Using Blender as a Personal Health Record

An open space where people who use Blender in an academic environment can share progress and the typical problems and needs of academic environments.

Moderators: jesterKing, stiv

tinytony
Posts: 3
Joined: Sat Sep 01, 2012 7:31 pm
Location: United Kingdom

Using Blender as a Personal Health Record

Postby tinytony » Sat Sep 01, 2012 8:07 pm

Hi All.

I am a new user to Blender, and the community, but not new to 3D programs having used Poser, 3D max and a few others. So excuse me fo not fully understanding Blender.

Having spent many years as a Healthcare Professional, and also working in IT design and implementation of Healthcare I have for the past 6 years working on a concept on how to show your personal health record using 3D and techniques used in 3D programs such as Blender.

I can show my record visually, my problems, my medications, diagnosis, operations, the concept can also share this information with other Healthcare professionals, such as Doctors, Dentists very easily, more importantly, it is so easy to understand any person would understand the record without having to know medical terminology. The other advantage is a doctor could share information back to me.

My problem is, I need to develop this concept with the help of a 3D developer program such as Blender.
It is not new technology that is needed, any 3D program can produce this, it would be how the information would be stored and generated and used that would need developing.

Those that I have shown this concept over the past 2 years love this, and I would love to find someone, a group, or a company that would be willing to start a new revolution in Personal Healthcare and Hospital Electronic Healthcare Records.
Tiny Tony

stiv
Posts: 3646
Joined: Tue Aug 05, 2003 7:58 am
Location: 45N 86W

Postby stiv » Sun Sep 02, 2012 1:19 am

A fascinating application. Some random thoughts:

The biggest challenge is to create a mapping from medical records terminology to specific parts or sites on a model. This might a good place for natural language processing/feature extraction techniques or maybe machine learning classification.

The 2nd big problem is labeling all the parts/sites on a human model.

There are some models of the human body floating around, some of which have been mentioned in this forum.

Seems like there would be some tele-medicine tie-ins.

tinytony
Posts: 3
Joined: Sat Sep 01, 2012 7:31 pm
Location: United Kingdom

Postby tinytony » Sun Sep 02, 2012 9:53 am

Hi Stiv.

Thanks for replying, and yes, The biggest challenge has been to create a mapping from medical records terminology to specific parts or sites on a model. it is why it has taken many year to resolve this, and where development and a slight change in the way 3D programs are used.

I have to be carefull here to protect the concept of how it all works, but secondary to this it can basically be mapped back to SnomedCT terminology which is an international coding system for clinical information.

At the moment there are approximatly 4.3 million codes in SnomedCT, a big challange is to map this, but as long as my concept and the development needed in this area is completed, then I am certain that international clinical organisation would open this with open arms.

Even in this area the cocept is not new, it is using existing technology in a way that has not been used before.

Therefore if anyone can help, or even Blender I would love to share with you in confidence the details to see if we can drive and develop this into a usable product that could potentially be used by anyone in the world.
Tiny Tony

pep
Posts: 32
Joined: Wed Dec 16, 2009 7:58 pm
Location: Costa Rica

Postby pep » Sat Sep 08, 2012 4:11 am

stiv wrote:...

The 2nd big problem is labeling all the parts/sites on a human model.

There are some models of the human body floating around, some of which have been mentioned in this forum...


Yes, we are working on the new version of Human Database from http://lifesciencedb.jp/bp3d/

The 3.0 version is a major update but it has "only" 934 parts and the hi-resolution version, imported in blender, is a 720 mb file - the lo res is a 170 mb.

Yet now all the parts have their proper name and are not truncated anymore :)

We are working at a way to have on the fly translation and are translating to spanish them. Maybe i'll have a opportunity to make a small demonstration on next blender conference but, if someone is interested to it, i can post the follow on of the story on the original thread.

tinytony
Posts: 3
Joined: Sat Sep 01, 2012 7:31 pm
Location: United Kingdom

Postby tinytony » Sat Sep 08, 2012 5:48 pm

Many thanks for replying pep.

You link is very interesting and yes the 3D models are interesting, if you could give me the link to the origional thread as I will not be at the Blender Conference this year.

I am not using reference points on the models for my concept, the XYZ coordinates would be too complex (Movement, and adjusting points would put any reference point out).

I am using the U-V mapping and the coordinates from this as reference points to certain parts of of the body and organs.

This way, even if the body changes, or points move, your reference points would not.

The colour used to place a mark on the model also used to drive the medical interaction (problem, diagnosis operation ETC) which is then used as reference to SnomedCT.

The U-V map is also used to link to any documentation received by the Doctor or Hospital.

That is the basic theory and concept, but need experts in these fields to help drive this forward.

Again thanks Pep, looking forward in seeing more of your work.
Tiny Tony

pep
Posts: 32
Joined: Wed Dec 16, 2009 7:58 pm
Location: Costa Rica

Postby pep » Sun Sep 09, 2012 4:22 am

Well, maybe i understood something less than half of what you describe :oops: (just a visual artist here)

but the link to the original thread is

http://www.blender.org/forum/viewtopic. ... 0adf465dfc

There you will find links to the 2.0 version imported to blender and the link to the original japanese database that was in the anterior post.

We have not yet uploaded the 3.0 version of the database model imported to blender. Main reason is that the import is far from complete. It has only some primitive texturization, the translation to spanish is not complete and no retopo at all.

The problem is that the Universidad de Costa Rica cut the places of the two assistant that were helping on this job, so we can work at it only in the spare time, but i think that we could motivate them, Lucia Gonzales and Luis Zarate (credit is due), to keep on if they see that someone appreciate their work.

Luis Zarate, who did the programming for the import, rename and group ordering maybe could help if you explain what is your idea of database management of the parts. And I hope Lucia keep on with the retopo job, to make the model manageable with smaller machines.


Return to “Academic & Research”

Who is online

Users browsing this forum: No registered users and 0 guests