The MUMPS (M) programming language has long been used in health-care. Its history and other details can be found here. Version 2.0 of a compiler to translate Mumps (M) to C is now available. Translations such as this make it easy for programs written in MUMPS to run on any computer. It implements
a substantial sub-set of the 1995 Mumps (M)
standard. This compiler includes support for limited indirection (“@” level
only at present), B-Tree based global arrays,
SQL (PostgreSQL) and extensions. In addition, it has code to facilitate development of web server-side applications. Details, download and documentation can be found here.
Monthly Archives: July 2000
GNUMed Alpha Server Possible in October 2000
Dr. Herb Horst of the GNUmed project has set out a tentative timetable for the availability of GNUMed “Provided people keep their promises…” The available schedule follows, highlights are Alpha releases in mid-October 2000 for server, December for client. Beta release in February 2001. “…looking for more developers to join, especially SQL wizards…” Note for US readers, the dates are written in Aussie DdMmYYYY format.
10.08.2000 alpha test SQL meta file: most entities and relations defined,
script to create an empty GNUMed database plus fully functional tables with
ICD-9/10 and other codes, some drug info, zip codes etc.
01.10.2000 alpha test server / API: working transaction server and most
logic on Interbase server functional. API for clients defined. From here
onwards it should be possible to write already functional clients
01.12.2000 first GUI client in alpha preview available: From here onwards
the system is theoretically functional. Design studies (executable ones)
exist already today but do not connect to the database yet.
01.02.2001 first beta test version (client and server)
Whooo-hooo. I hope we can make it until then.
The GNUMed project is looking for more developers to join, especially SQL wizards. The more people help, the earlier we can release. Looks as if funding will be available in the near future.
Stallman: Free Software Not Anti-Business
Douglas Carnall has an interview with the patriarch of free software Richard Stallman whom he interviewed at the recent Logiciel Libre conference in Bordeaux, France. An excerpt:
“Free software is not anti-business, says Stallman, it just changes the shape of the businesses that are possible. There are businesses to be made in the supply of physical
copies of software, in accrediting and guaranteeing particular installations of the software, and providing support and training for users of particular systems. But what should
not be done is to make money from restricting the supply of software through copyrights, patents, or trade secrecy agreements.”
Feature: Medical Projects at O’Reilly Open Source Convention
Several open source medical projects where discussed at the recent O’Reilly Open Source Convention in Monterey, California, July 17-20. Highlights were demos and secure architecture information on the GNUMed project by Horst Herb M.D. and David Guest M.D. from Australia as well as TK_familypractice from California with its XML based clinical modules and data mining which was represented by myself. Read on for a summary of the convention.
Dr. Horst Herb had a state of the project for GNUMed, saying that the GUI interface is designed and the overall program architecture is established. At present Herb and his team of developers are using Interbase and are working on a very sophisticated intermediate level server to provide encryption as well as other services necessary to ensure privacy and data integrity for patients as well as provide medico-legal protection. Complete logging of every transaction is being provided to ensure the highest level of data security, a foremost goal of the project. Dr. Herb educated the open source programmers in
attendance about the special needs in the medical information technology environment and is actively recruiting programming talent internationally for the project. He outlined his projected timetable for development. The complete elimination of paper
in the medical practice environment is the eventual goal as well as providing better quality records for both patient and provider.
A BOF (Birds of a Feather) session was held on Wedsday evening attended by a small but well focused group of attendees. Some were physicians but some were programmers from organizations in the medical field (Kaiser) as well as business representatives of some health related companies (Merck) which was encouraging.
Tkfp was demonstrated successfully on the O’Reilly RedHat 6.2 conference room computer after being downloaded off the net about an hour before the session and installed using it’s built-in install script. It ran right out of the box and was able to show prescription writing and display functions and demographic information manager that are all XML based, as well as the HTML (soon to be XML) based progress note generation features. The drug interaction scripts
that data-mined the Dr. Koop, PlanetRx and Medical Letter web sites for drug interactions was demonstrated. One of the attendees had worked on
the Dr. Koop site and commented that Dr. Koop’s developers would be “quite taken aback” (or maybe it was “appalled”) that it was being used in
that way, although he seemed to agree it was a cool idea. The MedMapper clinical decision making algorithms that integrate into the chart notes
were also shown.
Bernard Glassman M.D. of of the National Cancer Institute and Linda Lamb from O’Reilly headed a BOF session on Monday night about their ideas for patient-centric internet interfaces so patients and their families can
access appropriate information on their condition from reliable medical sources. The emphasis was on obtaining treatment protocol information
on cancer, but the ideas could be widely applied in the medical world. Lamb pointed out that O’Reilly publishes a line of patient guides
on various medical topics and is very interested in the medical world, as well as being the premier supporter of open source software in the
publishing world. Is there a possiblity of some connections being made there?
We were also able to get in some plugs for the other open source projects including – Arachne Circare Docscope EHCR Freemed FreePM GEHR
LAMDI LinuDent Littlefish Telemed OpenGalen OSHCA Prorec qRx Quick Quack Medical Manager STAR VistAXML Medical Dict.
Hopefully some of the open source coding talent attending the conference became more aware
of these projects and the opportunity to work on them with us in the medical world.
StarOffice 6.0 to be Open Source
It’s official: Sun will be open sourcing StarOffice 6.0 on October 13th, 2000 according to an article in Yahoo News business section. Sun Vice President Marco Boerries is quoted as saying
Sun’s open-sourcing of StarOffice Suite is the single largest open-source software contribution in GPL history and it adds a
key application suite to the open source portfolio,
Many other luminaries including Miguel Icaza of Gnome fame and Andy Hertzfeld are quoted. At the time of this writing, a Front Page article on Linux Weekly News has in depth coverage of why this is important.
By open sourcing StarOffice, Sun opens the door for any developer to modify, extend or fix problems in the actual program instructions of StarOffice. This should lead to rapid improvement and adoption of this already excellent Office software suite.
With this move, Sun hopes to break the lock that Microsoft currently has on Office suite software.
Finally: Clinician Feedback Via the Net
A Medscape.com (free login required) article discusses the Clinical Assessment of the
Reliability of the Examination (CARE) project which can be found at www.CareStudy.com. This project takes square aim at a problem many clinicians first encounter in medical school: horrendously long lists of possible signs and symptoms for a particular disease with no guidance at all as to which signs and symptoms are important and can accurately diagnose a condition. The article discusses how a textbook search of a frequently encountered disease Chronic Obstructive Airway Disease results in 40 different possible signs and symptoms. After receiving data rapidly over the Internet from actual practitioners, the CARE project was able to narrow the most important down to just four signs and symptoms. These symptoms will diagnose the condition with great accuracy and without lab tests.
The difference with the CARE approach and that of traditional clinical studies, is that CARE studies have obtained results at a 20 fold higher rate than traditional high-quality research. Traditional investigations can take years and still not result in precise criteria for accurate diagnosis. Using the Internet, groups of practicing clinicians can enter actual outcome data in only a few minutes per patient. The data is automatically collated on the CARE site, greatly increasing the speed of obtaining clinical data and results.
Email Impersonation: A Challenge for Health Care Organizations & Providers
I received an email solicitation recently from myself, offering me the opportunity to save 75% off my Dream Vacation. I have now joined a growing list of Netizens who have had their email addresses hijacked by a spammer. I’m told by those in-the-know that the practice of forging email headers to impersonate others is growing, as it is quite easy to do. Spammers that practice this form of deception do so to give the false impression of legitimacy to the scams that they are running. The capability to impersonate another person via email places health care organizations that use email for business and health care delivery at risk. Are there solutions?
Possibly. Companies like Disappearing, Inc., Critical Path and Infraworks offer new products that feature secure messaging. Development efforts such as these may indeed represent the future of communicating with each other, whether for business or pleasure, by the Internet.
Beyond Laparoscopy: Robotic Surgery
It looks like they are taking Laparascopic surgery to the next level, with the DaVinci surgical robot, which was recently FDA approved according to an Associated Press article. The new era was ushered in with a
Gallbladder removal, but it is being testing for other surgeries such as bypasses and heart valve replacement. The AP articles note that this is a step beyond laparoscopy. The robots have multiple lenses and sophisticated instruments on their arms which give feedback to the surgeon and unprecedented precision. Microsurgery with these types of instruments may become commonplace.
Windows CE Makers Quietly Moving to Linux
Todd Spangler of ZDnet has written an article reporting that many large manufacturers of Windows CE based handheld computers are quietly investing resources into developing products based on Linux. He expects Linux-based handheld products to arrive within the next 12-24 months from large manufacturers such as Compaq, HP and IBM. Spangler reports that these companies are moving to Linux because of the large number of developers for open source Linux and the $10-15 charge Microsoft requires for each Windows CE unit, as well as the $20 charge Palm gets for each PalmOS(tm) license. Currently Palm Computing dominates the market for handheld computers and is widely used in medicine due to its compact size and available software. However, the article states that Windows CE devices are more suitable for running Linux.
Does Patient Confidentiality Exist?
Are we living an illusion that patient confidentiality exists? It has been said that in the good old days there were only 3 people in the doctor’s office: the patient, the doctor and the patients lawyer. Now there’s a whole crowd in the room: insurance companies, HMO’s, government and increasingly the computer. The above organizations can demand and usually receive access to patient records. In addition, as organizations offer their employees the health-plan-dujour in search of savings, the paper trail of patient charts on its voyage through multiple companies could subject it to multiple scrutinizations. Is patient confidentiality an illusion? What can be done to ensure confidentiality?
Patient confidentiality likely is and has been an illusion for quite sometime. First-hand knowledge of how hospitals work and a white coat can typically give access to just about any patients data. However, the inconvenience and cumbersome nature of medical records is a physical barrier to snooping.
The handwriting is clearly on the wall that patient records will someday all be online with the convenient access that this entails. It is a paradox that this access will be seen as less secure than the current paper chart confidentiality which consists of a locked door, when in fact electronic records can be made far more secure than a paper chart. However, as long as there are organizations with large databases of patient data and no economic reason for maintaining confidentiality, confidentiality will not exist. A current example of this is the recent problems with Toysmart as reported by CNN in which the bankrupt company is now trying to sell its customer data despite promises not to do so.
In an ideal world patients could pay cash only and avoid any third party prying. Similarly, a physician could own his own computer disconnected from all but in-house networks, running peer-reviewed open-source software that has been certified as being free of ‘backdoors’ (access points to a system that a third party has placed legally or otherwise). While utopian, such systems are quite possible and can be constructed such that only patient record numbers and need-to-know data are transmitted to third parties. Perhaps one day systems that don’t fulfill these criteria will be the exception.