Andrew Huynh

Crowdsourcing the Unknown: The Satellite Search for Genghis Khan

Massively parallel collaboration and emergent knowledge generation is described through a large scale survey for archaeological anomalies within ultra-high resolution earth-sensing satellite imagery. Over 10K online volunteers contributed 30K hours (3.4 years), examined 6,000 km2, and generated 2.3 million feature categorizations. Motivated by the search for Genghis Khans tomb, participants were tasked with finding an archaeological enigma that lacks any historical description of its potential visual appearance. Without a pre-existing reference for validation we turn towards consensus, defined by kernel density estimation, to pool human perception for “out of the ordinary” features across a vast landscape. This consensus served as the training mechanism within a self-evolving feedback loop between a participant and the crowd, essential driving a collective reasoning engine for anomaly detection. The resulting map led a National Geographic expedition to confirm 55 archaeological sites across a vast landscape. A increased ground-truthed accuracy was observed in those participants exposed to the peer feedback loop over those whom worked in isolation, suggesting collective reasoning can emerge within networked groups to outperform the aggregate independent ability of individuals to define the unknown.

Visual analytics of inherently noisy crowdsourced data on ultra high resolution displays

The increasing prevalence of distributed human microtasking, crowdsourcing, has followed the exponential increase in data collection capabilities. The large scale and distributed nature of these microtasks produce overwhelming amounts of information that is inherently noisy due to the nature of human input. Furthermore, these inputs create a constantly changing dataset with additional information added on a daily basis. Methods to quickly visualize, filter, and understand this information over temporal and geospatial constraints is key to the success of crowdsourcing. This paper present novel methods to visually analyze geospatial data collected through crowdsourcing on top of remote sensing satellite imagery. An ultra high resolution tiled display system is used to explore the relationship between human and satellite remote sensing data at scale. A case study is provided that evaluates the presented technique in the context of an archaeological field expedition. A team in the field communicated in real-time with and was guided by researchers in the remote visual analytics laboratory, swiftly sifting through incoming crowdsourced data to identify target locations that were identified as viable archaeological sites.

Search and Discovery Through Human Computation

In the latest evolution of the Internet, human networks are becoming functionalized through collective collaboration frameworks. Questions are now being addressed as never before, by leveraging the easy digital accessibility of crowds to supplement the limitations of machine computation. This is especially relevant in the case of visual analytics where human intuition remains beyond the scope of existing computer object recognition algorithms. Distributing the effort over a massive network of humans not only succeeds in expanding the capacity of human based analytical power, but if set up appropriately, can also provide a statistical basis to pool human perceptive knowledge when identifying the unknown. Here we describe the impacts of this capacity in efforts of search and discovery, where massively parallel human computation can be used to identify anomalies of loosely defined characteristics within large volumes of ultra-high resolution multi-spectral satellite imagery. As human generated data is inherently noisy and subjective in nature, a statistical approach is taken towards consensus based data validation. We show that a spatial landscape can serve as the framework for collaborative computation through an overview of our initial efforts in archaeology, and the subsequent applications in disaster assessment, and search and rescue.

Limitations of crowdsourcing using the EMS-98 scale in remote disaster sensing

The combination of crowdsourcing and high-resolution aerial and satellite imagery have recently been explored in applications of post-disaster damage assessment and analytics. The large scale and distributed nature of these tasks produce overwhelming amounts of information that can be used to quickly assess post-disaster damage. While crowdsourcing provides fast, scalable analysis of a disaster area, the accuracy or usefulness of the data collected is not well understood. This paper investigates the variability and limitations in user assessment as observed by a standard damage assessment metric EMS-98, for each of the five damage categories. A case study is provided through the crowdsourced damage assessment of a region along the eastern shore of Japan that was effected by the 2011 earthquake and tsunami. High-resolution satellite imagery was captured after the event and was used to provide the images needed for damage assessment.

Mobile analysis of large temporal datasets for exploration and discovery

The increasing power and decreasing size of mobile devices and tablets provide a compelling new platform for mobile field geographic information system (GIS) operations with geospatial datasets. As these data sets become increasingly larger and more dynamic, it introduces the need for real time analysis and data collection in the field. Successful field research in the digital age requires computing power, functionality, and mobility. Location aware mobile computing devices enable new methods of knowledge synthesis by merging physical and virtual data layers. The geospatial dataset in this study is both large in scale and quickly transforms on a daily basis, requiring innovative strategies for effective application. This paper presents the novel combination of mobile tablets, GIS, and geospatial data to direct ground exploration and discovery of cultural heritage sites in Mongolia.

HIGHER PLATELET INHIBITION WITH THIENOPYRIDINE THERAPY, RATHER THAN SERUM FIBRINOGEN LEVEL, IS ASSOCIATED WITH PERIPROCEDURAL BLEEDING AFTER PERCUTANEOUS CORONARY INTERVENTION

Higher fibrinogen level and on-thienopyridine platelet reactivity are associated with ischemic cardiac events after PCI, while relationships with bleeding after PCI are unclear. Subjects pretreated with thienopyridines undergoing elective or urgent (acute coronary syndrome [ACS]) PCI were enrolled

ELEVATED SERUM FIBRINOGEN LEVEL IS ASSOCIATED WITH LONG-TERM MAJOR ADVERSE CARDIAC EVENTS FOLLOWING PERCUTANEOUS CORONARY INTERVENTION

Elevated fibrinogen is associated with short-term major adverse cardiac events (MACE) after percutaneous coronary intervention (PCI). The relationship between pre-procedural fibrinogen and long-term MACE after PCI is unknown. Individuals undergoing elective or urgent (acute coronary syndrome [ACS