Human Replacements

Hey guys, I am a computer engineering student hoping to get my MS /PHD in Nano-engineering/Biomedical engineering. With all of the developments that are coming out both in the medical field and now engineering is being incorporated I have a theory that within the next 10-20 years humans will be able to expand their lifespans by 40-60 years.

Idea:
You have self replicating Organic Nano-bots that live off of the environment around you.Thermal,etc. They are constantly fixing,repairing and regenerating new cells so that your body doesn't slow down and age.Without them of course you would, but with them you would age very slowly. Helping fix organ failures and kill deadly bacteria. I think NB's are the way of the future.


Second proposition is having bots that can sustain themselves long enough and then they die/poop them out and then you end up consuming them again.?

What do you think about the idea .
I know its grandiose and seems really sci-fi but what would I need to take into account ?

Other urls found in this thread:

youtube.com/watch?v=Cw5llLUrawc
pubs.rsc.org/en/Content/ArticleLanding/2012/CS/c1cs15262a
twitter.com/SFWRedditImages

?

Sure go ahead and make a NB with the capacity to do any mechanical work.

I'm of the opinion it's easier to program bacteria to do our work for us instead of making NBs.

Ok so how do you make self replicating Organic Nano-bots?

>> live off of thermal
No.

>> Fixing, repairing, and regenerating cells
How?

Ok go ahead and program bacteria to do this. First off, how are you going to get them to avoid the immune response? Second how do you coordinate them to get stuff done?

This sounds like an idea made by a 14 year old with no actual experience with Nanotechnology

>>/out/

Seconded. I work on nanotech(no not the glorified material science kind), and the field is way way way far off from making anything like OP described.

One conference talk I saw was on proving that this tiny nano-mechanism moved at all. This took a lot of very difficult and expensive lab work, all just to prove that it moved.

>being black

>You have self replicating Organic Nano-bots that live off of the environment around you.Thermal,etc. They are constantly fixing,repairing and regenerating new cells so that your body doesn't slow down and age.Without them of course you would, but with them you would age very slowly. Helping fix organ failures and kill deadly bacteria. I think NB's are the way of the future.

That's a Star Trek-tier popsci idea. If you have zero idea about the possibilities with curernt tech and the state of current tech then you probably won't even be a good CE.
How are you going to do NE or BiomedE if you don't even at least autodidactically learn things to avoid stupid shit like this. Take a look at bioinformatics so you can realise how complex a cell is.
Whether something is feasible depends on what you can do with current tech and how you can improve it and not on a popsci tier good idea.

>I know its grandiose and seems really sci-fi but what would I need to take into account ?
Around 1% of currently known shit and discovery the remaining 99% shit required.

>starts out by telling everyone about himself
Yeah you're not going to get anything done.

Actually everyone told me about the hurdles of doing this. Not sure if by the time I get done with graduate school that this will be an option. I am pretty sure everyone here is a notable scientist with good enough credit and not some loser fag who barely passed calculus.
But for those who put their honest opinion/knowledge I really appreciate it.

I'd rather say my field of study and goals so someone could gear me in the right direction.

I think Im pretty solid in CE nothing too hard. My goal is to dedicate time into the research side of things. I'll take your advise and look into bioinformatics.

Actually im Asian lol dont you see my long curry hair.

youtube.com/watch?v=Cw5llLUrawc

This guy has gone full retard.

wat. I'll write a more detailed response later if this thread is still alive, but I would recommend that you don't go into bioinformatics.

fucking nigger faggot

hold a strong magnet up to body and there goes your cute little nanobots you stupid nigger

Can you link some contemporary papers? I'm also interested in nanomedicine.

Also what is the current standard for building nanotech? Interested particularly in drug delivery. Are we able to build atom by atom or is chemistry the gold standard?

> (You)
>Can you link some contemporary papers?
"Great Expectations: can artificial molecular machines deliver on their promise"

Will put link later.

>> nanomedicine
That term is a fucking joke. Same with the term nanobot.

>Interested particularly in drug delivery.
Boooooooorrrrrrrriiiiiing and too applied

>>Are we able to build atom by atom
No. It's still a pretty big question whether we can even do this at all. Answering this question will take a lot of difficult and expensive lab work.

Protein engineering is starting to become a thing. People are starting to design proteins with specified 3d shapes just like we have been able to do with DNA nanotech

>Why don't engineers and nutritionists collaborate to make a perfect food machine?

>I just want my perfectly healthy and tasty cubes of food made by this machine that only needs my choice and prerequisite materials to be put inside.

Better make me my food machine :3

So here's the deal OP, what you have proposed is a common sci-fi trope. As you have not explained how said nanobots would actually accomplish these things, it
is no different than if you proposed to use magic instead of nanobots. Nothing good comes of discussing 'magical' solutions.

Second, we are way off from building such things. We can't build stuff up atom by atom and our synthetic chemistry isn't good enough to make complex machinery. It is an area of active debate whether we can build things atom by atom. No one has demonstrated this, not even in the lab.

Now that you're discouraged, there have been some interesting developments recently. Synthetic chemists are starting to make simple molecular machines and mechanisms. We've had molecular motors for a while, now people are starting to put them into simple systems.

The next big step appears to be organizing molecular machines into periodic structures. Fibers(1D), sheets(2D), and crystals(3D) are much easier to make than specific 3d structures.

Buckyballs and nanotubes are almost exclusively produced in gram quantities, Metal Organic Frameworks(MOFs), a special class of crystal, are produced in tonnage quantities.

It is definitely worth following recent developments in supramolecular chemistry.

We have almost figured out how to proteins with arbitrary geometry. This is pretty big. We have been able to do this with DNA for a while, only the issue with DNA is that it's very fragile. Proteins are not. DNA has a stiffness of thin gelatin, protein has about the stiffness of fingernails/crab claws/rhino horn.

So what can you do? Artificial intelligence for chemistry/materials science is getting pretty hot.(I work on this) IE searching for materials with specific properties and what not.

Computational chemical reaction planning is another emerging field. If we have a way to quickly find high yield reactions, we can make more complicated chemicals.(pic related chematica) If one can synthesize sufficiently complicated chemicals one can make nano-machines.

The aforementioned protein engineering work involves a fair amount of computational stuff.

I would also caution you against going into bioinformatics.

here is the link to "Great Expectations: can artificial molecular machines deliver on their promise"
pubs.rsc.org/en/Content/ArticleLanding/2012/CS/c1cs15262a

Hi I'm doing a PhD on machine neuron interfacing from a computer science perspective. Your idea is beyond retarded and you should take into account that superhero movies are not scientifically accurate.

Cool? How many electrodes does your microarray have?

60, but honestly we could do with far less

>environment around you.Thermal,etc. They
Babby tier thermodynamics tells you this is not possible. Stick with CS, it's what dumb people are best at.

Is it noninvasive?

What do you goys think of mechatronics?

It's in vitro. I'm not that interested in brain science desu, I'm interested in how neurons self organize to form adaptive and robust networks for solving tasks.

Check out multichannel systems MEAs if you wanna see what I work with.

Nah just wanting more details. That's preddy cool m8. It fucking sucks that we still don't know how real neurons work.