Generative Adversarial Network for Lupus Diagnostics and Medical Applications
Event Type
Student Poster
TimeTuesday, July 306:30pm - 8:30pm
LocationCrystal Foyer and Crystal B
DescriptionThe recent boom of Machine Learning frameworks like Generative Adversarial Networks (GAN), Deep Convolution Generative Adversarial Networks (DCGAN) and the development of high-performance computing for big data analysis has the potential to be highly beneficial in many domains and fittingly in the early detection of chronic diseases. The clinical heterogeneity of one such chronic autoimmune disease like Systemic Lupus Erythematosus (SLE), commonly referred to as Lupus, makes it difficult for medical diagnostics. This research employs unsupervised deep learning mechanisms to identify clinical manifestations of lupus from publicly available anonymous pictures of persons who present with cutaneous lesions like the butterfly rash, commonly seen in patients diagnosed with Lupus. We demonstrate the use of artificially generated butterfly rash images generated from GAN to train the discriminator model that differentiates Lupus from its other counter skin diseases using a Neural Network Classifier, as a use-case example. The expected outcomes are to help reduce the time in detection and treatment by gathering insights from its huge heterogeneous data clusters.