Ruby on Rails for Healthcare systems

Being interested in applying "Ruby on Rails" to the Healthcare sector (IT Healthcare), what do you recommend me to do as a next step after learning "Ruby on Rails"? In other words, what PATHWAY do you recommend someone to approach in applying Ruby on Rails to healthcare especially the CAD (Computer-Aided Diagnosis) systems?

Thanks a lot.

A smart person told me years ago, "You can't automate something that doesn't already work on paper." By paper, he meant "in real life". So I would say that all the Rails in the world won't help you unless you know what your application is supposed to do, and how it already works in the real world.

Walter

Walter Davis wrote in post #972903:

Being interested in applying "Ruby on Rails" to the Healthcare sector (IT Healthcare), what do you recommend me to do as a next step after learning "Ruby on Rails"? In other words, what PATHWAY do you recommend someone to approach in applying Ruby on Rails to healthcare especially the CAD (Computer-Aided Diagnosis) systems?

A smart person told me years ago, "You can't automate something that doesn't already work on paper." By paper, he meant "in real life". So I would say that all the Rails in the world won't help you unless you know what your application is supposed to do, and how it already works in the real world.

Walter

Thanks @Walter for your reply.

The application will most likely be a system that helps in diagnosing "Lung cancer", by for example comparing a scan with other scans in the database that look similar, based on which we can aid the physicican (Radiologist)in giving his diagnosis.

But, I really asked my question generally because I'm asking about the "Healthcare IT" in general since I think what I'm thinking about is part of this field. So, before being specific, I just wanted to know what is recommended as a pathway especially for Ruby on Rails developers, and that is why I chose to ask such question in a Ruby on Rails related forum.

I'm interested in the opinion of the Ruby on Rails developers in approaching such field.

Thanks.

Hello, I like your question, because we was working on several tutorials and books.

I can accept this advice about automating something. But it has nothing about study path, adn the question was about study path, isn't it?

I think Rails is for web applications. First of all you have to decide if your application will contact to the net? It seems not this is the main functionality of your application, and noone will check his cancer on Facebook. It means you probably don't need a webapp. Maybe later you need one which can communicate with your software.

I would ask it in Ruby forum and mailing list as Rails developer, and also go to a meetup on your local area, I'm sure someone will give you advice where to start. Best regards, Zoltan

I've built two intranet applications for a medical device manufacturer so far, more to come. I think that depending on the application, you can get a lot of mileage out of a web app. For one thing, the cost of supporting the desktop drops by a huge factor. For another, rolling out updates is completely trivial, with no need to traipse around to each desktop (or location, in a more widely-distributed organization) to make sure that everybody has the same version. I'm not saying that every problem is a nail, but you can get a long way through a lot of different types of problems without crossing the boundary into native application-land.

Walter

So @walter, what do you recommend to approach healthcare through RoR?

Healthcare is much too large a target for me to say "do this" or "read that". I worked for several years at Medical Broadcasting Corporation, now Digitas Health, and worked on a lot of different kinds of healthcare + Web projects (this was long before Rails came along).

* Some of these were basic static sites to promote a drug or to post the warnings somewhere so they didn't have to waste so much of the commercial time telling you it would grow hair on your palms or whatever.

* Some were quite ambitious intranet/extranet collaboration tools for the marketing team.

* Some were information-gathering systems for clinical trials.

* Some were very early social media attempts -- way too far ahead of the curve for anyone to try them.

* One was (I kid you not) a Web application to help people choose gifts for friends or relatives who were in the hospital or recovering from a serious illness. All sorts of magical fuzzy logic in there -- don't buy stuffed animals for a person with allergies -- that basically recreated common sense for the senseless. Over a million and a half dollars and it was pulled in less than a year.

So there's a LOT of different things under the "healthcare" tent that you could pick to work on.

What made it easy at MBC was the sales team -- all hand-picked from the cream of Big Pharma marketing -- who really knew the ins and outs of how a pharma company worked on the inside, and could think up clever ways for us to wiggle in there and build something cool for a profit.

If you can't replicate that part of the equation, I doubt that Rails or any other framework would help you much. But if you have landed a project to work on, Rails will help you build it much more quickly than traditional techniques.

Please do post some more focused questions and I (and others) will no doubt be happy to help you.

Walter

Thanks @walter for your throrough responce.

To be more specific where I would like to use RoR, I would like to use it initially in:

- Reading a DICOM image of a patient that is more likely suffering from lung cancer for example. - Compare the read DICOM image with other DICOM images in the databaase. - Returning the most similar case to the read image.

This in its case will help the radiologist diagnose the patient on lung cancer.

So, as you can see, I'm interested in applying RoR in CAD (Computer-Aided Diagnosis), and think that some image processing would be needed here.

For this, how can I approach it through RoR?

Being interested in applying “Ruby on Rails” to the Healthcare sector

(IT Healthcare), what do you recommend me to do as a next step after

learning “Ruby on Rails”? In other words, what PATHWAY do you recommend

someone to approach in applying Ruby on Rails to healthcare especially

the CAD (Computer-Aided Diagnosis) systems?

At Baobab Health Trust (Malawi), we are using Ruby on Rails to develop health systems. Our systems are web based (obviously), touchscreen based. They are more of electronic data capturing systems. Personally, I have been involved in the development of the following:

Thanks so much @Edmond for your very valuable reply. It is really VERY helpful.

@SW Engineer:

I just noted that my “bitly” links are not working fine. Here are the hard links:

Regards,

@SW Engineer,

I just noted that my bitly links (in my previous post) are not working because they miss underscores. I hope these one will work fine for you:

* Diabetes Mellitus and Hypertension Treatment(DMHT) System ([http://bit.ly/baobab_dmht](http://bit.ly/baobab_dmht))

* Antenatal, Maternity & Underfive System ([http://bit.ly/baobab_maternity](http://bit.ly/baobab_maternity))

* Inpatient Electronic Data System([http://bit.ly/baobab_spine_v_1_2](http://bit.ly/baobab_spine_v_1_2))

* Outpatient Electronic Data System ([http://bit.ly/baobab_outpatient](http://bit.ly/baobab_outpatient))

* Baobab's ART (BART) System ([http://bit.ly/bart_2](http://bit.ly/bart_2))

Regards,

Thanks so much @Edmond, really appreciate it.

What do you think of them is near to what I'm thinking about, that is a CAD system that diagnoses lung cancer through making comparison of DICOM images?

@Edmond. Yes, the bit.ly links work fine now. Appreciate it.

I'm ready to share more with you especially, on implementing expert CAD

systems.

Appreciate that @Edmond So, what is the pathway that you recommend me to follow do in order to reach the goal of for example: "Building a CAD system that helps in the diagnosis of lung cancer".

The first clear step is "Learning RoR".

Now, after that, what do you recommend me to do?

Can you give me some steps to walk along with?

Thanks.

I'm ready to share more with you especially, on implementing expert CAD

systems.

Appreciate that @Edmond So, what is the pathway that you recommend me to follow do in order to reach the goal of for example: "Building a CAD system that helps in the diagnosis of lung cancer".

The first clear step is "Learning RoR".

I think you are going about this in the wrong way. You stated in an earlier post that you wanted to build:

A CAD system that diagnoses lung cancer through making comparison of DICOM images.

I suspect the most difficult part of the job is that of analysing the images and determining the diagnosis, or do you already have an algorithm for this? If not then the first thing is to determine the algorithm for analysing the image and then determining the best language for performing the analysis. You may find that it is incredibly processor intensive, in which case an interpreted language like Ruby may not be appropriate. Perhaps you will need something like a C++ back end with an ROR UI layer on top.

Colin

Thanks @Colin.

Since there is a theoretical parts that can be taken care of (i.e; algorithms), I just want to be rest assured that I can implement that using RoR. So, the issue is, can I do such CAD stuff on RoR?

Thanks @Colin.

Please don't top post, it makes it difficult to follow the thread. Insert your comments into the previous message at the appropriate points. Thanks

Since there is a theoretical parts that can be taken care of (i.e; algorithms), I just want to be rest assured that I can implement that using RoR. So, the issue is, can I do such CAD stuff on RoR?

It is impossible to know whether it can be done in RoR unless the algorithms are defined. As I said, if the algorithms require huge amounts of processing power than Ruby may not be a suitable language.

As to whether RoR is suitable for the UI, that depends on the user requirements specification. If a web style interface is appropriate then the answer is probably yes.

You must work out what the software is supposed to do in more detail than you have given so far *before* deciding on the technology to use to implement it, not the other way around.

Colin

You must work out what the software is supposed to do in more detail than you have given so far *before* deciding on the technology to use to implement it, not the other way around.

Colin

I got what you mean. So, I think if not sure if it is appropraite, one can give it a try. Right?

You cannot give anything a try until you have sorted out some details of how to do the image analysis and how you want the UI to behave.

Colin