neuron text book and being disabled

[Jake_Trexel]

I am trying to find people that are willing to help me in my project of writing a text book,"The Mathematics of the Neuron".

I am disabled and home bound. Because of my health, I no longer have fellow faculty members to talk to. It has been since 1992 that I last worked as a professor of Mechanical Engineering. The internet is great, but it has its limits.

I am looking for help in the use of Mathematica, mathematics, and bio-physics. For example, how do I use Mathematica to solve the Hodgkin-Huxley Equations? Even with the demonstrations, I can not follow them because I have no books that match the equations that are being used.

Jake

http://www.maplesoft.com/applications/v ... &view=html

[Justin_Smith]

I am trying to find people that are willing to help me in my project of writing a text book,"The Mathematics of the Neuron".

I am disabled and home bound. Because of my health, I no longer have fellow faculty members to talk to. It has been since 1992 that I last worked as a professor of Mechanical Engineering. The internet is great, but it has its limits.

I am looking for help in the use of Mathematica, mathematics, and bio-physics. For example, how do I use Mathematica to solve the Hodgkin-Huxley Equations? Even with the demonstrations, I can not follow them because I have no books that match the equations that are being used.

Jake

The Hodgkin-Huxley model is a set of ODEs. You can find tutorials and overviews on solving ODEs in the documentation center (hit F1 or go to Help-->Documentation Center). Try typing tutorial/DSolveOverview into the search bar as a starting point.

[Jake_Trexel]

Thank you for your reply. I am not having a problems and solving a DE's mathemtica, but I am with the make up of the parts of this equation, please see the web site below.

For example, solving the equation for the initial values of "n,m and h", in the equation :
I= (Cm x dV/dt x gbarK x n^4 x (V- Vk)) +((gbarNa x m^3 x h x (V - VNa)) + (gbarL x (V-VL))

I calculate the gating parameters and rate constants, but then I have to find the the 3 unknowns with one DE, the n,m and h

My final equation, which I copyed and pasted to this post is
v'=(3.6*n^4*(v-Subscript[v, K]))+(120 *m^3*h*(v-Subscript[v, Na]))+(0.3*(v-Subscript[v, L]))

The paper that I am using came from the net, and the author said that he ran Turbo C on a XT computer and left it at that.

How do you come up with the values? Is it because they used experimental results or what?

Please look this site
http://fohs.bgu.ac.il/nia/NIA2003/neuro ... uivcrt.htm

Thanks

[telefunkenvf14]

I am disabled and home bound. Because of my health, I no longer have fellow faculty members to talk to. It has been since 1992 that I last worked as a professor of Mechanical Engineering. The internet is great, but it has its limits.

Jake:

Do you use any "speech to text" software, such as Dragon Naturally Speaking or Windows Speech Recognition? I've tried using both with Mathematica and have been frustrated with the results. Basically, I can get them to input only certain types of things, such as creating new sections, but actual dictation doesn't work well at all.

At the very least, I think WRI could take advantage of Windows Speech Recognition by including a custom 'dictionary' with Mathematica so that Windows Speech would automatically recognize the built-in functions and symbols. This would also dramatically improve the usefulness of the handwriting input---currently it's a pain to use because the Windows math recognizer is only set to recognize 'correctly formed mathematical equations'.

-RG

[Justin_Smith]

I am disabled and home bound. Because of my health, I no longer have fellow faculty members to talk to. It has been since 1992 that I last worked as a professor of Mechanical Engineering. The internet is great, but it has its limits.

Jake:

Do you use any "speech to text" software, such as Dragon Naturally Speaking or Windows Speech Recognition? I've tried using both with Mathematica and have been frustrated with the results. Basically, I can get them to input only certain types of things, such as creating new sections, but actual dictation doesn't work well at all.

At the very least, I think WRI could take advantage of Windows Speech Recognition by including a custom 'dictionary' with Mathematica so that Windows Speech would automatically recognize the built-in functions and symbols. This would also dramatically improve the usefulness of the handwriting input---currently it's a pain to use because the Windows math recognizer is only set to recognize 'correctly formed mathematical equations'.

-RG

Are you using Windows 7 handwriting recognition? In terms of getting speech recognition to work properly, it seems as if that can be generally customized entirely on the speech recognition's side:

http://www.pcspeak.com/hints/general/vo ... ists.shtml


Jake,

You may want to check with the paper's authors.



- Justin

[telefunkenvf14]

[Are you using Windows 7 handwriting recognition? In terms of getting speech recognition to work properly, it seems as if that can be generally customized entirely on the speech recognition's side:

http://www.pcspeak.com/hints/general/vo ... ists.shtml

- Justin

Dragon Naturally Speaking kinda works, but unless you pay the INSANE pro price, they pretty much lock you out of useful macro features that would be needed to execute commands, such as keyboard shortcuts in Mathematica. TRegardless, there is something funky about the dictation into the MMA front end---it never capitalizes right and correcting recognized text is nearly impossible.

Windows speech (on Windows 7) is better in some ways---there is a FREE tool for building macros, enabling one to vocally execute keyboard shortcuts in Mathematica (and they do work ok). However, regular dictation into text cells (or any other cells) remains completely unrecognized. I think it's because Mathematica isn't fully connected with the MS 'Text Services Framework'. (recalling name from memory... name may be slightly different )

The fallback mechanism in Windows Speech (when an app isn't connected with the framework) is supposed to be some sort of hidden copy/paste scheme, which, based on what I've learned about Mathematica, is probably being bungled up in Mathematica's own hidden clipboard for copying/pasting.

Based on all I've read, the easiest thing to improve may be the Math Input Panel. Perhaps WRI could kick out a list of all symbols and named functions and either force the Math Input Panel to recognize them as OK (i.e. 'correctly formed mathematical equations') or just generate a new vocabulary for the regular handwriting input panel. I've read some on how to do this on a mass scale...just don't have the time. (or the ability?---this would amount to debugging in two hidden/unfamiliar directions. Tough job without guidance.)

If someone from WRI wants a Tablet PC guinea pig... call me! (also up for working on Windows Speech Recognition)

-RG