Questions
0
Searching answer
Philips Intellivue MP70 Connection
Hi guys,
Sorry to be such a pain, but we are working on several devices in parallel and are getting stuck on all of them right now. We've been trying to connect to a Philips MP70 for several weeks now but to no avail.
The problem is that for some reason openICE doesn't connect and keeps sending association request messages, so we keep receiving what I believe is an association response. As far as I understand, there are multiple revisions of the communication protocol but I'm not quite sure how to find out which version our monitor uses. I was looking through the old topics and saw a picture posted by Jeff where the monitor had a total of 4 ethernet ports, two output and two input. Ours only has two ports - one input and one output (of course, there is also the LAN port, which we don't want to use).
Do you guys know what the problem is or at least where to start looking for it?
Thanks and sorry again!
Rado
P.S. Just so you don't think I'm only trouble, let me say that we are making good progress with the Drager Apollo and are getting ready to start collecting data from it soon, so thanks for that! :)
Sorry to be such a pain, but we are working on several devices in parallel and are getting stuck on all of them right now. We've been trying to connect to a Philips MP70 for several weeks now but to no avail.
The problem is that for some reason openICE doesn't connect and keeps sending association request messages, so we keep receiving what I believe is an association response. As far as I understand, there are multiple revisions of the communication protocol but I'm not quite sure how to find out which version our monitor uses. I was looking through the old topics and saw a picture posted by Jeff where the monitor had a total of 4 ethernet ports, two output and two input. Ours only has two ports - one input and one output (of course, there is also the LAN port, which we don't want to use).
Do you guys know what the problem is or at least where to start looking for it?
Thanks and sorry again!
Rado
P.S. Just so you don't think I'm only trouble, let me say that we are making good progress with the Drager Apollo and are getting ready to start collecting data from it soon, so thanks for that! :)
Customer support service by UserEcho
No worries, we like getting questions and feedback!
It's probably just a type, but you mentioned four ethernet ports, two in and two out. I just wanted to confirm that you meant serial ports. If you've been trying to use them as ethernet, well, there's your problem! ;)
thanks,
-Dave
I was using the term "ethernet" loosely. I meant that it has RJ45 ports, which are used as serial, as far as I understood. And we are getting structured messages from them, which openICE is able to parse. But we are stuck at the association stage.
BTW, our philips monitor only has 2 ports. The one that Jeff showed a picture of in an older topic has four, so I was just wondering if this is an indication of a different protocol or something like that.
Thanks,
Rado
Just checking in to see how things are going with your project. Where you able to connect the MP70 via serial connection successfully? If not, please write back and include your MP70 software version. This may help clue us into the problem.
Thanks,
Jeff
Thanks for checking in. We haven't had a chance to play with the MP70 recently but will go soon. We were waiting for you guys to first fix the other bug with Ubuntu not begin able to work offline since that project is higher priority. Do you know how that's going?
Thanks,
Rado
So I think there are compounded issues here. First, Ubuntu is supported and OpenICE will work. HOWEVER (big however) you are absolutely right, OpenICE will not work without and "up" network interface. Our current middleware substrate is a system called Data-Distribution Service (DDS). We implemented the vendor RTI's community edition of DDS. This middleware, RTI DDS Community Edition does not start without the host system having an "up" network interface. This is a known issue that we have been forced to work around. I would suggest re-running your testing with any network connection open.
I know the proof is in the pudding so here's a screenshot of my Ubuntu VM running OpenICE v0.6.2:
When I turn the VM network adapter off, my terminal says the following when attempting to start the application:
Let me know if this works for you,
Jeff
So the workaround is to simulate a network interface? How can I do this on a regular Ubuntu machine (i.e., not a VM)? Our goal is to run OpenICE on a beagle board, which will definitely not have internet access.
Thanks,
Rado
Is the Beaglebone that will be connected to the MP70 and running the OpenICE device-adapter software going to be connected to anything else? For example: another computer/Beaglebone running an app or the OpenICE Supervisor. Any network connection, including a local network (192.168.x.x), is sufficient for DDS to start.
Again, sorry, this is a known issue that we don't have a solution for. I'll look into it again and will write back if we find a fix.
Jeff
Our current plan is to just tape the beagle to a Puritan Bennett (and possibly an MP70, though it seems like that will take some more time) and let it collect data for a few weeks, while clinicians move it around and hook it up to different patients. So, I don't think we can expect nurses to also connect the beagle to a laptop.
Is there any way to disable the DDS code? We won't be needing it for this project.
Thanks,
Rado
Well... no, not at this point. We have never really built anything for the use case of integrating medical devices with OpenICE without the integration to any other system or device.
I will certainly make note of this for a potential future feature. Not being able to start OpenICE without a network connection a known issue. The problem with your scenario right now is that technically you are running two applications that are communicating over a network. 1 is PB840 device-adapter. 2 is OpenICE Supervisor data recorder app. In this case, the communication network you want to use is via the loopback adapter - which DDS doesn't currently accept as a network interface. Since this is a known issue with a use case behind it now, there might be something we can do but at this time it isn't possible.
Thanks,
Jeff
Don't know if this makes any difference but we are not using the recorder app that you guys provide (we're not quite sure what's the functionality of it). We're actually using our own ad-hoc data collector -- since the PB840 messages are human (and machine) readable, we're just storing the messages themselves. So essentially, we're just using the (modified) PB840 device-adapter.
So, if nothing else works, we may try to strip down OpenICE to just the PB code and only run that on the beagle.
But yes, from our perspective it would be great if you guys have an option to only provide device adapters. I guess this already works on Windows/Mac but unfortunately we can't figure out how to do it on a small board without using Ubuntu.
Thanks,
Rado
To get back to the MP70 discussion (we got a little sidetracked in this topic), we just went to have another go at the MP70 monitors at HUP (Hospital of UPenn). Turns out they are revision K (you guys seem to have J if I'm not mistaken). In any case, OpenICE gets stuck at the association step - we are getting some message from the monitor but I can't figure out what it is (I think I stored it but who knows - I just have byte array). Also, don't know if this is important or no, but since I updated to the latest version of OpenICE, the code seems to be doing something slightly different, though we are still unable to communicate.
Do you guys know what are the differences between rev. J and K? I'd imagine they're not so big if the two are just one generation apart but maybe I'm wrong.
Thanks,
Rado
Hi Alistair,
Did you have any luck with the new DB9 cable?
Thanks,
Rado
Not with my first try - I made the connections like I thought - based on the diagram- but still got no data. When I tried to pull out the wires to try a different configuration I broke one of them. Rather than order a new modular adaptor, I just ordered a $30 cable from medicollector.com so that I can have something that should work and that I can look at to make my own in the future. I'll let you know if the new cable works.
I saw that picture but like you, I assumed that it is the standard pin-out of the RJ45 to DB9 cable, so we got a few of those. Do let me know if the new cable does the trick!
Thanks,
Rado
Well, I got the custom RJ45-DB9 cable from MediCollector and repeated the whole adapter setup process. I can see the times are synchronized. OpenICE recognizes the adaptor and it shows up as an image, but I am still stuck at the Requesting Association, like before. I had to change the terminal command to address ttyO2 because that is the serial port on the BBB Revision C card that I have. I don't know if this is causing an issue or not. Frustrating. I've been trying to get this to work for months.
Alistair
Hi Alistair,
I just wanted to let you know we have made some progress with the rev. K devices. We did build the cable described in the programmer's guide (and it is not at all the same as Cisqo cable, as you pointed out).
At the Hospital of the University of Pennsylvania they have two kinds of MP70 monitors -- the first kind has 4 RJ45 serial ports (not counting the LAN port) and the second kind has only two such ports. I believe the latter is the same monitor as the one they have at the MD PnP lab though theirs is running rev. J (I paid them a visit a couple of weeks ago).
After lots of unsuccessful attempts with the 4-port version, we were finally able to talk to a monitor with 2 ports (running rev. K). The first time we tried with it, it worked right off the bat. The second time it didn't, so we played a little bit with the port settings -- after some fiddling we found that the configuration DtOut1 and GM worked fine, and we were able to talk to the monitor again. However, the LED lights on the back were misleading -- the port that was set up for data export actually had the arrow going in but we were still able to get the data.
Hopefully, next time we go we'll get to test with a monitor with 4 ports again (it's a bit of a hit-or-miss, depending on which room is available when we go there). But it seems that after some playing with port settings and power cycling, we should be able to get that to work as well.
Hope this helps.
Rado
hi Rado - thanks for the follow up. I actually have been working on a direct plug and play driver using C++ instead of Java. Doing this I've learned the MP series will only export wave data to one output - the first that requests it. So if the LAN port is in use then the RJ45 will only output numerics. Also the LAN uses IP/TCP protocol but the RJ45 can do fixed baudrate - which is what we are interested in using . The process for receiving and parsing data is complex and I've hired a professional programmer to tackle it. We can associate (at a baudrate of 115200) and receive data. We're currently fixing some memory caching issues. Overall this is an easier solution for us, but I hope to come back at some point to work again with OpenICE. Best of luck
I see. This is good to know, thanks! We suspected that the LAN port might be messing things up -- maybe that's why we had to cycle power before we could talk to the monitor.
Just out of curiosity -- does your MP have 4 or 2 RJ45 ports? Not sure if there is any difference, but we have only been able to talk to the 2-port variant as of yet.
Hm, do you have the programmer's manual for the Philips monitors? I wonder if it's just some minor revision in the communication protocol that is causing everything to break down.
Rado
2 ports. - one is for the EMR, the second we are using as Dtout2