Conclusion

The Semantic Web will revolutionize how information is shared and used in the future. It will allow users to consolidate and analyze vast stores of medical knowledge that have previously been unavailable. It will change the way patients interact with health care providers, and has the potential to improve how everyone manages their health on a daily basis. This technology is truly amazing, and will affect everyone’s lives in ways that are still difficult to imagine.

Possible Future Semantic Web Applications in Healthcare

The potential of the Semantic Web for delivering or improving health care is only limited by our imagination. Here are some ideas for how this technology might be used in the future.

Analysis of the Impact of External Conditions on Health

Semantic Web technology could be used to connect geographic, environmental, and cultural information from separate databases with patient information in electronic health records. This would give health care providers a more holistic view of the patient's living environment, and identify previously unknown external factors that may affect health. These factors could also be considered when the health care provider develops therapy or a wellness plan for the patient. This technology would also provide medical researchers with a means of conducting demographic analysis to evaluate how the complex interactions between environment and culture affect the health of specific populations.

Improve Computer Aided Robotic Surgery

Semantic Web technology might be used to consolidate and give meaning to information from a variety of electronic sources and then coordinated to support robotic surgery. For example, information from a patient's electronic preoperative tests, diagnostics, and electronic health records, could be synthesized into meaningful information that is mapped to robotics logic and computerized monitoring devices so as to improve the overall safety and efficacy of the surgery.

Telemedicine Communications Translator

Semantic Web technology could be used to translate words spoken into a voice recognition device into a foreign language that is “spoken” or printed by a computer on the receiving end of a VoIP connection. This type of technology would facilitate the delivery of telemedical clinical care to patients who speak different languages or to the deaf.

eHealth Record On-Demand Consolidation

Semantic Web technology could be used to consolidate electronic health records located in a variety of databases such as doctors, hospitals, pharmacies, and walk-in clinics, into a single patient file as needed. This would allow health care providers to make clinical decisions based on complete patient records. This technology could also be used to translate the consolidated health records into other languages to support the delivery of health care to patients regardless of their location and language. This consolidation and translation to other languages would also facilitate global medical research.

Interactive Health Care Management Trackers

The Semantic Web could be used to provide a customized interactive check list of actions that patients must take to manage their own health care. For example, the tracker could extract information from the patient’s electronic health care record to identify when annual examinations are required, and automatically contact the health care provider’s system to compare open appointments with open time on the patient’s electronic calendar. It could track insurance deductibles and individual payments for health care, and use historical information to estimate future health costs. An interactive health care management tracker could also track medications, automatically submit refill requests to a pharmacy, automatically request updated prescriptions from doctors, and use prescription and exam information from electronic health care records to query on-line pharmaceutical databases and extract customized information about possible drug interactions and side-effects.

References

Application of Digital Surgery in Orthopedics: THA and TKP. Retrieved 5/19/2010 from http://biomed.brown.edu/Courses/BI108/BI108_2005_Groups/04/orthopedics.htm

Bonsor, K. and Strickland, J. (200x). How Robotic Surgery Will Work. Retrieved 5/21/2010 from http://science.howstuffworks.com/robotic-surgery.htm

Dolbear, C. (2007). Ordonance Survey, UK Case Study: Semantic Web Technology at Ordnance Survey. Retrieved 5/20/2010 from http://www.w3.org/2001/sw/sweo/public/UseCases/OrdSurvey/

Federal Communications Commission. VoIP Frequently Asked Questions. Retrieved 5/21/2010 from http://www.fcc.gov/voip/.

Marescaux, J. MD, et al. (2002). Transcontinental Robot-Assisted Remote Telesurgery: Feasibility and Potential Applications. Retrieved 5/21/2010 from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1422462/

Public Health Situation Awareness with Semantic Web

A third example of the power of Semantic Web in health related areas is the public health situational awareness. Public health observation “is the ongoing collection and analysis of data to identify and respond to community health problems in a timely manner” [1]. It is common the important public health information is spread across many databases. Information about gastrointestinal or respiratory illness, water contamination, and air quality, is typically available through emergency room database records. However, the meaningful integration of these databases is necessary for them to be of use in public health surveillance systems.

The School of Health Information Science of the University of Texas has developed a system based on Semantic Web technology that integrates unstructured text such as doctors’ notes and patient complaints into structured electronic medical records. Code named SAPPHIRE (Situational Awareness and Preparedness for Public Health Incidences and Reasoning Engines), the system also facilitates the recovery and consolidation of all medical record information through a unified query interface, regardless of how (structured, unstructured) and where (text files, database tables, spreadsheets, etc.) the data is stored. Additionally, SAPPHIRE contextualizes data for a variety of different tasks. For example, creates models for identifying patients with influenza-like symptoms with neurological or gastrointestinal side effects to identify categories of outbreaks.

The prototype can be modeled to dynamically absorb the latest data feeds from external databases, then make new interpretations of the data accordingly. These capabilities were demonstrated during hurricane Katrina, where SAPPHIRE “was extended to accommodate data from a just-in-time PDA based questionnaire, a clinical information system from Katrina shelters and surveillance reports captured by the Houston Department of Health, and to support signal detection and reporting on diseases and syndromes defined by field epidemiologists. SAPPHIRE was the sole surveillance technology to address these needs within eight hours of the shelters opening, using Internet and Semantic Web technology”.

References

Mirhaji, P. (2007). Case study: Semantic web technology for public health surveillance, university of texas. Retrieved 5/17/2010, 2010, from http://www.w3.org/2001/sw/sweo/public/UseCases/UniTexas/

Pharmaceutics SURVEYOR

Another application for Semantic Web in health care is the drug regimen selection. The US population is highly concern about drug safety. Millions of Americans visit health care websites to identify side effects of a specific drug. Americans also investigate specific drugs to determine if interactions with other drugs will cause unanticipated health concerns. Thousands of deaths occur each year due to erroneous drug selections and drug interactions. Many of these deaths could have been avoided if the existing knowledge of drug side affects and interactions stored in databases has been effectively applied.

Nowadays, the availability of a vast array of drugs makes evaluating their safety a hugely complex and time consuming task. A single criterion is not enough to make a selection. Many factors such as drug interactions, side effects, efficacy, and impacts on certain health problems must be considered. Physicians must evaluate other special factors as pill size and the patient's relative weight, just to prescribe one drug.

A survey system has been developing to help in this task. The PharmaSURVEYOR allows a map-reading of this complex task to identify the best trade-offs for each patient. Figure 1, provides a view of the proposed PharmaSURVEYOR.

Figure 1.

PharmaSURVEYOR, two possible regimens are shown considering the actual regimen [1].

In the specific case shown in Figure 1, PharmaSURVEYOR provides choices based on optimizing treatment safety while also relieving the physical condition. The physician and the patient must make trade-offs in risks and side effects when choosing from the various treatment options identified by the system.

To achieve the optimal use of Semantic language (RDF and OWL) when querying drug data, the system provides a standardized and interoperable form of medical ontologies comprised of signs and symptoms that are delivered in a user friendly language that patients can understand. This ontology supports lateral and hierarchical data relationships in the underlying databased that are queried.

This application of Semantic Web technology provides a valuable method for sharing drug side affects and interactions between patients, physicians, pharmacists, and researchers. This collaborative scenario helps improve clinical drug therapy over time.

References

Von Schweber, E. (2007). Case study: Composing safer drug regimens for the individual patient using semantic web technologies. Retrieved 5/16/2010, 2010, from http://www.w3.org/2001/sw/sweo/public/UseCases/PharmaSurveyor/

Applications of Semantic Web

To show the benefits the Semantic Web offers to medicine, we present a set of application examples currently in use. These examples are data harmonization, drug selection, and public health behavior.

Data harmonization in Chinese Medicine [2]

It is well known that Chinese Medicine has an special perspective of the human body. Practitioners of this unique medicine consider that many components are part of the health condition of a patient. Some of these components are the ideas of “ying-yang”, “five-element”, food, and spirit.

On the other hand, Western Medicine considers Chinese Medicine an alternative method of healing rather than a scientific one. Conventional Western Medicine views the human body as parts, for example, body sections (upper or lower) or the different organs. Western Medicine has, however, scientifically proven the efficacy of Chinese Medicine in the prevention and therapy of some illnesses, and is interested in learning more from the vast stores of Chinese medical knowledge.

To that end, a bridge between the Western and Chinese medical databases is necessary. Currently, the greatest interest is in data related to biomedical and clinical research. A significant and growing amount of literature on these topics is available in a wide variety of databases. However, the language barrier makes Western access difficult because most of the literature on Chinese Medicine is written with Chinese characters.

Semantic Web (SW) technology has been used to solve this problem. The China Academy of Chinese Medicine Sciences has developed a method for sharing Chinese medical literature.

Because of the difference in view of medicine, a simple translation was not sufficient. An ontology mechanism was developed to correct translations between Chinese characters and western terms, and connect data in a semantic layer sitting on top of the existing relational databases. This makes knowledge of Chinese medicine available to the rest of the world. This integration of East and West is a positive step in collaboration among health care and life science.

Figure 1. Architecture of the Semantic Web layer and its role in unifying and linking heterogeneous relational data.

References

Cheung, K., & Chen, H. Semantic-based search and query system for the traditional chinese medicine community. Retrieved 5/17/2010, 2010, from http://www.w3.org/2001/sw/sweo/public/UseCases/UniZheijang/

Cheung, K., & Chen, H. (2010). Semantic web for data harmonization in chinese medicine. Chinese Medicine, 5, 2. doi:10.1186/1749-8546-5-2

Semantic Computing Technical Explanation

Semantic Computing is the technology that drives the engine that makes the Semantic Web possible. As described in other posts, the Semantic Web take human language and turn it into a machine language that can provide more powerful applications due to a similar use of language between user and system.

The Semantic Web's impact health care will be explained in later posts, but how it functions from a technical perspective will be examined here. Semantic Computing is the bridge to other applications and interfaces outside the web. It can be organized into 5 layers as seen below.

Figure 1 (5 layers of Semantic Computing)

1. Semantic Analysis is the layer that converts signals (pixels) into meanings (semantics)
2. Semantic Integration integrates information from the analysis layer into a unified model
3. Semantic Services are located at layer three, and they use the information provided from the Integration layer
4. Service Integration provides the inter operation of services to provide more powerful services
5. Semantic Interface is the GUI or consumer facing layer where manipulation and access of the various sources is available.

Of these layers the place where most impact will be garnered in health care are layers 4 and 5. They provide the access to more powerful services and can be changed and adjusted to fit the needs of the individual application or user.

Most of the interfaces that are used to provide health care impact will reside in those layers and will use some type of SSE or Semantic Search Engine.

Figure 2 (SSE Architecture)

The diagram can be read from the top down. The user interface takes the request from the user and is able to understand SNL or Structured Natural Language. The interpreter takes the SNL and converts it into Service Query Description Language (SQDL). The matcher combines the SQDL and the Service Capability Description Language (SCDL). At the final stages either a service invoker or the SCDL will make the final call to the Search Engine based on the converted language from the user that was established initially. The Service Invoker can be changed and customized to provide various items and was designed from the onset to be interfaced and programmed by smaller developers. This will allow this computing architecture to be versatile and agile in development and open to all service providers and developers alike.

This computing methodology is built to interface directly with Cloud Computing. The framework was established from the ground up to be web aware and to be able to send and receive information from the Cloud whenever possible. This is an important step moving forward, especially in the realm of health care. This feature will open new technological avenues in health care not often seen with the web, in terms of storage, search and delivery of data.

References

Sheu, P.C. et al. (2009). Semantic Computing, Cloud Computing, and Semantic Search Engine. Semantic Computing, 2009. ICSC'09. IEEE International Conference on , vol., no., pp.654-657, 14-16 Sept.2009. Retrieved May 3, 2010, from http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5298710&isnumber=5298512Article 3

Introduction

The Semantic Web describes a web-based technology which gathers pieces of data and combines the data jointly to be viewed, queried, and/or analyzed for use.

Semantic Web technology currently provides a user with the ability to search for information stored as data. Once collected, the user makes choices about how to make connections between data points.

The Semantic Web has enhanced traditional web technology by assembling the available content, including text, images, voice, and video, into a common design for faster reuse and plasticity of the data.

Formats for the Semantic Web were built on XML technologies, and use RDF, Resource Description Framework, for data exchange. RDF is a standard for creating data files, for voice, text, images and video, in a common file format. The RDF scenario provides an uniform platform for users to discover, distribute and recombine information. This is a very important concept because presently the Web aids in viewing of documents, whereas the Semantic Web will use technologies to exchange different data types in a uniform framework. In other words, HTML is to documents what RDF is to data.

An example of how the Semantic Web works is provided by Tim Berners-Lee, the creator of the World Wide Web. Mr. Berners-Lee compares Online Bank statements as an diagram of the Semantic Web.



With Online Banking, a consumer can download statements and analyze graph trends and pull statistical data from queries. Just imagine...data from that bank statement could be superimposed over your calendar to see when you wrote the check by jumping from one domain to another. The Semantic Web will allow for dynamic manipulation of data for maximum benefit for us mere humans.

References

Basic HTML and CSS Tutorial. Howto make website from scratch. podcast, youtube.com http://www.youtube.com/watch?v=OGg8A2zfWKg&feature=related

Billings, D. M. (2008). Quality Care, Patient Safety, and the Focus on Technology. Journal of Nursing Education , Vol. 47 (Iss. 2), p. 51-2 (pp.).

Brynko, B. (2010, 1). Semantic Search: Fact or Myth? Retrieved 4 22, 2010 from InfoToday.com: http://pqasb.pqarchiver.com/infotoday/access/1938697901.html?dids=1938697901:1938697901:1938697901&FMT=ABS&FMTS=ABS:FT:PAGE&type=current&date=Jan+2010&author=Barbara+Brynko&pub=Information+Today&edition=&startpage=27&desc=Semantic+Search%3A+Fact+or+Myth%3

Comming Soon: The Semantic Web. (2007, 1). PC magazine , p. 16.

Intro to Semantic Web. podcast. Youtube.com http://www.youtube.com/watch?v=OGg8A2zfWKg&feature=related

Lee Feigenbaum, S. M. (2007). Boca: an open source RDF store for building Semantic Web applications. Briefings in Bioinformatics , Vol. 8 (No.3), pps. 195-200.

Markoff, J. (2006, 11 12). Entrepreneurs See a Web Guided by Common Sense. The New York Times . New York, New York, USA.

That is the Semantic Web? (n.d.). Retrieved 4 22, 2010, from Whatis.com: http://searchsoa.techtarget.com/sDefinition/0,,sid26_gci214349,00.html

The Semantic Web of Data Tim Berners-Lee . podcast youtube.com The Semantic Web of Data Tim Berners-Lee Understanding XML. podcast youtube.com http://www.youtube.com/watch?v=UqwGSo82cwU