Jump to: navigation, search

Main Page

Revision as of 14:13, 17 November 2015 by Carlos Rizo (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)


NEW: patientSense is a collaboration between eHealth in Motion and Dataparc Communications Corporation, with contributions from Concordia University. It is supported by the Mitacs program.

[edit]

  • PatientSense is a tool and process for discovering and organizing digital documents, with an emphasis on the Web.
  • PatientSense builds on peer-reviewed methodologies to assist researchers in the manual curation of literature in three distinct ways:
  1. it highlights information automatically retrieved by our text mining system
  2. it provides researchers and web curators with an overview of the document content along with additional information and links to external resources
  3. it uses a proxy server approach, allowing a seamless integration with the user’s browser, and supports scraping of Web content
  • patientSense supports combinations of computer agents and human team members to effectively discover useful information using semantic analysis techniques and a powerful text engine
  • The software’s design is distributed, and it can supports web-scale search and organizing tasks with a focus on cognitive support for large scale tasks.

How to deal with information overlaod?

The Internet continues to expand at a rapid pace, with 2.5 quintillion bytes of data created every day. With estimates that 70% of patients access health information online, and roughly 10% contribute to online content, it is undeniable that patients are speaking out online and that someone needs to listen to what they are saying. PatientSense was created for this purpose.

  • PatientSense processes and manages vast volumes of structured or unstructured information found in the peer-review literature, the grey literature, and in patient conversations.
  • We will work with your group to organize your existing links and documents in patientSense, supporting a comprehensive search and faceted browsing interface with saved searches, search subscription and export to common formats
  • We will create a custom patientSense to systematically assess linked data at a wide scale, and support how you need to access it.
  • Our team approach includes sophisticated semantic agents designed around your research protocol.
  • The agents on patientSense will support researchers finding information in a large number of URLs, PDFs, and data, with relatively low standard errors of less than 5%, when using site-specific custom scrapers to precisely identify information.
  • Scrapers can include selection criteria, exclusion criteria, publication type, location, and so on.
  • We can also include grey literature or patient conversations on the web that are valuable to your mandate of for future projects.

Our results

  • Use of the patientSense approach, without the added scraping element, has reduced the time to triage publications by 21%.
  • patientSense obviates the need to limit search results on realist reviews, environmental scans or reviews, and keeps the searches and information extraction updating in the background based on the latest filters
  • patientSense preserves the format of the original document (including pictures, tables and embedded services) and annotations and tags can be used for future projects


David Mason

David is an Internet/Web based development of new systems. Multi million dollar, multi year projects in health and government public engagement (University Health Network, Toronto; Department of Foreign Affairs, Canada), international online human rights and independent media. Initiate and developing free/open software based systems, create and support teams, cultivate communities.

Davids specialties include: structured wikis, javascript/nodejs, project development, software development, host administration, people support.

Carlos Rizo MD

Carlos Rizo is a physician, medical futurist, patient activist, adventure sportsman, and advocate for personal healthcare, who also holds fellowships in Consumer Health Informatics and eHealth Innovation.

As a result of a spectacular accident in 2006, he became an expert in participatory and collaborative decision making models of patient care during life-threatening conditions. By tracking his spinal cord injury, retrieving his personal health information, and integrating it with the best available evidence from the Web, Carlos found answers not available in medical textbooks or expert medical minds, which enabled him to recover completely.

Being a patient now defines everything Carlos does. He has ample experience in research, global health systems, new media, social and open innovation, online collaboration, Internet startups and entrepreneurship. As a clinician with deep experience of the healthcare system, Carlos brings visionary and passionate insight to understanding needs at all levels, from the individual to entire health systems.

Using a range of digital tools to track, analyze and manage his health condition and the incorporation of traditional and social evidence gathered from a world-wide-web of medical and patient experts, he follows state-of-the-art, unique and practical approaches to data gathering, evidence integration and analytical methods to ensure that you benefit from the best patient focused data collection and evidence integration experience in the market.

Patients are talking online

The Internet continues to expand at a rapid pace, with 2.5 quintillion bytes of data created every day. This data comes from searching information, sharing digital pictures and video, creating and sharing websites, and social media interactions. Some comes from patients who search, browse and contribute daily to online forums. While formal surveys exist to measure active patient participation on the Web, there are no universal, organic indicators to track ongoing real-world consumer activities or capture opinions about health care. Although a knowledge-base regarding the factors that influence and initiate online consumer involvement in health care choices exists, an understanding of how patient use social media during illness has not not been well defined. The identification and description of patterns of information gathering, opinion expression, comparisons and sentiment about health care can help identify the utility of social media in decision-making over time. With estimates that 70% of patients access health information online, and roughly 10% contribute to online content, it is undeniable that patients are speaking out online and that someone needs to listen to what they are saying. PatientSense was created for this purpose.

We are listening to what patients say

With origins that trace back to 2009 at the Innovation Cell (innovationcell.com) and peer-reviewed publications, PatientSense is a joint effort of eHealth in Motion, a patient-focused innovation consultancy in Canada in partnership with dataparc, a semantic web innovation firm based in Montreal. The focus of this collaborative is to co-develop a tool that allows the observation and analysis of patient stories and interactions available on the web. PatientSense is among the first patient-focused projects using semantic media technologies to observe and listen to patient stories. The ultimate goal of PatientSense is to learn and support how patients think, helping one another through the evolution of health journeys, and how healthcare innovation are influenced.

Unlike brand tracking or social media monitoring, this project observed what consumers ask, compare or feel about specific health care issues (e.g. are consumers prospectively trying to make choices, how are they retrospectively interpreting what already happened?) With the variety, volume and velocity of patient conversations online, we opportunistically choose medications (i.e, antidepressants) or procedures (i.e., total knee replacement) as the focus of the exploration

We use PatientSense to listen better

PatientSense front-end was established with a set of starting assembly components, which enables describing initial executions (web scraping and analysis) in a distributed environment to process execution parameters. Upon completion of an execution, results are stored on the front end in a report format, with the ability to create annotations and evaluations. Based on evaluation, execution elements such as search terms are improved, and additional analysis methods are added to the search. We integrate Open Source and validated research technologies to built a centralized location for collaboration that identifies word-strings to distinguished forward-looking, prospective comments / questions / concerns from past-looking, retrospective comments / complaints / results about specific health issues.

Our team uses peer-review to improve the quality of the listening filters according to specific guidelines. Then non-relevant information is filtered via an iterative approach. Finally our team suggests rules and modifications to filter to improve the reliability of the information extracted from the web.

  • PatientSense provides cutting-edge web technologies that use Big Data approaches to better support our clients’ knowledge management infrastructure and automate the knowledge discovery and management process.

Twitter