with an osteoarthritis severity classification task. We devised a survey of 30 real and 30 DeepFake images for medical experts. The result showed that on average, more DeepFakes were mistaken for real than the reverse. The result signified sufficient DeepFake realism for deceiving the medical experts. Finally, our DeepFakes improved classification accuracy in an osteoarthritis severity classification task with scarce real data and transfer learning. In addition, in the same classification task, we replaced all real training data with DeepFakes and suffered only a $$3.79\%$$ 3.79 % loss from baseline accuracy in classifying real osteoarthritis X-rays.

Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions

“…Convolutional neural networks have significantly impacted computer vision in medicine 48 – 50 . Unfortunately, with the increase in neural network complexity comes difficulty in interpreting the clear etiologies of predictions, especially on a per-instance basis.…”

Transplant nephropathology: Wherefrom, wherein, and whereto

“…48 In this new realm of pathology, with AI so much in charge, we will begin to truly understand pathology and morphology for the very first time in ways never possible before, because of the potential of reverse engineering the light microscopic data. Reverse engineering occurred in radiology a few years ago, 49 where generative AI was shown to produce x-rays indistinguishable from those seen clinically, that are generated from the clinical data, predicting exactly what the radiology will look like in a set clinical situation, deep fake x-rays! Within the next year or two, we should be able to do the same with light microscopic pathology, whereby deep fake histology pictures are entirely created from the clinical data, predicting what the pathology will look like in specific disease states of various severities and under the influence of specific parameters.…”

Artificial intelligence in knee osteoarthritis: A comprehensive review for 2022

“…In [ 59 ], two GANs that can generate an unlimited number of KOA radiographs at different KL grades was proposed. The KL0 and KL1 grade images were merged into the KL01 class, whereas KL2, KL3, and KL4 ones were merged into the KL234 class.…”