Hi Mario :-) Apology and some questions--Ira Laefsky
Ira Laefsky
Hi Mario:
I hope that you are not offended or put off that we in physiosense are looking at applications that affect health fitness and the study of physiology, but are not strictly speaking for
Certified Medical applications. I firmly believe that we have a lot to learn from each other even if we are involved with slightly different applications of very similar technologies.
In particular I/we are currently looking at the Cooking Hacks Arduino/Raspberry Pi E-Health I/O Board that you recommended to our attention as one was of obtaining analog health
Signals such as GSR & EKG, in addition to obtaining digital signals from such “appliances” as Zeo, Fitbit, Neurosky, MyBasis, etc. Since we both are considering development around
a Linux Distribution and we wish to obtain analog data from the Cooking Hacks E-Health Shield, I/we are considering either transferring this analog data (along with eventually digital signals obtained from the sort of digital health appliances I mentioned to a Linux system (either the Raspberry Pi or initially on a Linux PC) --the PC allowing easier I/O for Wifi, Bluetooth Zigbee and USB devices. Since we both want to enable both local data logging, on-screen menu selection, and communication with an Internet of Things Site we are investigating a
software systems architecture under Linux that will permit simultaneous development of different individual programs (possibly in C and Python) by different Open Source contributors.
If you are able and willing to share it what is the general systems architecture of your “totem system”? That is what are the principal software modules that you are developing and
in what languages and how do they fit together?
Thank you very much for your previous information and I hope that we can learn and share much with each other even if some of our intermediate goals (as far as distribution) and
certification may differ.
Much Thanks and I pray no offense
--Ira
From: Ira Laefsky [mailto:lae...@comcast.net]
Sent: Saturday, January 05, 2013 11:17 PM
To: 'Mario Baldini'; 'Charalampos Doukas'; physi...@googlegroups.com; 'Emery Premeaux'; 'Mike Dodaro'; 'Daniel Neis Araujo'; lae...@comcast.net
Subject: MARIO & ALL Some Questions Some Answers Let's Keep The Discusssion Going
Mario & Other Participants:
Your work sounds quite impressive!
You are quite correct in pointing out the rigorous steps required for certification of a device that is going to be used for patient care in the medical community. In the US
The FDA (Food and Drug Administration) a U.S. Government Agency http://www.fda.gov/medicaldevices/productsandmedicalprocedures/deviceapprovalsandclearances/510kclearances/default.htm
requires a 510(k) device certification involving many months of testing and agency approvals for any device that is being used for a “medical purpose” in patient care.
According to a friend of mine who manufacturers EEG Equipment this process can take up to two years and a total of over $1,000,000 (US) for which he had to obtain
outside venture capital in spite of already having a successful going concern.
We as hobbyists and professionals on the Open Health/Physiology Aggregator project are seeking to fill a real need that we and 10,000’s of other experience which does not
require this investment (unless we later decide to license its use in patient care). There are in the US and around the world a community of fitness enthusiasts, lifeloggers
(who collect data on their own life and health concerns, and a very large community of Quantified Self http://quantifiedself.com/ followers who track their own health and fitness
Data and purchase many millions of dollars worth of fitness and health monitoring equipment such as Fitbit, Zeo, Neurosky, Emotiv, Polar, Basis, etc. (all commercial successful for profit companies). This non-medical equipment not sold as treating any medical condition, and which must be safe in general and if a commercial entity is marketing it for profit must avoid product liability which has a much lower standard and does not require any particular certification although some devices have UL or EU certification. There are several existing open source projects which are serving this market such as
Open EEG http://openeeg.sourceforge.net/doc/ and Open Health Tools http://www.openhealthtools.org/ .
The particular need we in physiosense/Open Health Physiology Aggregator seek to address is that of individuals (usually healthy) who have either purchased commercial digital health and fitness equipment such as Fitbit, Zeo, Neurosky, Emotiv, Polar, Basis, etc., and or have built their own mostly analog devices such as pulse, or SP04 and who wish to track their data in a
Database or Web-based environment that they themselves control. Some members of our group such as myself and Charalampos also wish to conduct academic publishable research using these devices aggregated to the Internet of Things for Data Analysis. But I should emphasize that Self Trackers, Fitness Enthusiasts and Quantified Self enthusiasts are a huge market
in which much commercial profit can be made even with Open Source Systems (hardware and software). And this is a need that does not require making claims or obtaining registration as
medical equipment.
Now a few questions on your project and how we can work together. I am curious as to what sorts of Analog Processing with ADC’s (analog to digital converters) and operational amplifiers for signal conditioning you are using to complement the digital smarts of the Beaglebone? As I believe you pointed out earlier, Libelium a Spanish company is selling to the hobbyist market a E-Health Sensor Platform for the Arduino and Raspberry Pi that provides an interface to many of these Health Sensors
http://www.cooking-hacks.com/index.php/documentation/tutorials/ehealth-biometric-sensor-platform-arduino-raspberry-pi-medical?utm_source=banner_ehealth&utm_medium=banner and I have been in communication with their management. What do you hope to do with Sensors and Analog Electronics to
collect the data and will this electronics be a commercial product before being Open Sourced? As much as you care to share on your project what are the principal components of your
software architecture on the Beaglebone? What are the main modules of your software and what do they accomplish? To what extent do you share our goals of collecting data from
commercial digital and hobbyist health and fitness monitoring aids and sending that to a database and or the Web of Things for charting and further analysis?
I look forward to the many ways in which our efforts intersect and through which we could collaborate. How would you see us working together?
Many Thanks
--Ira Laefsky and physiosense
From: mario....@gmail.com [mailto:mario.o...@gmail.com] On Behalf Of Mario Baldini
Sent: Friday, January 04, 2013 12:37 PM
To: Ira Laefsky; Charalampos Doukas; physi...@googlegroups.com; Emery Premeaux; Mike Dodaro; ira.laefsky; Daniel Neis Araujo
Subject: Re: Info On Your Linux Health Projects/Collaboration
Dear Ira and fellows,
Happy New Year! Just now with the holiday season I was able to proper respond some pending e-mails in my inbox.
We have been using Beaglebone as a hardware tool to ease development of a platform around the TI's AM335x. This SoC was chosen because its performance capabilities (LCD control, high level OS -- android), while keeping a relative low cost when compared to other embedded computers.
In the future we plan to build a custom hardware from scratch, using this SoC.
The short time goal is to develop a "totem" style product, for homecare use. We are currently integrating an OEM module to acquire Blood Pressure. We also have partially developed a module for electrocardiogram acquisition.
This personal hub would then transmit the data to a central (using GSM network). Some data would be directly acquired by the totem, or also by external modules (like wristwatches, etc). For other portable modules we are still evaluating some technologies, like AVR vs ARM based microcontrollers, or 2.4Ghz vs 433Mhz band.
Since we are a startup (with no big capital), we are developing this as a commercial product for a specific customer, but the ownership is still ours. As we gain financial stability, we intend to make this opensource and give the proper support.
In our view, just throwing the source code and schematics isn't the right way to do opensource. Specially in the healthcare field, we need to take special care of it, in a way that the final product really reaches the patients, avoiding that the project doesn't simply implodes.
Regarding the goals of Physiosense, we basically share all of them. It would be great to contribute with anything possible.
One big issue that we didn't quite figured out how to solve, when it comes to making it opensource, is how to build something that is really usable in the real world, and not just for academic / educational purposes.
There are many tools (ie. arduino based) for health data acquisition. These tools are *great* for learning purposes, or niche scientific use. But mostly aren't legally approved for deploys with real patients.
I personally don't think that the technical aspects are the main problem. I see you investing a great effort in the specification of inputs, sensors, etc for the platform.
At least in the Brazilian healthcare market, we learned that most of the work is non technical. It's difficult to see a project like this to succeed and gain large scale without a proper market strategy and financial roadmap.
By "how to put in in the market" I don't mean commercial / how to sell it. But how to legally use in the real world patients.
I constantly see medical products that are technically simple, but costs *a lot*, simple because there are few vendors and most solutions are imported.
What we are trying at Atto is to create products ( and opensource them) to be effectively used by patients, in the real world treatment, since most of the available solutions are proprietary and overpriced.
What is the roadmap of the project, regarding the certification process ( and financial support of it ) ?
Forgive me if I misunderstood its goals, and to certificate it isn't one of them. But in my view, if we want that this platform become widely used, by real patients, the product must be properly certificated.
Certification process implies (at least in Brazil) that someone becomes directly responsible for it (liability). Who would fulfil this role in the project? Would a company/foundation be in charge of it?
Also, is there any intersection point with the Continua Health Alliance ( http://www.continuaalliance.org/index.html ) ?
I'm not trying to just point out problems. I am sincerely exposing the main issues that we have facing, as a way to discuss it further.
Thank you,
Best regards,
Mario Baldini
2012/12/17 Ira Laefsky <lae...@comcast.net>
Hi Mario:
I am eager to learn more about your Linux projects involving collecting data from Health Sensors, and to learn more about how we might work together for the benefit of Health Self-Trackers (e.g. the “Quantified Self” community) and Open Source data aggregation. Pls. let me know some more about what you have done in the past and your thoughts on our documentation.
--Ira Laefsky
SKYPE: ira.m.laefsky
Gooogle Groups: Physiosense
From: mario....@gmail.com [mailto:mario....@gmail.com] On Behalf Of Mario Baldini
Sent: Tuesday, December 11, 2012 8:28 PM
To: Ira Laefsky
Subject: Contact
Dear Ira,
Sorry I couldn't reply you sooner. I just came back from a business trip and had many issues to resolve.
Please don't see this as a lack of interest. I am very interested in contributing to your proposal.
Best regards,
Mario Baldini
--
Atto Engenharia de Sistemas
Caixa Postal 10170
Florianópolis - SC
CEP: 88062-020
Brasil
--
Atto Engenharia de Sistemas
Caixa Postal 10170
Florianópolis - SC
CEP: 88062-970
Brasil