WikiBI/Applications

From AI Wiki

Data/text/web mining

Contents 1 Abstract 2 Introduction 3 Biomedical Informatics: An Overview of the Field 3.1 Important Concepts 3.2 Applications of Biomedical Informatics 3.2.1 Electronic Medical Records 3.2.2 Data Mining and Knowledge Discovery in Databases 3.2.2.1 Clinical Decision Support Systems (CDSS) 3.2.3 Computerized Physician Order Entry (CPOE) 3.2.4 Clinical Workflow Systems 3.2.5 Telemedicine 3.2.6 Public Health and Disease Outbreak Monitoring Applications 3.2.7 Human Genome Project 3.2.7.1 The Cancer Genome Atlas 3.2.8 cancer Biomedical Informatics Grid™( caBIG™) 3.2.9 Radio Frequency Identification for Biomedical Informatics 3.2.9.1 RFID for Patient, Personnel and Asset Tracking 3.2.9.2 RFID for ID verification and Recognition 3.2.9.3 RFID for Senior Citizens 3.2.9.4 RFID for Medical Monitoring 4 Case Study 1 : Biomedical Informatics on the Web 4.1 Introduction 4.2 Their Applications and How they are Used 4.2.1 Blogs 4.2.2 e-Portals 4.2.3 e-Multimedia 4.2.4 Online Support Groups 4.3 What are they Addressing/Solving 4.4 Benefits of these New Applications 4.5 Challenges of these New Applications 4.6 Resources: Biomedical Informatics on the Web 4.6.1 Blogs 4.6.2 e-Multimedia 4.6.3 e-Portals 4.6.4 Online Support Groups 4.7 Summary 5 Case Study 2 : Service-Oriented Architecture applied for Biomedical Informatics 5.1 Introduction 5.2 How would SOA be utilized for Biomedical Informatics 5.3 Issues that are Addressed with SOA 5.4 Benefits of SOA in Biomedical Informatics 5.5 Challenges to SOA in Biomedical Informatics 5.6 Summary 6 Overall Concluding Remarks 7 References 8 Online Resources 9 Questions for Discussion

Abstract

Preventable medical errors kill the equivalent of more than a jumbo jet full of people every day in the US and about 25 people per day in Australia. ^[1]Stopping these deaths due to such errors is an urgent matter. Biomedical Informatics are the answer to stopping all these errors. Applications like RFID for patient identification, Electronic Medical Records, and Disease Outbreak Monitoring are few examples. New interesting areas are emerging in the biomedical informatics field; biomedical informatics on the web and SOA for biomedical informatics.

Introduction

In the September 22nd 2005 edition, Time Magazine featured the story of Melinda Amedee, a Hurricane Katrina victim. Melinda was scheduled to have a tumor removed from her kidney at Ochsner Cancer Institute, a New Orleans hospital, on August 30. Due to her painful 25 year history of kidney problem, Melinda who was 39 at the time, was worried for any further delays in treatment. She quickly arranged to have the surgery at the MD Anderson Cancer Center in Houston instead. But like thousands of other patients evacuated after the storm, Melinda presented her new doctors with a challenge: no medical records, and no way of contacting her Louisiana kidney specialist. ^[2]. Electronic Medical Records and Biomedical Informatics may be the answer to all these challenges.

In the this chapter, we will overview the field of biomedical informatics including important concepts, literature review of the biomedical informatics applications, examine case studies of new and promising biomedical informatics tools, and finally pose some questions for future discussions.

Biomedical Informatics: An Overview of the Field

In this section will review the filed of biomedical informatics' applications and their uses. We will discuss important concepts, and review the literature concerning applications of biomedical informatics starting from the more traditional or conventional applications to more of the new and promising applications.

Important Concepts

Biomedical Informatics: are defined by Edward Shortliffe “as the scientific field that deals with biomedical information, data, and knowledge – their storage, retrieval, and optimal use for problem-solving and decision making”. ^[3]

Medical Informatics: Edward Shortliffe and Robert Greenes both defined Medical Informatics as: “the field that concerns itself with the cognitive, information processing, and communication tasks of medical practice, education, and research, including the information science and the technology to support these tasks.” ^[4]

Electronic Medical Record (EMR): EMRs are a patient’s medical record put in a digital format thus providing physicians and healthcare providers access to the patient's medical records from multiple locations across space and time. EMRs capture patient’s clinical diagnosis, lab results, prescriptions, allergies to certain drugs as examples. ^[5]

Computerized Physician Order Entry (CPOE): is an electronic information system used by physicians to communicate promptly their treatment instructions and recommendations of a certain patient to other medical staff, including other physicians and nurses, over a computer network. They allow for safer healthcare practices due to the fact that these systems have the ability to check if the patient has an allergy to a certain drug for example.^[6] They also eliminate errors of dispensing the wrong medication because of a mistake interpreting the physicians hand-writing. ^[7]

Service Oriented Architecture (SOA): SOA can be defined as "the architectural style that supports loosely coupled services to enable business flexibility in an interoperable, technology-agnostic manner. SOA consists of a composite set of business-aligned services that support a flexible and dynamically reconfigurable end-to-end business processes realization using interface-based service descriptions." ^[8]

Applications of Biomedical Informatics

In this section we will review the literature around the applications of biomedical informatics and how they are used in the field. Although this is not an exhaustive list of all biomedical informatics’ applications, it provides a good sense of these applications and how they are used. More traditional or conventional applications will appear first, and ending with the newer and promising application.

Electronic Medical Records

Huge investments and massive government initiatives world-wide are after this giant application/project. The goal is to integrate all patient's data in one securely accessible digital database. One of the benefits EMRs provide is the prevention of adverse drug-drug interactions that may happen because the patient's medical record isn't complete or the information about a drug they are taking is not captured. Unfortunately many EMR systems/applications are still not adopted; "According to a U.S. study earlier this year, fewer than a third of hospitals and well under a fifth of private-practice physicians use electronic medical records." ^[9] Examples of EMR systems that are either developed or still under development include the U.S. Natioanl Health Information Network, France's Dossier Médical Personnel, and the U.K.'s NHS Connecting for Health system.

One initiative aimed at boosting the efficiency of EMR systems while enabling them to be more "user-friendly", is the University of Maryland's Human Computer Interaction Lab's project: LifeLines for Visualizing Patient Records.

"LifeLines provides a general visualization environment for personal histories ... For a patient record, medical problems, hospitalization and medications can be represented as horizontal lines, while icons represent discrete events such as physician consultations, progress notes or tests. Line color and thickness can illustrate relationships or significance. Rescaling tools and filters allow users to focus on part of the information, revealing more details." ^[10] The application will enable physicians and healthcare providers to spot any anomalies and trends in a patient's medical history with less chance of overlooking some information. It also allows for a stream-lined access to details with it's menu-like Graphical User Interface.

Data Mining and Knowledge Discovery in Databases

With advances in algorithmic innovations, data-mining techniques now provide a way to discover and search for patterns in data stored in large data-warehouses and repositories and perform various analysis scenarios in a fast fashion. Such techniques are very useful in biomedical informatics applications and can provide a model for predicting drug-drug interactions for example; optimizing these algorithms will ensure patients' safety by preventing harmful drug-drug interactions from being prescribed in the applications like Clinical Decision Support Systems (CDSS). ^[11]

Clinical Decision Support Systems (CDSS)

Long have application like CDSS been needed for physicians finding it hard to provide case-specific medical diagnosis, advise, or prescription in stressful emergency situations. CDSSs actively seek and link the patients’ information including past medical history and diagnosis with medical knowledge systems/databases thus generating safe case-specific medical advice based on those linkages and findings. ^[12] For example, if a patient has a certain medical condition and is currently taking Vicodin, a CDSS will alert the doctor not to offer the patient Naltrexone due to the serious drug-drug interaction that takes place if both drugs were taken together. ^[13] Not only do they offer safe medical practices, CDSS also offer economic savings; when a CDSS was coupled with antimicrobial management teams at the University of Maryland Medical Center they helped realize savings of $84,194 over a three month time period. ^[14]

Computerized Physician Order Entry (CPOE)

As discussed earlier, these systems offer a way of digitizing the process of physician’s health orders accurately, safely, and promptly. Sengstack et el. said that "... order entry systems tout the ability to create a legible, complete order and apply logic-based rules to patient information to prevent errors.” ^[15] Newer CPOE incorporate physician's voice commands, hand-written instructions from tablet-PCs, etc ... ^[16] Such applications eliminate the errors that occur due to mistakes in reading or interpreting physicians handwriting and thus providing a safer and more effective treatment.

Although such systems bring a lot of benefits, have a great deal of potential, hospitals have struggled to engage physicians in using CPOE. It's hard to change habits of people or get them to use a new technology that has a steep learning curve, let alone changing the ways physicians do their work in a busy and hectic environment where errors due to misuse or time delays have zero tolerance. In other words, mastering a new way of doing their work, like using CPOEs, requires effort and writing orders electronically takes more time than just doing it the traditional way - using pen and paper. ^[17]

Clinical Workflow Systems

When processes mature and become routine to an organization, automation of these processes yields higher efficiency rates. Automation also allows for little to no room for human errors that result from information overload, or poor access to data at any given time; put in other words, it prevents poor judgments by empowering the machines to operate these process instead. Such systems are known as workflow systems that streamline communication channels and linkages between various disparate best-of-breed healthcare systems that may be in place at a healthcare facility.

They fundamentally use the same approach as commercial workflow systems, but customized to the healthcare industry.

In 2005, Andrew & Associates in collaboration with the Center for Advanced Professional Productivity conducted a survey of leading clinical workflow systems. The survey, "Workflow Management Survey: Ambulatory EHR Systems”, surveyed 52 solutions on various aspects including: ambulatory patient encounter and scheduling, visit preparation, patient pre-visit assessment, patient registration, pre-encouter care delivery, physician encounter, patient release, order and result management, patient post-visit assessment, communications, and billing. ^[18]

An example of a clinical workflow system that was evaluated in the study is e-MDs. It provides a workflow system tailored for clinical settings; the system shares information through a central database illuminating data redundancy while streamlining the processes. The system has four main modules, as depicted by figure 2: pre-visit, check-in, treatment, and checkout. ^[19]

Other examples of such systems include AcerMed, GE Healthcare, and MediTech.

Telemedicine

In today's world a need emerges for new ways of delivering healthcare services and seeking expert medical advice across physical and temporal borders. Telemedicine is the answer; it becomes especially valuable in remote areas via enabling the delivery of healthcare at a distance. Many variations of telemedicine applications across the spectrum can be found: as simple as communicating via phone or email, to robotic surgery via satellite technology.^[20] It has been a growing area of interest especially in warfare via providing urgent care to wounded soldiers in the battle fields' "hot-zones". ^[21]

NASA Extreme Environment Mission Operations (NEEMO) project, is an example of a telemedicine application. NASA and the U.S. Army are collaborating to design a surgical robots operated by human surgeons from a distance. These robots are envisioned to keep doctors and physicians off the "hot-zones" and still provide quality urgent care to injured soldiers. NEEMO 9 was an 18-day experiment exploring the use of these robots for telesurgery; with the involvement of the U.S. Navy and the National Oceanic and Atmospheric Administration, a surgeon in Ontario, Canada operated the M7 robot when it was 4 kilometers underwater in Key Largo, Florida. This simulation is significant especially considering the fact that a submarine can't surface until the end of the mission because the human body needs time to clear out the nitrogen. So if a submarine crew member had a medical emergency and needed immediate medical attention, he/she have to wait until the end of the mission to be treated.^{[22] [23]} Telemedicine will definitely be a solution to such problem.

One major contributer to this field is the U.S. Army Telemedicine and Advanced Technology Research Center (TATRC). Another example of a telemedicine program, is the Arizona Telemedicine program at the University of Arizona. In 2001, the program was awarded the "American Telemedicine President's Award : Top Honor in Telemedicine" ^[24].

Public Health and Disease Outbreak Monitoring Applications

In 2003, the World Health Organization announced 5865 probable Severe Acute Respiratory Syndrome (SARS) cases with 391 deaths that have been reported from 27 countries.^[25] In events like this, public health and disease outbreak monitoring applications have proven beneficial. Such applications and systems are used to gather, store, monitor, and track data concerning public health and disease outbreaks in a region or geographic location. Their existence is valuable due to the information they can provide about disease outbreaks in a timely manner and address any concerns or prevent any bio-terrorism efforts. Examples of such applications include the Public Health Information Network's BioSense, the University of Arizona’s BioPortal, the University of Pittsburgh's Realtime Outbreak and Disease Survailance (RODS), the U.S. Department of Homeland Security's BioWatch and National Biosurveillance Integration Systems ^{[26] [27]}.

Human Genome Project

The Human Genome Project,was a 13-year effort coordinated by the U.S. Department of Energy and the U.S. National Institutes of Health. The project identified all the genes in the human DNA, the sequences of chemical-base pairs that make them, and provided a platform for data storage and analysis. ^[28] This platform is available for utilization by commercial and non-commercial entities for the development of robust medical applications and advance research projects such as predicting a person's disease susceptibility based on their genetic predisposition. The project's outcomes recognized the major role of bioinformatics; project leaders said that "one of the key research areas was bioinformatics. Without the annotation provided via bioinformatics, the information gleaned from the HGP is not very useful." ^[29]. Biomedical informatics has given researchers the ability to conduct genetic analysis that can predict future health conditions to foster personalized medical care/treatments, do informed family planning, appropriate lifestyle changes, and find cures for diseases. ^[30] In a collaboration between Oak Ridge National Laboratory and the University of Southern California, new hope for the blind started to shine; a protein found in spinach may be useful in restoring sight to them. ^[31]

Another project the U.S. Department of Energy is sponsoring is the Microbial Genome Program that spun-off the Human Genome Project. In 2001, using biomedical informatics to analyze genome sequences of a certain microbe, scientist from Belgium and the Brookhaven National Laboratory are collaborating to engineer bacteria that cleans contaminants such as microbes from soil. ^[32]

The Cancer Genome Atlas

Built on the same principals, but addresses specific problems related to cancer, is the The Cancer Genome Atlas (TCGA). TCGA can be viewed as an extension of the Human Genome Project while capturing and categorizing genetic changes associated with cancer in humans. ^[33]

cancer Biomedical Informatics Grid™( caBIG™)

The grid provides a working space for researchers, physicians, and cancer patients through collaboration via pooling various resources from heterogeneous environments . The initiative built a community of researchers and professionals that developed and released a variety of bioinformatic tools. Such initiatives and systems will accelerate the problem-solving process while fostering innovation and cutting-edge research to the benefit to all human-kind.^[34] caBIG™ "represents a colossal initiative involving virtually all aspects of cancer research and may help to transform the way cancer research is conducted and data are shared." ^[35] This is done by leveraging the tools and computing capacity grids can offer to solve complex research quests.

Radio Frequency Identification for Biomedical Informatics

Radio Frequency Identification, or RFID for short, is "a technology that incorporates the use of electromagnetic or electrostatic coupling in the radio frequency (RF) portion of the electromagnetic spectrum to uniquely identify an object, animal, or person." ^[36] Researches from RSA Laboratories and Massachusetts Institute of Technology, Ari Juels and Stephen Weis, described how RFID technologies work. "RFID systems consist of two main components: tags and readers. Tags are radio transponders attached to physical objects. Radio transceivers, or readers, query these tags for some (potentially unique) identifying information about the objects to which tags are attached." ^[37] Applications of RFID are growing exponentially and the technology stores a lot of hope. Some of the RFID applications for biomedical informatics include, patient, personnel, and asset tracking; identification and verification; elderly emergency dispatch; medical monitoring.

RFID for Patient, Personnel and Asset Tracking

• In 2004VeriChip™: was the first approved RFID tag by the U.S. Food and Drug Administration as a human-implantable RFID microchip for medical applications. ^[38] The chip is implantable in the back of the upper right arm of a person. VeriChip™ offers several applications including: patient identification, infant protection, wander prevention, and medical asset tracking. ^[39] Figure 4 illustrates the implantation and actual size of VeriChip™.

• NEC Unified Solutions announced in February 2007 the launch of WiFi RFID solutions embedded in it's Univerge system to help large healthcare organizations and hospitals to track their patients, staff, and medical equipment. ^[40]

RFID for ID verification and Recognition

With the flexibility in programing RFID tags, it has become easy to encode them for security purposes of medical facilities and in healthcare organizations. If an unauthorized person or patient wanders into a restricted area, RFID tags will be detected by RFID readers in place and alert the staff at the facility of such violation to take the appropriate action. Examples of systems that have such capability include VeriChip™ and Univerge discussed earlier.

RFID for Senior Citizens

• Smarthouse was developed by Medical Automation Research Center at the University of Virginia to help in-home, non-invasive monitoring of senior citizens. The system is composed of "a suite of low-cost, non-invasive sensors (strictly no cameras or microphones), and a data logging and communications module, in addition to an integrated data management system, linked to the Internet." ^[41] Figure 5 shows a diagram of the system's implemntation and how it would integrate in a floor plan of a building. When Smarthouse detects a low/high pluse, fall alerts, or even stove temperature alerts, it dials the caregivers' pagers or mobile phone and leaves a numeric message with the senior citizen's ID and alert code. ^[42]

RFID for Medical Monitoring

In 2006, Eastman Kodak Company filed for a patent at the U.S. Patent and Trademark Office for an edible and digestible RFID enabled tag used to monitor the ingestion of medicines. Figure 6 illustrates the patent for RFID device. Kodak is envisioning that the new RFID system will be used for monitoring various aspects of a patient medical condition. One of the projections of how the system will be used is that to monitor for example the digestive track when administering medication. The company also claims that it can be embedded in artificial hip joints and knees so that when the tag wears-out, it can provide a signal to the physician that a surgery is needed.^[43]

Click here to access the documentations and picture files of the patent.

Case Study 1 : Biomedical Informatics on the Web

Introduction

Just like commerce, biomedical informatics has been affected by the Internet revolution in today's Information Technology era. Because the Internet world is fast-paced and and constantly on the move, technological advances are common and people, entities, and organizations are racing to adapt them. As we move into the era of web 2.0 the way we use the Internet and web changes as well. We start to notice more web services, blogs, the creation of online communities and collaboration forums, MySpaces, Wikipedias, and so forth.^[44]

Biomedical informatics are also infulanced by the movement towards web 2.0 technologies. Researcher, physicians, and patients are now blogging on their own virtual spaces online about their research, new applications of biomedical informatics, or even their experience at a clinic or a hospital. It's no surprise when one navigates from their web browser to YouTube.com, a multimedia website that allows individuals to upload their own multimedia content and make it available to the public online, to find information on biomedical informatics in a media-rich format. MedGadget.com is another example of this move towards web 2.0; this online blog lists the latest inventions and newest medical applications, systems, devices, and gadgets. Matthew Herper, a staff writer at Forbes magazine, had to say this about medical blogs "there are dozens of Web logs about the practice of medicine written by doctors, scientists, and other health professionals. Almost all of these blogs are shockingly good, providing riveting reading about how people are diagnosed, treated and sometimes cured." ^[45]

Their Applications and How they are Used

The above-mentioned generation of biomedical informatics on the web applications are being observed, used, and utilized in various ways and in many forms.

Blogs

Medical blogs are used by different people, including physicians and patients, to share their experiences, opinions, and even feelings. An example blog is NHS Doctor Blog. Blogs are also utilized as an "underground" or unofficial source of information of concepts and developing medical technologies and devices that are not out in the field yet. An example blog is MedGadget.com.

e-Portals

e-Portals on the other hand act as electronic information portals where individuals can obtain and learn more about a certain medical topic or a particular disease. WebMD is a great example of an online information portal that provides and presents medical information to the general public in a simplified language. Another good example is the * Genetics Home Reference; after the completion of the human genome map in 2003, the U.S. National Library of Medicine developed an online information portal that makes "the connection between genetics and disease more understandable for the general public." ^[46]

e-Multimedia

Wether the purpose is to educate over a distance, advertise, or simply express an opinion and share an experience, online multimedia can go a long way. Such multimedia is available in a variety of topics in many flavors to anyone who has an internet connection (with a decent speed; DSL or higher) to learn more about a certain topic. It truly caters to visual learners, or people who rather watch a YouTube video in almost real time instead of reading a book. Professional and academic seminars or conferences can be made available on the web too. One example of an entity capitalizing on e-multimedia for biomedical informatics is the Stanford Center for Biomedical Ethics. The center produces films and documentaries that educate physicians and others on ethical and social issues concerning the medicine and biomedicine fields.

Online Support Groups

Joining support groups for patients with long-term illnesses, such as breast cancer patients, is very helpful in the treatment process and overall well-being of the patients. These support groups offer their members various benefits including emotional support, informational and practical support benefits. ^[47]

Amongst the famous online support groups, are those who are for breast cancer patients. A good example is the Breast Cancer Support. The site provides a virtual place where cancer survivors can seek support, information, or find someone else with their same diagnosis. ^[48] People can register with the site or simply post anonymously to the various chat boards like surgery related or chemotherapy related issues. The website is also a venue to inform breast cancer survivors and others of local events that take place such as fundraising or informational events. Another example is the Pink Link Foundation where breast cancer patients and survivors can be linked virtually. It was established in 2005 by Victoria Tashman. The foundation provides an online database that creates the link between patients and survivors and is built on the "mentor-patient" relationship to effectively combat the disease.

What are they Addressing/Solving

In this era of technology advances, the internet age, and the web 2.0 era, biomedical informatics needed to go this route for them to be accessible and available to all people. Having to conduct seminars or local support groups has it's limitations in terms of time, place, and resource availability. Having biomedical informatics applications on the web, wrapped in a new creative electronic way, made it easy for people to connect anonymously online/offline in a rich multimedia environment that can be personalized and tailored. Access to the information or applications is easy and only requires a personal computer with an internet connection; this means that anybody from any part of the globe at any given time can access blogs, online support groups, and so forth.

Benefits of these New Applications

These new biomedical applications on the web have several benefits to their users.

• Accessibility: All you need to access these applications and utilize them is a computer with an internet connection that has a decent speed like DSL. You can access these applications from anywhere on the globe at any time you find convenient. Accessibility has also been simplified for computers as well; cross-platform communication in heterogeneous environments is now possible since such applications are on the web.
• Wide Reach: Because these applications are accessible from the web, many people and users can be reached thus crossing spatial and temporal borders. People will not be limited to a certain geographic area to attend a conference or learn about a certain medical issue.
• Rich Multimedia: Since all the applications are digital, and with the advances of technology and programming languages, it has become easier to provide users a rich multimedia experience that tailors for different types of audiences that have various learning styles.
• Flexibility: Because these applications can be accessed through the web, scientist or researchers, and even enthusiast users, can work from home or the local coffee shop while sipping some coffee instead of driving to work or central location where these applications may reside. Flexibility in terms of access and collaboration truly harnesses innovation and rapid developments.
• The True Voice: Because individuals and groups can create blogs or post biomedical comments online and anonymously, if they choose to, expressing honest personal opinions and stories has grown. There is little room for fear of being prosecuted or discriminated against (assuming the comment is not a national security threat or similar) for expressing personal opinions, feelings, or experiences.
• Collaboration: Exposing your biomedical application on the web and sharing it with the millions of users around the entire globe is a great way to seek collaboration. People who have the same interest as you do can easily get a hold of you and communicate as to what they would like to collaborate on.
• Creativity: Since ideas and examples for biomedical applications are available online, people can evaluate them, build on-top of them, or customize them to their specific needs thus stimulating creativity. For example, after reviewing the human genome project, a person or group of people may decide to adopt the same approach, but this time for plant genomes to create evergreen trees.

Challenges of these New Applications

• User Training and Education: Since these applications are on the web, users must be computer literate, have some experience navigating/browesing the web, skillful at searching for the appropriate applications that meet their needs, and know how to navigate and operate the specific application. There is a learning curve here, and maybe a turn-off for so many people especially with applications that are very powerful, flexible, and by nature very complex.
• Cost of Ownership: For developers, expenditures for the infrastructure, the underlying design, and programming of the applications can be a financial burden. Expenses include the underlying hardware (servers, switches, ...), software development, and ongoing maintenance and support for the hardware and software. For users, owning a computer or having access to one that is connected to a reasonably fast internet connection can be difficult especially in developing countries. The question then remains: who pays for all these expenses? Do developers charge users for using their applications, or do governments subsidize these entities to support them? Should consumers bear the costs and on what basis (per usage, one-time fee, ... etc)
• Credibility: Because people can post and upload their applications freely online and without any evolution mechanisms, it becomes difficult to establish credibility and liability. It is especially hard to evaluate those applications that are anonymous or are pushed for political agendas or sales espionage purposes.

Resources: Biomedical Informatics on the Web

Below is a list of resources for biomedical informatics on the web:

Blogs

• MedGadget.com
• NHS Doctor Blog : "Best Health Policies/Ethics Weblog 2006 - MedGadget.com"
• Health IT Guy Blog : "Best Medical Technologies/Informatics Weblog 2006 - MedGadget.com"
• Anxiety, Addiction and Depression Treatments : "Best Clinical Sciences Weblog 2006 - MedGadget.com"
• Intueri.org : "Special Mention 2006 - MedGadget.com"

• Biological Informatics Blog

e-Multimedia

• Stanford Center for Biomedical Ethics - Bioethics and Film
  • Hold Your Breath Series
  • Worlds Apart Series
  • Grave Words Series
  • Vanishing Line Series
• YouTube.com
  • YouTube.com Video: DNA and the Transcription Process
  • YouTube.com Video: DNA and Microarrays definitions

e-Portals

• Genetics Home Reference
• WebMD
• Dr. Koop.com

Online Support Groups

• Healthy Place Online Community
• University of Cape Town's HIV/AIDS online support group
• Breast Cancer Support
• Pink Link Foundation

Summary

Biomedical informatics on the web are truly the new trend that is changing the way applications are designed and used in biomedical informatics. Such applications include biomedical blogs, support groups, and multimedia websites. They offer many valuable benefits such as accessibility, and a platform for collaborations between people across borders and time zones. Despite their great benefits, one must also closely consider their challenges such as credibility and cost of ownership. All in all, we believe biomedical informatics applications on the web have yet more and more to offer and that they are definitely the future in this field.

Case Study 2 : Service-Oriented Architecture applied for Biomedical Informatics

Introduction

In the computing world, Service-Oriented Architecture, SOA, has become a buzz word. SOA can be defined as: "the architectural style that supports loosely coupled services to enable business flexibility in an interoperable, technology-agnostic manner. SOA consists of a composite set of business-aligned services that support a flexible and dynamically reconfigurable end-to-end business processes realization using interface-based service descriptions." ^[49]

Recently, there have been several discussions of how may SOA benefit the medical and biomedical arena and revolutionize the way these functions operate. In this case study, we will review SOA for biomedical informatics, how might it be utilized, it's benefits and challenges, and conclude with closing remarks.

How would SOA be utilized for Biomedical Informatics

Since SOA is implementable over the web and allows for easy access to services in heterogeneous environments it has great potentials for biomedical informatics that have grown rapidly with dozens of standards and on various platforms. Omar and Taleb-Bendiab discussed a way of how SOA can be utilized for electronic health monitoring. ^[50] They suggest that hospitals implement such architecture to support their patient monitoring activities. The architecture will monitor the patient's vital life signs, such as temperature and blood pressure, and alert the nursing staff or caregivers once abnormalities in the patient's condition occur. Figure 7 summarizes how the system functions:

Another example where SOA may be utilized for biomedical informatics applications is in electronic medical expert systems. Once a patient's MRI is taken, the image is then compared to a rich external data-warehouse that is maintained by a separate entity and return the result as to what the patient's condition is in a form of a service. This definitely encourages and enables collaboration and information sharing in a heterogeneous environment. ^[51]

Furthermore, SOA can be used to leverage the use of grid computing technologies to solve complex biomedical informatics problems that require vast amounts of computing power. Grid computing is analogues to that of the power grid technology, where functionality can be leveraged and economies of scale realized by connecting several nodes together that communicate in a shared standard to solve complex problems via dividing the tasks and aggregating the results at the requesting node. ^[52] SOA and Grid technology seem to be promising for applications like tissue-based diagnostics. Such combination can dig deep in molecular genetics of the patient, acquire images of the tissue at hand and run recognition algorithms for image analysis and communicate accurate results electronically in an efficient and fast manner thus ensuring correct diagnosis of the patient's condition while ensuring the patient's safety. Figure 8 illustrates how such Grid architecture combined with SOA would function:

Issues that are Addressed with SOA

With SOA, many issues related to biomedical informatics applications can be resolved and performance improvements can be realized. Open and shared standards is a key characteristic of SOA. This allows for intercommunication between different systems and platforms that have varying architectures. Since there are many standards in biomedical informatics, such as HL7; DICOM; PHINMS, a standard means of communicating between systems that use one of the previously mentioned standards or others is essential to ensure interoperability between systems in order to yield system synergy along with larger gains and benefits. SOA also simplifies biomedical software design and implementation via decomposing the processes and problems into smaller and more manageable units called services. Moreover, SOA allows for object/service reusability and thus saving limited IT resources. In addition, SOA improves the adaptability to the changing business or functional requirements that are often rapidly changing in the life sciences and biomedical informatics fields. ^[53]

Benefits of SOA in Biomedical Informatics

• Reusability of Services/Objects: Since SOA allows for the reuse of code/objects, developers do not need to spend lengthy time in development of the same code/object thus saving time and costs. It also ensure consistency since the same service will be called from various places.
• Interoperability in Heterogeneous Environments: Because SOA is built on the idea of open communication with a shared standard, the most popular being XML, allowing for virtually anyone using any system to call the service over a web-based architecture. This is very important since many applications have been developed and invested in heavily already; SOA enables these applications to communicate and collaborate with each other despite the fact that they came from different environments.
• Scope and Flexibility: SOA also allows for scope redefinition, whenever needed, very easily. To re-scope, the service is changed once and the change will carry on and is true for every one who calls that particular service. This also translates to more flexibility to changing requirements or scope. If the service is now to be utilized by other entities and requires a more/less detailed information, it can be done by modifying the targeted service.
• Accessibility/Availability of Services: Having the services published in a networked environment or over the internet, makes it easy to access the services and utilize them. All that's needed is a published dictionary of all the available services and the communication protocols governing them.
• Complexity Encapsulation: SOA provides the users with ease of use since it only displays the outcomes of the services and hides the complexity of the underlying operations that take place when the service is called. The users needs only to discover the service from the dictionary, call it, and the rest will take place accordingly.

Challenges to SOA in Biomedical Informatics

• Liability: While SOA can be utilized in CDSS to enhance the decisions on patient treatment to deliver a safe medical practice, liability becomes a concern especially when the service is called from an external entity to the service requesting organization. Assume that the external entity was not rigorously and appropriately maintaining/updating the service "drug-drug interactions", the service requesting organization can be in huge liability situation if it made a decision on outdated information provided by the service and prescribed a new drug that has adverse effects when combined with the current drug the patient is taking.
• Costing/Billing: It is projected that services for biomedical applications will be commercialized in the future which raises the question on how will service requesters be charged? A flat rate or per-use? Who will in the end bear the costs; patients, hospitals, governments, ...?
• Privacy and Security: Because these services are accessed from external users, and potentially open for everybody, there is a concern that these services may be abused and the privacy and security of patient's data maybe violated. Consider an outsider gains access to a service in place that reports to Department of Health outbreak cases of the Avian flu and in turn discovers the identifies of the outbreak patients. Or in the case of electronic health monitoring know the location of a particular patient at a given time.

Summary

SOA is a new technological approach that has some challenges that yet have to be addressed, but carries a lot of potential in the future. SOA allows for cross-platform communication between systems in heterogeneous environments thus allowing for collaboration and building-on existing applications and architectures. Examples of it's applications include electronic health monitoring and tissue-based diagnosis in a Grid/SOA computing architecture. Some of the challenges SOA currently faces in biomedical informatics include liability of the service providing entity, costs and billing associated with commercialization, and the privacy and security of the patients data. In our humble opinion, we believe that SOA has far more potential in the future for biomedical informatics applications and will be the standard when it comes to software development for biomedical applications on the web.

Overall Concluding Remarks

As discussed above, biomedical informatics applications save lives; they allow for safer medical practices, collaboration, and information sharing. Applications like EMR and the Human Genome Project have been invested in and explored for quite a period of time. Newer applications like RFID, biomedical informatics on the web, and Service-Oriented Architecture for biomedical informatics have huge potentials and are promising. Along with the merits, these applications also face challenges to implementing them. Such challenges can include privacy and security concerns, costs of ownership, ... etc. The information technology tools available nowadays can heavily leverage the ability of biomedical informatics and what they can provide. As technology advances, so will biomedical informatics applications.

References

1. ↑ Leape, L. (1994) “Error in medicine” Journal of the American Medical Association 272(23): 1851-57
2. ↑ Bower, A., (2005). “Katrina's Lingering Medical Nightmare” Time Magazine. http://www.time.com/time/nation/article/0,8599,1107826,00.html
3. ↑ Shortliffe, E., et al. (2001). Medical Informatics: Computer Applications in Health Care and Biomedicine. Second ed. New York, Springer-Verlag.
4. ↑ Greenes R, Shortliffe E. (1990) “Medical informatics: An emerging academic discipline and institutional priority”. Journal of the American Medical Association, 1990(263). 1114-1120.
5. ↑ Bell, T., "Medical Records: From Clipboard To Point-and-Click" IEEE Institute. <http://www.theinstitute.ieee.org/portal/site/tionline/menuitem.130a3558587d56e8fb2275875bac26c8/index.jsp?&pName=institute_level1_article&TheCat=2201&article=tionline/legacy/inst2005/dec05/12w.fMedical_Records.xml&>
6. ↑ Horsky, J., Kuperman, G., Patel, V., (2005). “Comprehensive Analysis of a Medication Dosing Error Related to CPOE” Journal of the American Medical Informatics Association, 12(4). 365-376.
7. ↑ Kuperman, G., Bobb, A., Payne, T., Avery, A., Gandhi, T., Burns, G. Classen, D., Bates, D., (2007). “Medication-related Clinical Decision Support in Computerized Provider Order Entry Systems: A Review” Journal of the American Medical Informatics Association, 14(1). 29-40.
8. ↑ Arsanjani, A., Borges, B., Holley, K., (2004). “Service-Oriented Architecture” DMReview, 12 Nov. 2004. <http://www.dmreview.com/article_sub.cfm?articleId=1013602>
9. ↑ Bell, T., "Medical Records: From Clipboard To Point-and-Click" IEEE Institute. <http://www.theinstitute.ieee.org/portal/site/tionline/menuitem.130a3558587d56e8fb2275875bac26c8/index.jsp?&pName=institute_level1_article&TheCat=2201&article=tionline/legacy/inst2005/dec05/12w.fMedical_Records.xml&>
10. ↑ (1999). "LifeLines for Visualizing Patient Record". University of Maryland - Human Computer Interaction Lab. Accessed 3/30/07. <http://www.cs.umd.edu/hcil/lifelines/>
11. ↑ Boyce, R., Collins, C., Horn, J.R., and Kalet, I.J. "Modeling Drug Mechanism Knowledge Using Evidence and Truth Maintenance". IEEE Transactions on Information Technology in Biomedicine, in press, 2007.
12. ↑ Teich, J., Osheroff, J., Pifer, E., Sittig, D., Jenders, R., (2005). “Clinical Decision Support in Electronic Prescribing: Recommendations and an Action Plan - Report of the Joint Clinical Decision Support Workgroup” Journal of the American Medical Informatics Association, 12(4). 365-376.
13. ↑ Drugs and Treatments - Vicodin Oral http://www.webmd.com/drugs/drug-3459-Vicodin+Oral.aspx?
14. ↑ McGregor, J., Weekes, E., Forrest, G., Standiford, H., Perencevich, E., Furuno, J., Harris, A., (2006). “Impact of a Computerized Clinical Decision Support System on Reducing Inappropriate Antimicrobial Use: A Randomized Controlled Trial” Journal of the American Medical Informatics Association, 13(4). 378-384.
15. ↑ Sengstack, P., Gugerty, B., (2002). "CPOE Systems: Success Factors and Implementation Issues". Journal of Healthcare Information Management. 18(1): 36-45
16. ↑ Kaushal, R., Bates, D., (2001), "Computerized Physician Order Entry (CPOE) with Clinical Decision Support Systems (CDSSs)". Agency for Healthcare Research and Quality. ,July 20, 2001.
17. ↑ Metzger, J., Fortin, J. (2003). "Computerized Physician Order Entry in Community Hospitals: Lessons from the Field" California Healthcare Foundation. <http://www.chcf.org/documents/hospitals/CPOECommHospCorrected.pdf>
18. ↑ Andrew, W., Bruegel, R., (2005). "2005 EHR Workflow Management Review and Survey Results". ADVANCE for Healthcare Information Executives. Accessed 4/3/07. <http://health-care-it.advanceweb.com/Common/editorial/editorial.aspx?CTIID=1620>
19. ↑ (2006). "e-MDs Solution Series clinical Workflow". e-MDs Solutions. Accessed 4/3/07. <http://www.e-mds.com/solutions/clinical_workflow.html>
20. ↑ "Telemedicine: A Brief Overview". American Telemidicne Association. June 23, 1999 http://www.atmeda.org/news/overview.htm
21. ↑ Rosen, J., Hannaford, B., (2006). "Doc at a Distance". IEEE Spectrum. http://www.spectrum.ieee.org/oct06/4667
22. ↑ Burnett, M., (2007), "Robotic Relief". Military Medical Technology - Online Edition. (11):1 http://www.military-medical-technology.com/article.cfm?DocID=1886
23. ↑ (2006). "Mission News:NASA's Undersea Crew is Heads Above Water". NASA. Accessed 3/10/07 <http://www.nasa.gov/mission_pages/NEEMO/NEEMO9/index.html>
24. ↑ (2001). "Arizona Telemedicine Program: Awards. University of Arizona. Accessed 3/12/07 <http://www.telemedicine.arizona.edu/awards.html>
25. ↑ (2003) " Severe Acute Respiratory Syndrome (SARS) - Multi-country outbreak - Update 44". World Health Organization. <http://www.who.int/csr/don/2003_05_01/en/index.html>
26. ↑ (2006) "Fact Sheet: Protecting the Homeland Post September 11" Department of Homeland Security. <http://www.dhs.gov/xnews/releases/pr_1158339830666.shtm>
27. ↑ "Prevent Proliferation of Weapons of Mass Destruction" The White House. <http://www.whitehouse.gov/nsc/waronterror/2006/sectionV.html>
28. ↑ (2005). “The Human Genome Project” Oak Ridge National Laboratory. http://www.ornl.gov/sci/techresources/Human_Genome/project/about.shtml
29. ↑ (2004) "Bioinformatics" The Department of Energy. <http://www.ornl.gov/sci/techresources/Human_Genome/research/informatics.shtml>
30. ↑ (2007) "Research". Standford Center for Biomedical Ethics. <http://scbe.stanford.edu/research/>
31. ↑ Krause, C., (2001) "Spinach Protein Offers New Hope for the Blind" Oak Ridge National Laboratory Review 34:(2)
32. ↑ (2001) "Scientists Decode Genes of Microbe that Thrives in Toxic Metals". Department of Energy. 11:3-4. <http://www.ornl.gov/sci/techresources/Human_Genome/publicat/hgn/v11n3/18ralsto.shtml>
33. ↑ Hanauer, D., Rhodes, D., Sinha-Kumar, C., Chinnaiyan, A., (2007) "Bioinformatics Approaches in the Study of Cancer". Current Molecular Medicine. 2007:133-141
34. ↑ (2007). “cancer Biomedical Informatics Grid™” National Cancer Institute. https://cabig.nci.nih.gov/
35. ↑ Hanauer, D., Rhodes, D., Sinha-Kumar, C., Chinnaiyan, A., (2007) "Bioinformatics Approaches in the Study of Cancer". Current Molecular Medicine. 2007:133-141
36. ↑ http://www.directedi.com/press/edi-glossary.html#RFID%20(radio%20frequency%20identification)
37. ↑ Juels, A., Weis, S., (2006) "Defining Strong Privacy for RFID", unpublished - work in progress <http://saweis.net/pdfs/JuelsWeis-RFID-Privacy.pdf>
38. ↑ (2006). "Company". Verichip Corporation. Accessed 3/11/07 <http://www.verichipcorp.com/company.html>
39. ↑ (2006). "Solutions". Verichip Corporation. Accessed 3/11/07 <http://www.verichipcorp.com/solutions.html>
40. ↑ (2007). "NEC Adds Health-care Functionality to Its Wi-Fi". RFID Journal. Accessed 3/11/07 <http://www.rfidjournal.com/article/articleview/3108/1/1/>
41. ↑ http://marc.med.virginia.edu/projects_smarthomemonitor.html
42. ↑ Alwan, M., Dalal, S., Mack, D., Kell, S., Turner, B., Leachtenauer, J., Felder, R. (2006). " Impact of monitoring technology in assisted living: outcome pilot" IEEE Transactions on Information Technology in Biomedicine. 10(1):192-198
43. ↑ http://www.newscientisttech.com/article/dn11162-invention-edible-rfid.html?DCMP=Matt_Sparkes&nsref=edible-rfid
44. ↑ O'Reilly, T., (2005). "What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software" http://facweb.cti.depaul.edu/jnowotarski/se425/What%20Is%20Web%202%20point%200.pdf
45. ↑ Herper, M., (2003). "Best Medical Blogs" http://www.forbes.com/2003/10/03/cx_mh_1003medblogs.html
46. ↑ Spatz, M., (2004). "Genetics Home Reference". Medical Library Association. 92(2): 282–283 <http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=385319>
47. ↑ Till, J., (2003). "Evaluation of support groups for women with breast cancer: importance of the navigator role". Health and Quality of Life Outcomes. 1:16
48. ↑ Breast Cancer Support Group. Accesses 16 march 2007. www.bcsupport.org
49. ↑ Arsanjani, A., Borges, B., Holley, K., (2004). “Service-Oriented Architecture” DMReview, 12 Nov. 2004. <http://www.dmreview.com/article_sub.cfm?articleId=1013602>
50. ↑ Omar, W., Taleb-Bendiab, A., (2006) "E-Health Support Services Based on Service-Oriented Architecture". IEEE IT Pro magazine; 2(06): 35-41.
51. ↑ Thanh, D., Jorstad, I., (2005). "A Service-Oriented Architecture Framework for Mobile Services". IEEE Proceedings of the Advanced Industrial Conference on Telecommunications/Service Assurance with Partial and Intermittent Resources Conference/ELearning on Telecommunications Workshop; (2005):1-6.
52. ↑ Görtler, J., Berghoff, M., Kayser, M., Kayser, K., (2006). "Grid technology in tissue-based diagnosis: fundamentals and potential developments". Diagnostic Pathology; 1:23.
53. ↑ Nadkarni, P., Miller, R., (2007). "Service-oriented Architecture in Medical Software: Promises and Perils" JAMIA; 14:244 –246.

Online Resources

• American Medical Informatics Association (AMIA)
• Journal of American Medical Informatics Association (JAMIA)
• Journal of American Medical Association (JAMA)
• The Bioinformatics Organization, Inc.
• United States Department of Health and Human Services
• Biomedical Engineering Online
• Healthcare IT News

Questions for Discussion

• Can new IT concepts such as Enterprise Application Integration, or Business Process Networks help advance biomedical informatics applications? If so, how?

• Human gene data are being evaluated for inclusion in future implementations of EMR systems for the benefits they bring. What other biomedical information may be helpful if coupled with EMR?

• How can creativity be encourage in biomedical application development? Are open and shared standards the solution? What are their limitation (if any)?

• What other application principals/analogies/idea/archiectures can be applied for developing or tailoring biomedical informatics applications?