Hrishikesh Deshpande

André Gooßen, Hrishikesh Deshpande, Tim Harder et al.

Pneumothorax is a critical condition that requires timely communication and immediate action. In order to prevent significant morbidity or patient death, early detection is crucial. For the task of pneumothorax detection, we study the characteristics of three different deep learning techniques: (i) convolutional neural networks, (ii) multiple-instance learning, and (iii) fully convolutional networks. We perform a five-fold cross-validation on a dataset consisting of 1003 chest X-ray images. ROC analysis yields AUCs of 0.96, 0.93, and 0.92 for the three methods, respectively. We review the classification and localization performance of these approaches as well as an ensemble of the three aforementioned techniques.

Hrishikesh Arun Deshpande

Today, internet and web services have become an inseparable part of our lives. Hence, ensuring continuous availability of service has become imperative to the success of any organization. But these services are often hampered by constant threats from myriad types of attacks. One such attack is called distributed denial of service attack that results in issues ranging from temporary slowdown of servers to complete non-availability of service. Honeypot, which is a sort of a trap, can be used to interact with potential attackers to deflect, detect or prevent such attacks and ensure continuous availability of service. This paper gives insights into the problems posed by distributed denial of service attacks, existing solutions that use honeypots and how a mesh of virtualized honeypots can be used to prevent distributed denial of service attacks.