- Its the age of Great Leap Forward..
People are happy, work goes on.. life is kind of strange with no BSOD but people are still carrying on.
One thing they are unhappy about.. They have to work. Gone or the days when their trusted OS used to giveway under then leaving them staring at a BSOD so that they could make a quick call to the IT in charge and go out for a coffee.
Excuses for work delay have come down.
So how did we get ourselves in this mess ?
We had to struggle.
What follows is a first hand account and a general guide for all Doctors who wish to implement Open Source Solutions / Software or Linux based systems at their workplace but are afraid to take the plunge.
For the technically minded Mr. Kumaran (http://kums.in) is preparing a more technically heavy note full of implementation details. Please watch that space.
– Sit down and think what we want our computers to do for us and define a budget.
– This will dictate what software & hardware you need to get that
– Make the right Hardware and Software Investments. It is very easy to
get convinced by the vendor to invest more in features you may never
use / need.
– You need to choose you FOSS projects carefully.
gives you indices that help you decide the popularity of the software.
– Look Before you Buy: Always check the Linux Compatibility of the
hardware you are going to purchase after you have installed Linux.
While many will work out of the box, some will work after tweaking,
some may not work.
– Get the right persons for the job. There will be an element of luck
here. If you are, then you will find somebody passionate abt this. Consult them.FOSS enthusiasts and ready to help.
– Do your Homework: I worked for 5 years trying out one solution after another before settling on the ones I wanted. Since I know the work flow of my hospital, I naturally know what suits it the best.
– Keep it cheap. It can be done. Cost is important, even if you are willing to pay
more. Bargain hard. Buy only what you want. Dont by what you are
– Interoperability within and without your office. This is important.
When going down the not-so-beaten-before track. Detractors are many.
Complaints will be raised. It will be irritating. Don’t expect users
to share your idealistic vision. Every complaint needs to be
listened to and most of them can be safely ignored. Some of them
will give an idea of the gap that needs to be bridged.
– After-installation-service – accessibility, response time, ?365X24X7.
This needs support. If you have an in house reponse team that knows
FOSS solutions, then thats good, otherwise you may need to buy
support. My Open Source friends tell me that Red Hat support is good
and is value for the money paid. Decide on support. Lord Murphy is
always watching. Depending on the level of in house support there may be some investment here.
– Staff training had to be spot on so that there is no excuses
to put up your hand and say `open office is too tough`. This is the
most important. Implementation will fail otherwise. Management will
be inundated by complaints of `non-working’ computers. In our
Hospital. Mr. Kumaran cleverly themed Cent-OS to look list Win7.
Most users tech-savviness ends are very `iconic`, so thats all that
What did we do ::
– Even when we had the WinSystems I stood firm and never let the management invest in MS office. It was unnecessary. Complaints were many ranging from inability to open external files to difficult UI.
Some of the complaints, especially UI related were somewhat relevent, but I decided not to pay heed. These have settled down.
– Avoiding repeated investments to tide over a initial improper investment: The Pros kept telling us to invest in Teminal Server License but we stood firm. I kept telling the management that that takes away the advantage of cost and that I could run it better without any further investment by installing Linux. There needs to be a strong FOSS enthusiast in whom the management has trust. Otherwise the Corporate win.
– Kept to a budget: When we cut off all the frills like antivirus, firewall, terminal license we cut off all the recurring expenses.
Know what you want and buy only that. They will try to sell you more; thats their job. Let them do theirs, and you do yours.
– Choose your FOSS projects you wnt to implement: Now that you have got the courage to implement, use only well maintained and mature FOSS projects. We have Cent OS , Ubuntu, Zope & Plone, mysql, postgresql, dcm4che,JBoss – all mature and well maintained. For EMR you can use Open EMR, Open MRS, GnuMed, GNU-Health. There are several DICOM viewers you can use depending on the data you want : Oviyam, Mayam, Osirix(Mac), Aeskulap, GingkoCADx, inVesalius and many more. ImageJ is useful with its Java plugins for image analysis.
-Use the LTS version of the OS if possible. Install it on the staging system first. Train all the staff, get them to use it everyday and stress the system. This will give you an idea as to what kind of server you need now and when you expand in future.
– Training. It is important to train the staff. This is paramount. Our Admin Officer (AO) who also looks after IT complaints goes around doing all the minor trouble shooting and doubts. Once the staff realise work has to be done with this, they keep quiet and try to learn. We do Plone/Zope training many times a week. The software vendor and our AO do the Hospital Software training. The basic use of a Linux thin client interface has been taught. The theming is very handy for bridging the gap.
– Firewall. Set up a linux firewall and set up basic access rights to prevent unauthorised access to internet. Setup a secure Wireless. This will help access to Intranet, PACS, EMR over mobile and touch devices and also allow remote access.
Accept the Sore Areas:
Medical Presentations: Right now its only MS office. Even Mac guys have trouble presenting many times; atleast in India. This will need users to use their personal MS office products to prepare their presentations. Sadly Libreoffice Presenter leaves much to be desired. It can do the basic .ppt export, but cannot be relied upon
Printer Support: This will need you to look before you buy. If you have expensive existing systems please evaluate. We could not use two network printers (Sharp)after switching over. They are both aged, so replacement is an alternative.
External DICOM applications: Many patients come with CDs with bundled DICOM viewer software. This may not work in Linux ; even under wine. Depending on the type of practice this may be a problem. In our current practice most patients come with printed films so it is ok. Some of the DICOM viewers open / the CDs can be opened with Aeskulap, but GE PACS cd usually dont.
.docx files: People send you MS Office files of all kinds. You cannot expect them to save it in a legacy format. Ubiquity of MS Office and Win make them expect a reply when the files are received. MS office 2007 – 2010 files are an issue. Simple ones open. Extensive use of animation, graphics causes improper rendering. Usually Office communications are bland letters which open without an issue. For academic presentation use, we suggest you rely on a personal copy of MS Office in you laptop.
Hopefully these basic guidelines will help people to switch. This is first hand account. That should be encouragement enough.
We have made it happen at our Hospital. It is comfortable. Transition has been very smooth with most users not even knowing the OS has changed. They only feel the computer is faster.
Please feel free to leave comments in case of any implementation queries.
As I said before watch http://kums.in for technical updates.
This will be integrated into the EMR.
As of now this is Standalone but this will be integrated into the EMR.
This is still very-much-beta. Design & Functionality will change considerably before I go final.
Am having a repository developing Bluebream (Zope 3) simple contact management application at BitBucket.
Its here: https://bitbucket.org/dreaswar/bbcontact/
Its just starting. I suppose pluggable modules can be created to make it like an ERP
Care to join ?
The Great Dream …
This post is my dream .. or has been for about 4 years.
A day that my hospital runs on full Open Source Software.
First a little about me. I am a practicing Paediatric Orthopaedic and Spine Surgeon. I am a Open Source enthusiast. I started using Linux 6 years ago and for past 4 year I am using it almost exclusively at work and home. The only time I use Windows is for the odd gaming. I do Python programming – web & desktop.
When we moved into our new hospital premises, set up 2 years ago, to start on a good note and to save start up cost I set up an Open Source Intranet (Plone) and Open Source PACS for my use at the Hospital. I also started developing my own EMR project that I speak abt in this blog.
I dreamed that the hospital will implement an all Open Source Solution. I advised them likewise.
They seemed to listen. Then FUD (Fear -Uncertainty-Doubt) took over them:
How can we follow the advise of this Non-Professional ?
What happens if we have a trouble in future and then he cant help us ?
Where do you find the Linux certified guys to help you ?
What happens if Linux is sold off to come company and it becomes a paid?
(Yes, they did actually say that ).
So they sell themselves to the “Professionals” . Our Management was no different.
It was advised that our Hospital will need the best server, a professional firewall, latest antivirus, and all windows machines. Individual desktops were advised, even for Reception area ! . This along with the latest MS Office and all the great accompaniments. … It was not cheap – It was not meant to be. The offer looked really good. The guileless management was tempted to say ‘yes’.
Then they thought they’d double check with me, just in case.
The Great Deliberation followed.
One looked at the order, I chopped off half of it. Why do you need a comp with Core2Duo, 320 GB drive, 4GB RAM and Win7Prof for Reception, Cash ?. All that they ever use to is to login to our Hospital Software.
I suggested all Thin Clients, Open Office, Linux on Server. Firewall with Linux. No antivirus software. Intranet with Open Source and Open Source PACS system. Desktops only for Doctor chambers and Media editing.
After mustering courage and ample dosage of FUD from the Professionals, Management decided on Windows Server & Win Thin Clients. They were whining all the while saying that my idea is ill advised. I suppose he would considering that I trimmed the budget at least 5 -6 lakhs of Rupees.
So it was going to be a huge server we may never use with features and specs that many small IT companies may not have or want or need or use : Sonic wall firewall, MacAfee antivirus, MS Office on all Desktops the list went on and on… There were ThinClients at all stations and Desktop at Doctors rooms and other important admin areas only. IT was Windows everywhere.
The Professionals offered to set up the Server with domains all the security stuff. It was bundled with the purchase.
Things had barely gotten off the ground after a year of struggling to set it up. Then a 2 year jinx started.
The Great Depression followed.
The Thin Clients which needed to work with only our Hospital Software (written in Java) needs Firefox. Most stations needed this software for the daily work. Even though we had purchased the ThinClients after testing with Firefox and our Software, after implementation it was horridly slow. CPU was clocking 100 % the moment we use the software on Firefox.
The ‘Pros’ blamed it on Thin Clients. They told Mangement we told you so. You needed Desktops. Buy it and it will run things smoothly. Blame game between the ‘Pros’ and the software vendor started.
They could not sort it out for over two years. Work at the hospital suffered. We needed to replace the Thin Clients at the high workflow areas with old desktops to that we could serve.
All the while I kept telling them you move to Linux it will be all right, but they needed a ‘Pro’ to tell them that. Of course, they would not. They tried to sell us more. After `considerable study` at their HQ and evaluation of the software, they said it will be all right if the server did all the processing and asked us to shell out more for the Terminal License for Windows.
I put my foot down and said no. I said I could make it run smoothly under Linux if they wanted, so they had better come up with a better option.
Things dragged on with no news from them even after 1 year. The Management was frustrated. Then Windows cracked and light came through. The Management decided to Open the Doors.
The Great Revival followed.
Enter Mr. Kumaran (http://www.kums.in)
Cent-OS it was then for the main server with customised `really thin` clients. One week of testing with VirtualBox and couple of flashed thin clients and the Managements saw the light of what I was saying all the while. All processing at the server. Hospital Software is fast. Staff are happy. Work gets done. Management is happy. Work get done with no money spent.
Now we have fully converted. Printing is an issue because of the WinPrinters we had purchased. Luckily most of the are aging and needs to be replaced.
The Great Leap Forward …..
Open Source Intranet Platform
- Intranet has been customised and installed.
- Used everyday at our Hospital now.
- It runs Plone with add-ons and custom scripts.
- Currently we are `fitting the gaps`
- Scheduled implementation is due next month.
- Staff seem to like the concept and are getting the hang of it.
Open Source PACS
- Set up and running at full tilt
- Currently in testing.
- Scheduled implementation in a months time.
Our setup now contains
- A main Cent-OS Server serving the Thin Clients
- A Desktop with Ubuntu 12.04 PACS Server (Staging)
- A Desktop Ubuntu 12.04 Intranet Server (Staging)
- A Desktop Ubuntu 9.04 Server for the Hospital Management Software Maintained by Software Vendor. We need to migrate this as it is aging.
So, what are we leaping to ?
- Open Source EMR
- Open Source ERP
- and more…..
The entire application interface is basically a tabbed top and bottom container for Patient management.
Patient list grid is basically a list with Datagrid with a jsonRest store. The Django view will return the json. The row click will fill the bottom tabbed panel with appropriate contact, phone numbers, email, guardian, admission and visit info for that patient.The double click will allow editions / deletion of the patient if the user has the permissions.
Application currently uses Django 1.4, PyYaml, PIL, ReportLab, PISA(html5lib) and Dojo 1.7.2
The icons currently being used are from KDE but that may change.
The second sreenshot is the comparative interface in jQuery. There is no bottom pane here. Left sidebar shows the context info on patient selection from the list. The Right sidebar is hidden if the screen resolution is below 1024 and shows itself on zooming out / clicking the icon at the top left.
This interface uses jQuery, jQuery UI and plugins for the table with heavy CSS and jQuery customisation. The icons used are mostly from the tango project with a few from the silk collection for web. I have made some of my own. Some icons are just place holders and i need to work on them to make my own ones.
As I noted few days before, I have split the project and both are maintained.
The Dojo front end is called OrthoDocx and jQuery front end is called OrthoEMR.
It is Orthopaedic specific as of now.
Will upload comparative screen shots soon.
Hi, As usual updates are infrequent.I have moved from jQuery to Dojo as the front end development in jQuery was not satisfying for me. Dojo is more polished. I like the Dojo approach of bundling all the plugins and integrating it tightly rather than the add as you like approach of jQuery. Dijit is superb.
Of course Dojo documentation can do better. However, it has become so much better than a few releases ago.I have found that if you persist with it after some initial frustration, it is rather nice once the development picks up speed.
I chose to do the forms with minimal modification of Django code. I did not resort to Dojango which i feel binds the Django code too much to Dojo. What i have done is to set up series of JSON views which tries to remain as front end agnostic as possible.
I didnt have the heart to do way witht the jQuery project. So i decided to maintain both. Currently the focus is on Dojo. I am using both at my clinic and it seems to work fine. Development is progressing at a good pace.
Expect some screenshots soon …
Hello wandering web surfers,
It has been a long time.. the spider was asleep.Just to announce that long time go when this blog started i had mentioned about designing my own EMR.That is taking shape. I have been quietly at work for about past year and 3 months at it.
Choice of Desktop v/s Web:
Well i went for the web..
Python was the language i know .. at least a little bit.
A little bit of googling suggested Django was the most popular, so i jumped straight in. Initially i did think of trying out wxPython as the base like Gnumed but .. for no specific reason decided on Django. There was however some inspiration as i found at least one project which has used it..
There were times when i didn’t have time to teach myself all these and thought of switching back to wxPython, but i stuck on and i am now reasonably satisfied.
I had made a Google code project page .. DjangoEMR. This could not be filled up as i had not anticipated the kind of delay.
I thought, rather stupidly that Agile Web development will be faster and i will be up and running in no time. I must say these 20 min blog tutorials are so misleading !
What it is::
Django EMR is a Orthopaedic Small Clinic / Physician EMR.
This can be extended to a multi-doctor practice, though i have not put in any effort on that as the project stands now.
The aim has been to optimize it for personal record keeping.
There also has been no specific attempt to incorporate things like HIPAA / HL-7 standards.. simply because i don’t have time to do that stuff. The aim has always been to create a personal record keeping.
One thing, this is designed as a Orthopaedic software, but with little effort this can be easily extended to other Specialities.
Features:: (most of it done, some of them almost)
- User Registration ( basically built into Django)
- Patient Contact Database
- Admissions ( No Bed management / Medical Orders ).. remember this is not a Hospital Information System
- Physical Examination and Daily Progress notes
- Investigations and Labs result entry
- Discharge Summary Generator
- Detailed Diagnosis and Classification of each patient complaint
- OPD Visit Management ( No Scheduling / Appointments)
Other planned features::
- Physician Dashboard ( No of patients, admissions, events, reminders etc..)
- OPD visits scheduling and Appointments
- OT Booking and Planner
Well, there are some more things i want to finish before i populate my project in Google code.
Stuck with the kind of busy days as i am nowadays really cannot give a time line.
watch this space.
Will i need Help ? ::
Most certainly i would.
Please do let me know here.
As usual.. i was doing my daily Python exercise..
Since i am a noob,,, i could not solve this simple Python problems of finding prime numbers !!
i came across the wonderful blog that helped me..
have a look.
In the beginning, I did not exist….
there was this great void on the web….
… then on one lazy Friday after noon.. after an enriching cup of hot Milo, the Spider decided enough was enough and in a great maniacal flurry of creative hurry, decided to leave marks of its existence on the web of life.
This Blog is a product of that afternoon. Now i exist on Orkut, Facebook, LinkedIn, WordPress and Twitter..
Enough about Me…. Now on to something completely different………….
Me…( Ya.. i like Monty Python)
I am an Orthopaedic Surgeon,, that is to say i try to mend Broken bones.. and fix worn out Joints for a living.
I grew up in a small town called Palakkad; a dry and windy town of Kerala State in South of India.
After training and doing Fellowships in various subspecialities that has now come up to split up the once broad speciality of Orthopaedics.. i have come to subspecialize in Paediatric Orthopaedics, Deformity Correction and Spine Surgery.
When i get bored, i tweak my Linux and i code my Python, PHP.. I have a great project of building my own EMR.. ( that is an electronic medial record)
Jokes aside fellows,
Great to be here.. Just feeling my way around here..
Hope to fill the blog in coming days with Orthopaedic notes, Patient education write ups. Also plan to start a Python code snippet collection.. someday..
Thanks for peeping in..
Comeback sometime . Hope to have the contents ready.