World's Smallest 3-D Printer

Posts: 4,080
http://www.popsci.com/technology/article/2011-05/worlds-smallest-3-d-printer-could-be-your-first-tabletop-fabricator

At the Vienna Institute of Technology, a group of engineers claims they've created not only the world's smallest 3-D printer, but also one that's so light and inexpensive that it could conceivably pave the way for truly domestic 3-D printing. Lost an earring or a cuff-link? Print one out (and congratulations on your fancy life). That's the future, and it might not be far off at all.
3-D printer manufacturers sometimes think big, but there's just as much of a movement to think small, to bring this sort of fabrication to the masses. Our roundup of 3-D printing dream projects includes both--Enrico Dini may want to put a 3-D printer on the Moon to build houses out of moon-dust, but Hod Lipson wants cheap 3-D printers in every classroom. This project, hailing from Vienna, is more in the second group.
The university claims this is the world's smallest 3-D printer, designed to print with a special kind of synthetic resin that instantly and precisely hardens when hit with an intense beam of light. That gives it the ability to print very intricate as well as very sturdy objects. It uses a focused beam of light, hardening layers only a twentieth of a millimeter thick, which is delicate enough that the university says it can be used to print finicky objects like hearing aid parts.
The team says the prototype is "no bigger than a carton of milk," about 3.3 pounds in weight, and can be sold for 1200 Euros (about $1,700 USD). The size and price are both flexible, and could go down if the printer sees mass demand. No word on whether they'll attempt any sort of mass production, but this is a pretty intriguing look at our future. It's not hard to imagine a 3-D printer on every counter, alongside the food processor and coffee machine. In fact, it's great fun to imagine that.
Reply Quote
Posts: 4,080
Yeah, it's just starting to snowball, but their are nearly unlimited applications for this.

We need to increase the variety of materials that can be printed, I know people are working on that.

My primary interest in it at the moment, is the possibility it has for making the transition to a post-scarcity society.

Also, on a side note, imagine what we can do when 3D tissue/organ printing is combined with the da Vinci surgical robot (people are working on building an AI that will remove even the surgeon from the entire process)?!?!
Reply Quote
Posts: 2,190
Surgical robots will only be guides/aides to surgeons for a while to come (imo). I doubt we will see a robot (without human intervention) removing organs, or performing surgery; at least on other humans.
Reply Quote
Posts: 4,080
05/18/2011 02:21 PMPosted by UberNuB
Surgical robots will only be guides/aides to surgeons for a while to come (imo). I doubt we will see a robot (without human intervention) removing organs, or performing surgery; at least on other humans.


No, actually many people are actively working on fully automating the robotic surgeons with AI.

Consider this, as an illustrative example (there are many more too, and many more to come):

http://robotzeitgeist.com/2010/07/robots-performing-surgery-with-full.html

Robots performing surgery with full autonomy
Robotic arms have been used in medical procedures for a while now, providing surgeons a level of steadiness and precision that few human hands can replicate. Now, however, things are moving forward to a future where these robot arms will be able to perform such operations almost entirely on their own.

At the present we are only talking about biopsies, or dealing with dead patients. A safe way to start, but scientists at the Duke University in North Carolina have already seen these robots achieve a 93% success rate when cutting into prostate tissue. A dead turkey, whose flesh has a similar texture to humans, was used in the experiments. The robotic arms used ultrasound to locate the exact placement of the organs, and then took real-time 3D information which told them what to do next.

The leader of the team, Professor Stephen Smith, explained that the next test they will undertake is to try out the arm on a human mannequin. This dummy will have a “stiff bra cup” with a g!!%* embedded inside, to mimic a cancerous lesion. The robot’s job will be to remove this lesion while following correct medical procedure and saving the person’s life (theoretically). One of the main problems that will need to be addressed is improving the robots’ speed when it comes to obtaining and processing the data from the ultrasounds, but a more powerful processor and a more effective algorithm can help overcome this challenge.

The professor is hopeful that success in these tests will pave the way for a lot more robots doing surgeries on their own, not just biopsies. This would save patients time and money, which is one of the biggest problems in the healthcare industry today. Hopefully, they’ll be able to offer some type of medical guarantee as well.

A brief video showing a medical robot undergoing trials follows.
Edited by BanesidheVex on 5/18/2011 5:10 PM PDT
Reply Quote
Posts: 2,190
Eh, I mean, cutting into a dead person/animal is cool and all, but it doesn't necessarily translate into actual surgeries.

When the patient is dead, you can take all the time in the world to analyze the data and calculate the "next step"; when you have to react quickly to not bleed out your patient, or need to adapt to something different, it is a much more complicated problem.

The technology exists to perform basic operations, but I just don't see a fully automated robot performing intricate surgeries in the near future.
Reply Quote
Posts: 4,080
Eh, I mean, cutting into a dead person/animal is cool and all, but it doesn't necessarily translate into actual surgeries.


You realize why they're doing that right?

In preparation to eventually move on to live people. That's how testing begins.
Reply Quote
Posts: 4,080
05/19/2011 09:40 AMPosted by Astrai


You realize why they're doing that right?

In preparation to eventually move on to live people. That's how testing begins.


Not to mention that the speed of a computer's processing is orders of magnitude faster than a human; and the accuracy is as perfect as the materials allow for it to be.

A properly programmed and trained robotic surgeon would be more precise, faster, and respond more readily to problems it has programming to deal with.


And cheaper. This should reduce the costs of surgery substantially.
Reply Quote
Posts: 2,190
Not to mention that the speed of a computer's processing is orders of magnitude faster than a human; and the accuracy is as perfect as the materials allow for it to be.

A properly programmed and trained robotic surgeon would be more precise, faster, and respond more readily to problems it has programming to deal with.


A computer can perform calculations significantly faster, I agree, but humans can process information and react to situations at an incredible rate. No matter how much computing power you have, a human will almost always be able to understand what they are looking at faster than a computer.

Robotic movement can be very precise, but there is a difference between being able to follow a path perfectly, and having the ability to touch and feel.

I just don't imagine the real-time reaction of a computer as being fast enough to perform surgeries that have a high degree of variability.
Reply Quote
Posts: 4,080
Not to mention that the speed of a computer's processing is orders of magnitude faster than a human; and the accuracy is as perfect as the materials allow for it to be.

A properly programmed and trained robotic surgeon would be more precise, faster, and respond more readily to problems it has programming to deal with.


A computer can perform calculations significantly faster, I agree, but humans can process information and react to situations at an incredible rate. No matter how much computing power you have, a human will almost always be able to understand what they are looking at faster than a computer.

Robotic movement can be very precise, but there is a difference between being able to follow a path perfectly, and having the ability to touch and feel.

I just don't imagine the real-time reaction of a computer as being fast enough to perform surgeries that have a high degree of variability.



bwahahahahahaha
Reply Quote
Posts: 4,080
All you have to do is watch this video:

http://www.youtube.com/watch?v=bfdHY26E2jc

One of the most frustrating parts of all this, is bringing people up to speed ;)

C'mon, human species, let's get with the program.......we're already past these kinds of debates =)
Reply Quote
Posts: 2,190
05/19/2011 04:52 PMPosted by Astrai
All you have to look at in order to see how wrong is robotic assembly lines; they perform complex and demanding tasks which require a high degree of precision very much faster, more accurately and more cheaply than a human assembly line could. And that's with technology which is already decades out of date.


There is almost zero variability, if any at all, in an assembly line. Any variability which does exist is a controlled variability. Same thing for the example(s) in the YouTube video; they are tracking one object which is significantly easier to isolate than anything in the human body.

I am not arguing that the kinematics is lacking, by all means robots can do exact as programmed (which makes them extremely efficient in environments with low variability).

They aren't ready in the realm of sensing, which is what humans do a far better job at than any computer, assuming a "real world" environment. This is why I say they are still a ways off from performing complicated surgeries.
Reply Quote
Posts: 2
Uh, so...lets see here, Astrai and BanesidheVex, you are comparing a three variable system to the human body? THAT is your proof? I am sorry, UberNuB is totally right here. At the current time, computer technology is not where you believe it is.

While the robot is nice in the video you showed, it essentially amounts to a closed system with only an X, Y and Z component. Yes, it does it fast, that does not mean it is doing the calculations at lightning speed, it just means it has a programming algorithm to predict the X, Y and Z accurately and can process the few snapshots of data it receives quickly.

Believe me, I am currently working on a handwriting analysis patent for a major US corporation. The problem, is that human handwriting is too varied. The process works, in much more terms that I cannot discuss fully, by stripping the handwriting to 2 pixels wide, and taking any "shift" out of it, and straightening it. Then, normalizing it to a size in which it is used to process. The best accuracy is only about 80%, and that is with a limited number of words to recognize.

Yes, computer technology is an amazing thing, but it is NO WHERE NEAR being able to handle the complexities of a human body. Facial recognition software is just finally becoming reasonably accurate (Most good ones will have about 80% correct in real world circumstances). When you cut someone open, you can KNOW what you are looking for/at and still be confused by the body. People are all different and their organs are not the same size, or in the same places exactly. Not to mention, a cancerous/diseased organ looks, in some cases, ENTIRELY different from a healthy/normal one. It would take decades upon decades of data, and a ridiculous amount of man-hours to allow a computer to understand the knowledge of a human surgeon.

Think about every little part of your body, they all work in harmony in some way, shape, or form. Now think about a family member, and how you are different in size and shape from them. It isn't only aesthetics on the outside, it extends to the inside. One would have to take a 3D CT basically with radiocontrast injected into the blood before every surgery, that data then uploaded to the computer. While that is reasonably possible, not everyone can do that, or afford that at the current time, nor would that dose of radiation for every surgery be healthy. Then, you have the 3D image of the body, but that is only true at the ideal condition of the CT scan, when things deviate, as the human body likes to do, everything gets shot to hell.

Tl;dr, UberNuB is right, human body=too many variables.
Reply Quote
Posts: 2
The robot surgeons you speak of are actually CONTROLLED by humans, they are not robots in the programmable sense. Thus, that argument is not applicable. They are basically remote controlled drones used for surgery that create smaller incisions and smaller actions done a surgeon cannot do adequately/with as great as precision.

Yes, the technology exists, the technology has existed since the first ever action was done via robot, it is just an intense series of commands and variables needed to be programmed. A feat that would take DECADES to accomplish, and has not yet been finished.

The method is not in question of the way to teach them, it is the way to allow them to compensate for the variables. One can repeat an appendectomy ad nauseum and allow a computer to capture where to make the incision, how to get through the layers of tissue and muscle, but when for some reason a patient has an adverse reaction to the anesthetic, a vein is cut, an organ is not somewhere it should be, it is a iteration of the system that the computer does learn more, but it cannot predict with any accuracy where the "normal" is, and when to use the "normal" and when not to use it.

The technology exists to program it, but the problem is, there is no real way to program it with any confidence to be marketable. Thus, you cannot say that merely by teaching proper input it may be accomplished. As proper input is not possible due to the variables in the human body. Which, is the crux of my (and UberNuB's) argument. The string of the discussion was not missed, you are the one who failed to pick up the opposite side of the argument. This is not merely a simple one side argues X and the other side argues the opposite of X. This is a X vs X is not feasible argument. Big boy arguments tend to not be strictly negative defenses.

Way to completely and utterly miss the spell check button.
Reply Quote
Posts: 8,700
Awesome!! I've been wanting a 3d printer foreverrr! :D
Reply Quote

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)

Reported!

[Close]