Given my interest in technology transfer and an entrepreneurial spirit, I along with two colleagues, Massimiliano Versace and Anatoli Gorchetchnikov have founded Neurala LLC, a startup company focused on the development of brain-based technologies to assist in more intelligent analysis of data and problem solving.
Along these lines we have been working to develop a technology platform to be used in various application domains including but not limited to acoustic scene analysis, remote sensing, financial time series forecasting, and intrusion detection, just to name a few. Two of the major projects that I have been directly involved with are highlighted below.
Neurala applications
 |
 Identifying objects in remote sensing data is a challenging task that requires highly skilled and costly human experts to analyze massive amount of digital data. Within Neurala, I have led the efforts in realizing a remote sensing software platform for the purpose of developing automatic methods to delineate water flow and soil loss from LIDAR data. The application of bio-inspired machine vision and classification algorithms is key to accurately and quickly process large amount of remote sensing data, extracting knowledge that is then applied to make better and faster decisions in soil conservation practices. |
 |
Classifying sounds in real time in mobile platforms has great potential for civilian and military applications. I have led Neurala in a collaboration with Biomimetic Systems (BMS) to perform acoustic scene classification for military applications. This work is part of a larger project called Red Owl, a collaboration between Boston University, iRobot, Insight Technology, and BMS. The system, assembled for use on iRobot’s unmanned PackBot robots, models of which are now deployed in Iraq and Afghanistan, is designed to enhance the situational awareness of soldiers in the battlefield. The system is undergoing extensive testing by multiple groups within the armed forces. |
Watch the videos of Neurala's Technology Platform while it classifies sounds in real-time