Real-time tissue identification can benefit procedures such as stereotactic brain biopsy, functional neurosurgery and brain tumor excision. Optical scattering spectroscopy has been shown to be effective at discriminating cancer from noncancerous conditions in the colon, bladder and breast. The NASA Smart Probe extends the concept of ‘optical biopsy’ by using neural network techniques to combine the output from 3 microsensors contained within a cannula 2.7 mm in diameter (i.e. the diameter of a stereotactic brain biopsy needle). Experimental data from 5 rats show the clear differentiation between tissues such as brain, nerve, fat, artery and muscle that can be achieved with optical scattering spectroscopy alone. These data and previous findings with other modalities such as (1) analysis of the image from a fiberoptic neuroendoscope and (2) the output from a microstrain gauge suggest the Smart Probe multiple microsensor technique shows promise for real-time tissue identification in neurosurgical procedures.

No abstract

No abstract

The Ensemble project has consistently identified common patterns pertaining to activity planning across missions, analogs, and field tests. In this process, we have developed a unified planning repository that serves Mars Science Laboratory, International Space Station, Analog Field tests like Pavilion Lake Research Project, and many others. This repository encapsulates complexities associated with persistence, indexing, access and synchronization of data required for distributed planning. By providing a uniform Restful interface to the clients, the plan repository enables access from a variety of clients spanning multiple NASA centers. These clients range from shell scripts and browsers to mobile apps and complex desktop applications. It supports multiple representations, or formats, of plans to support multiple clients. With a focus on interoperability, end-users from disparate clients are able to interact with each other seamlessly. With support for rich querying capabilities through an inverted index of all of the activities and plans stored in the repository, we allow operators and scientists to search for the plans based on tactically relevant criteria with structured and unstructured (full-text) queries. Lastly, in order to support distributed planning, the repository incorporates a messaging layer to synchronize plans across multiple geographically dispersed clients in real-time. In this paper, we describe common patterns in tactical planning and the feature sets we have built to support them. Furthermore, we share key insights and lessons learned from deploying plan repository on a variety of missions and field tests. We provide an overview of the novel integration of technologies that support planning repository, and we discuss the role of cloud computing in our deployment strategy. We conclude with an overview of future research and upcoming features in the plan repository.