Deep learning models have demonstrated high-quality performance in areas such as image classification and speech processing. However, creating a deep learning model using electronic health record (EHR) data, requires addressing particular privacy challenges that are unique to researchers in this domain. This matter focuses attention on generating realistic synthetic data while ensuring privacy. In this paper, we propose a novel framework called correlation-capturing Generative Adversarial Network (corGAN), to generate synthetic healthcare records. In corGAN we utilize Convolutional Neural Networks to capture the correlations between adjacent medical features in the data representation space by combining Convolutional Generative Adversarial Networks and Convolutional Autoencoders.

Deep learning models have demonstrated high-quality performance in areas such as image classification and speech processing. However, creating a deep learning model using electronic health record (EHR) data, requires addressing particular privacy challenges that are unique to researchers in this domain. This matter focuses attention on generating realistic synthetic data while ensuring privacy. In this paper, we propose a novel framework called correlation-capturing Generative Adversarial Network (corGAN), to generate synthetic healthcare records. In corGAN we utilize Convolutional Neural Networks to capture the correlations between adjacent medical features in the data representation space by combining Convolutional Generative Adversarial Networks and Convolutional Autoencoders. To demonstrate the model fidelity, we show that corGAN generates synthetic data with performance similar to that of real data in various Machine Learning settings such as classification and prediction. We also give a privacy assessment and report on statistical analysis regarding realistic characteristics of the synthetic data.

Deep learning models have demonstrated high-quality performance in areas such as image classification and speech processing. However, creating a deep learning model using electronic health record (EHR) data, requires addressing particular privacy challenges that are unique to researchers in this domain. This matter focuses attention on generating realistic synthetic data while ensuring privacy. In this paper, we propose a novel framework called correlation-capturing Generative Adversarial Network (corGAN), to generate synthetic healthcare records. In corGAN we utilize Convolutional Neural Networks to capture the correlations between adjacent medical features in the data representation space by combining Convolutional Generative Adversarial Networks and Convolutional Autoencoders. To demonstrate the model fidelity, we show that corGAN generates synthetic data with performance similar to that of real data in various Machine Learning settings such as classification and prediction. We also give a privacy assessment and report on statistical analysis regarding realistic characteristics of the synthetic data. The software of this work is open-source and is available at: https://github.com/astorfi/cor-gan.

Uber has teased a look at what its futuristic Skyport flying taxi hubs could be like when UberAir comes to life. At the firm's Elevate Summit in Los Angeles, Uber unveiled elaborate concept images of the Connect system developed by Corgan that could provide infrastructure for the vertical take-off and landing craft. The modular system can essentially be installed anywhere, be it an open site, atop a parking garage, or even on the roof of a skyscraper, according to Corgan. Uber has teased a look at what its futuristic Skyport flying taxi hubs could be like. At the firm's Elevate Summit in Los Angeles, Uber unveiled elaborate concept images of the Connect system developed by Corgan that could provide infrastructure for UberAir Uber has plans to begin its first flight demonstrations as soon as 2020, and begin taking passengers by 2023.

One major caveat, however, is that unlike regular taxis which can freely zip about the streets, UberAIR taxis need access to the sky and a place to land. That's where "Skyports" come in: special areas localized specifically for launching, landing, and customer pickup, and they're looking appropriately futuristic. During the second day of Uber's Elevate Summit 2018, the company revealed concepts for its air taxi ports. While still at the early developmental stages, Uber plans to support over 4,000 passengers per hour, per Skyport. These initial blueprints come from Corgan, an architecture firm keen on "transforming urban air mobility" with what it calls "Connect," an infrastructure that will enable up to 1,000 Uber eVTOLs (Electric Vertical Take-Off and Landing) each hour.