|
Okay, somthing new. I have designed, and developed a prototype, medical robotic platform. It has the capability to take Blood Pressure, test diabetics, and even do EKG's. All of the medical reporting equipment has been verified by Doctors, and Nurses, as being accurate. The Device is being built as an interface between Veterans and the Veterans Administration medical center. I call it Val or Valerie. The acronym is V.H.A.L. which is [V]eterans [H]ealth [A]ssist [L]ink. I am now starting to build the voice menu for the Robot. The unit has a Kinect sensor which means it has a microphone array. My problem is this. I would like to build a limited amount of A.I. along with the robotic commands for movement. With the Kinect unit I also have the capability for voice distinction along with Facial feature recognition. I need a kick in the butt to get going. I would appericiate any help from any source if there is anyone out there that would care to help.
|
|
|
|
|
David P Smith wrote: I need a kick in the butt to get going.
Start here[^]. Write us a nice article on the technology and how you used it, and we won't kick too hard 
Bastard Programmer from Hell 
if you can't read my code, try converting it here[^]
|
|
|
|
|
Not to kick you while you're demotivated, but, there are some big challenges ahead of you. I tried to do some similar product ideas in the health arena, and there are two things that will quickly become the bane of your existence. Just wanted to share these with you so that you don't get surprised later on:
1) FDA Medical Device Clearance: If it's a device, and used in health care, chances are that it needs some type of approval/clearance. If it touches a human in any way (like blood pressure, EKG's, etc) you are guaranteed that you will need approval before you can sell or make claims. Here is a resource to help: http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/default.htm[^]
2) Health Insurance Payment Authorization: If Doctors/Nurses can't charge it back to an insurance provider in some way, good luck getting them to use it. First, you will need the FDA approval on the device. Then, you will need to hunt down people who can make these decisions (if they actually exist). Many of them will say that they just follow medicare payment guidelines for DME (durable medical equipment) payments. So then you try to get a determination from CMS (center for medicare/medicaid services)... HA! Then, you will need to make your case as to why your device is better than all the rest, and how it will improve patient outcomes. To do that, you need clinical studies... to do that, you need doctors interested in actually being a part of the research (it is quite a commitment), then you will need funding for the research, so you begin writing a grant proposal. Then, you will need to buy a wig because you have pulled out all of your hair.
Also, on the payment front, if you try to make the argument of physician oversight billing for time spent reviewing results from the device, you may be out of luck. Physician oversight payments for tele-health are pretty restricted.
Be The Noise
|
|
|
|
|
Hello there!
I'm a masters degree student from Norway. For my thesis I have developed four Android 4.0 apps. These apps are similar to discussion forums. The apps have different design solutions, and I'm trying to find out which design is the best for making good discussions.
In order to do this, I was hoping some of you would want to try out one of the apps! Basically, I am planning to have a "test period" for about one week, where people can use the app for discussion. After this I have a quick survey with some questions relating to the application, and the discussion itself.
If you would like to participate, please let me know by sending an e-mail to debatortest@gmail.com or leave your mail adress here so I can get back to you.
Cheers,
Audun Sørheim
|
|
|
|
|
I want to share a project idea hoping to get feedback and possibly serious collaborators who are ready to give it a try in hope it can bring concrete gain and success.
The idea is easily understood given the following parts of it :
- A rules execution server
- A .Net UI component for rules creation
- A .Net component to communicate events/facts to the server and to receive rules execution responses.
So it is easy to see that this new approach aims at helping developers easily integrate expert reasoning in their applications. they will have to deploy the server with their application, integrate the editor in their UI and wire up the server with their app as follows :
- communicate events when they happen.
- listen for actions triggering and do the needed jobs.
- when user creates a rule, just save it in the engine.
So the developer does no longer have to change his code when the end-user asks for a particular logic, and he gives maximum freedom to the end user to make sophisticated treatments.
I can guarantee the server part, i hope to find collaborator on the .Net side.
Thanks!
Push Framework - now released !
http://www.pushframework.com
|
|
|
|
|
Ahmed Charfeddine wrote: So the developer does no longer have to change his code when the end-user asks
for a particular logic,
Wont work. The reason is simple.
To solve complex problems complex logic is needed. Thus programming languages exist. Complexity also allows for more ways to create errors. Complex errors.
And that is why programmers exist.
A general solution would either be so simple as to provide no benefit or so complex that the "users" would need to be be programmers. Thus losing all possible benefit.
Now if you have a specific problem domain and extensive expertise working in that domain (and not just programming for it) then you might be able to create a 'language' that is intutive for most users of the same domain. Then you can wrap everything else you said about it around it.
Other than that you can just hack it as a learning experience. If so I suggest you keep the language simple. You might want to look into some of the existing gaming systems such as the ones that let you program 'robots' (software entities only) to fight against each other. The language is simple and there are existing implementations that let you constrast against. I am certain that there are other types of games systems like that as well.
|
|
|
|
|
He's talking about Business Rules Engines. There are quite a few commercially available ones available right now that do just this. I worked on one back in the early 90s.
|
|
|
|
|
Pete O'Hanlon wrote: I worked on one back in the early 90s.
Presumably you are referring to a generic/general system and not one that was specific to a industry/company.
By the way I created a "business rules" engine in the 90s as well. And worked with another commercial one then as well. And another system that could be deemed that as well in 2000.
All of them required developers to create the rules. Or specially trained operators who spent all of their time doing that which certainly suggests a programmer to me.
Pete O'Hanlon wrote: He's talking about Business Rules Engines. There are quite a few commercially
available ones available right now that do just this.
And how many sales people create rules? How many HR managers? How many call center employees? How many clerks?
|
|
|
|
|
I knew the answers will start to come as when the sun reaches America.
Well, jschell, as Pete said, this is feasible. People have done this before. But there is a new approach.
The solution will be very simple, both for the end users and for the developers as well.
Here are some example scenarios (our application will be named S) :
Example 1
Company A develops an emailing applications. To provide its end users with the ability to make sophisticated treatments, A integrates our solution S this way :
it adds the rules editor to the front end, it publishes all sort of events/facts, when they occur to the engine part of S (incoming email, with properties, contacts with properties..)
and adds code that listens for actions triggering by engine so that it fulfills them.
The developer will describe to S, what is the set of possible facts/events and their arguments, also will describe to S the set of possible actions.
As you can see the developer did not puzzle himself how to create some langage or parser. To the end user, I imagine he can do this :
When all of the following is true
There is an email E from a person X
Person X belongs to contact group G
Then
Move email E to Folder named G+"Emails"
Example 2
Company B develops an access control system that allows other companies to restric access of employees to secure zones. After integrating S, now, the end user can write this rule :
When all of the following is true :
There is access request from reader R by person P1 (the system should interpret the card id to a person id)
Reader R is an input door to zone Z
Person P2 is owner of zone Z
Person P2 is within zone Z
Then
Open door attached to reader R.
This rule allows the system to grant access to a zone only when its "owner" is in it.
End users can come with many rules like this.
So the company will take care of the hardware communiaction part plus some interpretations of the signals, then it sends evrything to the engine and reacts when the engine triggers some action. End users will be more happy this way.
Is the idea clearer now ?
What are your thoughts ?
Thanks!
Push Framework - now released !
http://www.pushframework.com
|
|
|
|
|
Ahmed Charfeddine wrote: The solution will be very simple, both for the end users and for the developers
as well.
I have created a rules engine. I have used several others. I have created custom solutions. I have created pseudo work flow systems and real work flow systems as well. I have created scripting languages and created solutions that were delivered with scripting languages. I have created dynamic data driven systems. And of that lot I was the designer on a number of them.
So I have a fair bit of experience in creating dynamic solutions that match business models.
Ahmed Charfeddine wrote: As you can see the developer did not puzzle himself how to create some langage
or parser.
What makes you think that a parser is necessary to solve this problem?
Ahmed Charfeddine wrote: it adds the rules editor to the front end, it publishes all sort of
events/facts, when they occur to the engine part of S (incoming email, with
properties, contacts with properties..)
Where exactly do those come from?
Ahmed Charfeddine wrote: Reader R is an input door to zone Z
Person P2 is owner of zone Z
Person P2
is within zone Z
Where does the physical access to the door come from? Specifically how exactly does the door get opened by the software? How exactly does person P2 come to be within Z? How exactly does P2 come to own Z?
Lets take a real scenario rather than an imaginary one.
- A call center employee makes a service call to an account.
- The employee is tasked with getting a payment for the account (the task.)
- The employee can have a number of contact numbers along with comments about each number written by previous employees (usually about how 'good' the number is, what the person is like, when available, etc.)
- Certain payment options require a manager override which can either be initiated by email (task ticket) or personal presence.
- The time spent on the task must be measured in detail.
- Each person (employee and/or manager) must be tracked as to actions on account.
- Resolution and or further action must be recorded.
- Updates to at least two systems can occur depending on resolution. One system is a standard database the other is a custom database that requires a screen scaping interface.
- Real time monitoring of the call center app in use by the employee must be possible by one or more managers.
- The call center app must also act as a time clock keeping track of employee arrival, departure and breaks.
Some background - the call center managers bonus is based on revenue received and minimal time spent on calls. Thus the manager spends their time insuring that the employees are doing their job and making sure that when a manager override is needed that they are on the spot to do it.
So how to you see your system solving the above problem?
Exactly how are you going to convince the manager, the one mentioned above, to take time out to learn your system, create with it and debug with it, given that doing those actions guarantees that the bonus will be negatively impacted (because they won't be doing the part that actually drives that)?
How do you see a manager that is good at this specific job, managing call center employees, is going to be equipped with the skills necessary to create business rules, debug them and even for that matter be able to logically derive new ones using nothing but your system?
|
|
|
|
|
Hi, and I appreciate a lot your knowledgeable answer.
I don't agree with the example you gave though, and I think it is just because my idea is still not clear. Those are to me static specifications of how the application is going to work, and I would not recommend to solve it using business rules either.
It is rather about applications where, at certain stage of functionality, we want to allow end-users the ability to specify custom treatments, but since the possibilities are numerous, we seek a sort of a solution.
Push Framework - now released !
http://www.pushframework.com
|
|
|
|
|
Ahmed Charfeddine wrote: It is rather about applications where, at certain stage of functionality, we
want to allow end-users the ability to specify custom treatments
Which is business specific and/or company specific and it requires an infrastructure to support the the allowed rules in the first place.
And much of my previous example could have been handle dynamically via a work flow system. However that doesn't mean the call center manager (the "user") would have implemented it.
|
|
|
|
|
Just released the latest alpha version of TakeStock 2 (a powerful, easy-to-use and completely free personal portfolio manager). Would appreciate any and all feedback, suggestions and bug reports. Thanks!
TakeStock 2 blog[^]
/ravi
|
|
|
|
|
I'm trying to put a group together for a side project. No expected gain from it.
I have a dedicated linux server at my house for it. Right now I have gentoo on it, but I can do a different distro... not that it matters 'that' much to the dev team. The system scripts will be in perl/bash anyway.
We'll be using (probably) apache, node, jquery, mysql. Not opposed to php and/or python if we get people who want to use that (I don't know python yet though >,<). I'd like to use mongo but if that is going to happen I'd need someone more familiar with it, otherwise I'm going with mysql 
Some of the things we'll be making from scratch:
On site IM and mail/messaging.
Blog and forums
collaborative editing
calendar/schedular
chat server for chat rooms
job posting/search
We havent ruled out any technologies yet. So far it is just me and one other person. (neither of us are designers). We just want a fun project where we get full control. If you are interested and want another tech, we are open to that.
If it moves, compile it
|
|
|
|
|
If you're looking for an easy-to-use utility to keep track of your investments, I'd appreciate it if you gave my app TakeStock 2 a whirl. Completely free and actively supported since 2002!
TakeStock 2 Blog[^]
Thanks,
/ravi
|
|
|
|
|
I have an idea for a phone app. It involves motion detection and also some speech to text or text to speech.
Anone interested in partnering, or if you're willing to provide assistance alone the way, drop me line.
Thanks
If it's not broken, fix it until it is
|
|
|
|
|
What languages do you plan on using?
|
|
|
|
|
Probably WPF in C#
If it's not broken, fix it until it is
|
|
|
|
|
If it's a phone app, it's not going to be WPF.
|
|
|
|
|
Windows Phone?
It will be developed for Windows and multiple mobile devices.
If it's not broken, fix it until it is
|
|
|
|
|
Windows Phone is Silverlight or XNA. No WPF there.
|
|
|
|
|
I can help if it's an Android one
|
|
|
|
|
Ive been working hard on my site for my webdesign business and i wanted to get some feedback as to how you like the site(layout,functionality,ect..)and anything i could do better. I appreciate any responses! 
Sagedread Designs Site
|
|
|
|
|
1: Dislike the main graphic and font: looks a bit harsh and wastes a lot of 'selling' room.
2: The font for the blog is quite hazy and hard to read: font choice is questionable.
3: The main advert has some spelling and grammar mistakes.
(Viewed on IE9)
Other than that it's okay though it wouldn't make me want to use your services: if you're a web designer and your site is not 100% why would I pick you to design my web site? (Realize that is a bit subjective but you did ask).
"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." Red Adair.
nils illegitimus carborundum
me, me, me
|
|
|
|
|
Yes i agree the logo is too big. Ive fixed that as well as the font issues. Thanks for the feedback.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
