Panel I

Detecting and Using Document Structure

The detection of discourse structure of scientific documents is important for a number of tasks, including biocuration efforts, text summarisation, and the creation of improved formats for scientific publishing. Currently, many parallel efforts exist to detect a range of discourse elements at different levels of granularity, and for different purposes, including extraction of information from complex documents, alignment of parallel corpora across languages, and support for document summarization (particularly multi-document summarization).  Another interesting class of applications comes from "bibliometrics" and "scientometrics".  For example, for analysis of argument structure in full text articles from the scientific literature, it may be important to know where a particular reference is cited or where a particular statement is made (Background, Discussion, etc.).  Another application might include tracking over time where (in what sections) an entity or concept is mentioned,  to determine whether the mentions migrate from research claims into the "Background" or eventually to the "Methods" sections of articles, as the concept moves from "foreground" (subject of the research) to "background".

In this panel we would like to, explore compare, contrast and evaluate different scientific discourse annotation schemes and tools, in order to answer questions such as:
  • What motivates a certain level, method, viewpoint for annotating scientific text?
  • What is the annotation level for a unit of argumentation: an event, a sentence, a segment? What are advantages and disadvantages of all three?
  • How easily can different schemes to be applied to texts? Are they easily trainable?
  • Which schemes are most portable? Can they be applied to both full papers and abstracts? Can they be applied to texts in different domains?
  • How granular should annotation schemes be? What are the advantages/disadvantages of fine and coarse grained annotation categories?
  • What correlations occur among document structure, argumentation, and rhetorical functions?
  • Is there a common framework that could be used for domain-independent document structure annotation?
Sophia Ananiadou, NaCTeM, University of Manchester, UK
Maryann Martone, Department of Neuroscience, University of California, San Diego
Ágnes Sándor, Xerox Research Europe, Grenoble 
Hagit Shatkay, University of Delaware, Newark, DE
Anita de Waard, Elsevier Labs, Burlington, VT (Panel Chair)

Panel II

Increasing Data Availability and Focus of Analytical Research for 4P’s (Patient, Provider, Pharmaceutical and Payer) of Healthcare: Trends, Opportunities and Gaps

Healthcare systems around the world are struggling to keep up with patient needs, and improve quality of care while reducing costs at the same time. At the same time, more and more data is being captured around healthcare processes in the form of Electronic Health Records (EHR), health insurance claims, medical imaging databases, disease registries, spontaneous reporting sites, and clinical trials. The panelists with extensive experience and representing the 4P’s of healthcare: Patients, Provider (hospitals, labs, clinics), Pharmaceutical and Payer (insurance companies, government) will discuss questions such as the following:
  1. What are the data sources that are becoming available across the 4Ps? How do they vary by geographies?
  2. What are the problems that are becoming feasible to be addressed by the Analytics/Data Mining community and who is the beneficiary for the addressed problems? In short, who and what can data analytics help with?
  3. Success stories and failures.
  4. Where are the gaps in the problems that are being addressed by the community and what can be done to bridge those gaps?
James Brady, Director, Innovation and R&D, OptumInsight 
Faisal Farooq, Sr. Scientist, Siemens Healthcare (Panel Chair)
Jyotishman Pathak, Asst. Professor, Mayo Clinic
William Reiter, MD, Board Member, HealthShare Montana
Shusaku Tsumoto, MD, Professor, Shimane University, Japan

Panel III

Towards Data: A Human/Machine-oriented Approach of Medical Data Collection

More than twenty years have passed since clinical data were computerized as a hospital information system, whose stored data include all the histories of clinical activities in a hospital, including accounting information, medical image, laboratory data and electronic patient records. Due to the traceability of all the information, a hospital cannot function without its information system. Furthermore, if it is extended into electronic healthcare records, it may not be a dream for each patient to benefit from their personal database with all the healthcare information. Recent advances in data mining will support this trend: analysis of such large scale databases enable us to visualize and capture what we have not seen only from clinical sites. However, it is notable that conventional computerized data are described by medical staff or measured by various kinds of medical instruments. They can be viewed as summaries of clinical processes, which remove background information behind the observations. Thus, analysis based on these data cannot overcome their limitations. One of the reasons why data mining is not successful for medical risk is that it takes a long time even for medical staff to interpret the results obtained and fill the gap between their knowledge and extracted patterns.  

If we want to go beyond these summaries and to get more information in order to capture the whole clinical actions or patients' behavior, we have to monitor and store their details through more sophisticated methods, such as sensor networks. For example, if we want to prevent medical incidents, in-hospital infections we have to monitor the behavior of medical staff, and if we want to prevent chronic diseases (metabolic syndromes) in an efficient way, we have to monitor the measurements of behavior of patients. Thus, although "On data" approaches are important for future medicine and healthcare, Ones of "Towards data" are more important for IT-oriented future development of these fields, which has been proposed by the organizer. These data collection and their analysis will be new challenges for healthcare IT and thanks to the recent developments of sensors and devices: it will not be a dream or science fiction. 

This panel gives recent advances in IT technology for data collection, their problems and their future vision: not only hardware-based or sensor-based, but also human-oriented, or human-agent-interaction based will be discussed by the panelists who play important roles in building up these fields. 

Shusaku Tsumoto, Shimane University (Panel Chair)
Mihoko Otake, University of Tokyo
Hiroshi Nakajima, OMRON
Takayuki Fujita, University of Hyogo
Yan Chow, Kaiser Permanente Information Technology

Panel IV

Social Media for Consumer Health

Community-based social media, characterized by its ability to glean collective wisdom through supporting communication, information sharing, and collaboration between individual users, is imposing a greater impact on people’s daily information seeking, knowledge construction, and decision making. Unsurprisingly, the impact of social media has extended to the health care domain, as consumers have begun to share health-related experiences and knowledge online [4]. Currently, about one-third of Americans who go online to research their health use social networks to find fellow patients and discuss their conditions [1], and 36% of social network users evaluate and leverage other consumers’ knowledge before making healthcare decisions, such as choosing providers, determining a course of treatment, and managing their health risks [3]. In a recent study, Pew Research declares that social media sites are becoming important hubs for health advice [2]. 

As social media emerge as a platform for disseminating and sharing of health-related information, more research is needed to elucidate the nature and working mechanisms of this new medium. This panel will explore the potential of social media, such as social networks and online health communities, for promoting consumer health from the following aspects: 

  1. How do end-users use social network sites (SNSs), a popular form of social media, for health- and wellness-related purposes? How do they perceive SNSs as a place for health information and communication? What kind of questions do they ask their social ties? What factors impact their decisions of using SNSs for health information? 
  2. Do we have the tools that end-users can use to discern high quality information from those that are of dubious quality on SNSs and other social media sites?  In the event of viral rumor-spreading that can happen quite rapidly in social media, what can public health organizations do to counter false information? How do we develop tools to enable national and international health officials to monitor the spread of false information and intervene fast to prevent its spread? How do we improve the existing tools ranging from those used at the individual scale to that at the aggregated national and international scale? 
  3. Online health communities have enabled people to obtain emotional support from other users [5]. What are important factors associated with the positive impacts of online health communities?  The success of these communities, to a large degree, is critically dependent on the active engagement of leaders. What are the characteristics of leaders in these communities? How can we facilitate the dissemination of valuable patients’ experiential knowledge and stories to the right people at the right time using social media integrated with online health communities?
  4. Research on social media often focuses either on regularities or patterns at the network level (e.g., social network analysis) or cognitive or affective behavior at the individual level. It is, however, possible that the interactive impact to individuals and their social context (as shaped by the online environment) can result in the emergence of new regularities. What are the benefits of bridging studies on individual information processing with higher level approach such as social network analysis or social computing method in health informatics? 
Yan Zhang, University of Texas at Austin (Panel Chair)
Wai-Tat Fu, University of Illinois at Urbana-Champaign
Prasenjit Mitra, Pennsylvania State University
John Yen, Pennsylvania State University