 So thank you very much for coming. My talk today is about how I copied design that is used for surgeons into a 3D printable design. And before I go into the details about the part I'm talking about, I wanted to talk about why I'm doing this. So the first, there's a problem that some countries or some places in the world have access to a 3D printer, but they don't have access to medical devices. So actually, they could print their own medical devices, but they can't, because right now I'm telling you that the project I'm doing has been done before. It has been done by university, but they published a paper, didn't publish the files, and nobody can actually print those items. The files are not available. So goal for this project is to make it open to hardware, to make it available for anyone in the world to print out and actually use this technology. And on this project, like any project, I'm learning a lot about things. On this project, this was about how different materials react and how can I change them apart from a different material to another material, and what are the steps that I need to do to change them according to the material properties. And of course, I was getting paid for this, so I was very glad to know about that. So this is the guide of the presentation. Where do you want to go? Where do we want to get to? What do we have and what it took to get there? So let's start with the first point. Where do we want to get to? So this is a hemostat. So what does it do? I don't have any formal medical training, so I'm going to explain it with my words that how I understand it. So it's used to clamp things, especially blood vessels. And this is used to stop you from bleeding out while being operated. So you can probably imagine how important this is for a surgeon and probably more important for the patient. And this is to be used for this thing out of metal should be made in plastic, 3D printed. And yeah, so the tools I have at hand. So I'm working with open source software only. What I'm using is OpenSCAT. That is basically a program that turns code into a 3D model. But that wasn't enough for me for my purposes. So I previously made this thing called Crystalscat that enables me to program complex models in Ruby. And this produces OpenSCAT code, which then produces the 3D models. Then I have my editor of the choice, G-Edit or G-Mate. I used GIMP. I will show you in a bit what for. And I've used my 3D printer. And now to what it took to get there. So we have this thing. And before I go on, I will talk a bit about how testing of this works when I do the design. Or how does the design process work? And I'm using a thing called iterative design. That is what it means is that I program a thing. I test it, like I printed it up in my 3D printer. I inspect it. I learn from the mistakes or the things that are actually good and make changes to that design and repeat. So I have a little prototype counter on the bottom left. And maybe you can start making guesses on how many prototypes I needed to copy the function of this thing. So another thing is that it doesn't need to look the same, but it needs to have the same function. So another bit about testing. There are three stages of testing. The first one is that I print it out. I test it in my hands and see or try to figure out how it works and think about does it work like it should work. Then there is my employer who is a doctor. He prints it out and gives me feedback about it. Does it work for him? What does not work? The third stage is he gives it to an actual surgeon or many of these and get feedback from them. So I will talk about at a few stages, I will talk about where I went in that. Many prototypes I did just for my stage one because I saw a lot of errors in my iterations. And I think I can go on higher. There's just one thing. Why couldn't I just copy and paste this thing? So there are 3D scanners around. So they could basically copy the outer shape of this object. They could make a really nice copy of that. But they can't go into the inner details about this object. So for example, this thing has a hinge inside of the object. And a 3D scanner cannot scan these parts. So when you would send this to a 3D printer, just like from the 3D scanner on, you can get maybe the correct shape, but it wouldn't have a function. It would be just fixed and not movable like it should be. So to add this thing is made out of steel. So I talked about I needed to change material properties. So this thing is quite thin. Look at the arms on here. They are maybe, I don't know, a few millimeters thick. And if you make the same out of plastic, it will just bend or snap and not apply any force to the actual tool head. So let's start with my timeline. On April 9, I've got a 2D scan of this stuff. So why a 2D scan? So on the bottom of this picture, there's a ruler. And this gives me the two-dimensional, this makes me able to measure it in two dimensions. And the way I did it was I will open my GIMP program, copied the ruler around, rotated it, and took measurements. So I was able to do a basic shape of the object I was making before actually having access to the real tools. So then I decided to just start with the thing that most of the parts have. This was the hand grips. I will not go through this code here because you can, on the end of the presentation, there's a link to my GitHub. And you can download the code there and see it in detail. On the progress of making this, I improved my cat program, my Ruby 2 OpenSket program, add basic things like a ruler to measure what I'm doing. I'm just skipping through there quite quickly. I used the library to make the arms. It was really simple in code. I've made a simple hinge and a simple tool hat. This was also very little work. And I printed it out just for, I know it didn't work properly, but I just printed it out and see how it feels in their hand and the dimensions somewhat match. And would I be able to use it? And so I got the first prototype out and it has a working hinge. Well, the tool hat doesn't grip. The arm shape was a bit off, but I didn't know better this time. It doesn't have any locks, but actually you could hold it in your hands and see what's going on. So I went on and I got this. So I've got the picture of the locking mechanism on the scan and it didn't tell me much. So there was these three pins there and I couldn't figure out how they are wanged. It looks like they're pointing upwards. So I figured out I can probably make this as simple as stacking some cubes along the cube and let it lock from, make it able to just lock by pushing on it. So I put it on my cat model. I was trying to print, or I was printing out and trying to use it and, well, what could personal we go by? Disaster broke. Well, not a big disaster because prototyping with a 3D printer is a really, really cheap thing. So this didn't cost me a fortune. Just a few cents of material cost and time wasted. So not a big loss. So in my code that was really quick fix actually to make it stronger. I tried to make every parameter parametric. So you could just change the width, the height and the pin count and make another prototype in a few minutes. But before going on, I was also working on getting a better grip of the tool head because the previous prototype resulted in a really flat surface along the tool head, which didn't provide much grip. So I deleted a few cylinders along the sides of the tool heads. And I was sure this wouldn't print as it were on the OpenSCUT model because my printer's line width is a bit higher than what's shown here. But actually, I got a server that wasn't smooth out of it. So that was a plus. So this is number three. I've got a working hinge, a better working tool head for my thinking, some arm shapes that I didn't know how it worked at this time. And I had a locking mechanism that actually worked. It didn't break. But at this point, I sent it to my employer and he printed it out and said to me, this is not quite right. There are some things that doesn't really work like Legitude. Maybe wait till you get the actual model and see what's going on. But this data was already on the mail. Three days later, I've got a nice package getting all these tools. So let me get a sip of water for explaining. So I've got some tweezers, some scissors. Then again, from the top there are two pairs of tweezers. Then there's a towel clamp. Then there's another pair of scissors. Then there's a needle driver. And on the bottom, there's the Amostad that I was working on first. I will get to the details of the other parts later or some of them. So I had a closer look at the actual Amostad out of steel. And as you can see on this slide, if you look at the arm structure, you see some major differences. So what I didn't get from the 2D scan is how it actually operates, how it functions. So what I didn't know is that when it's not closed, the metal tool, it already closes the front of the front grip. And the more you push, the more pressure it applies to the front tip. So on my prototype, I was thinking about when you push it all the way through, it should close. So that was obviously wrong. But I didn't know better. I didn't have it in my hands at this time. So there was another thing. And this was the locks. So I made a close-up on the locks, which is on the actual Amostad, which is on the left. And as you can hopefully see there, it's somewhat angled. So what happens if you just push the grips in is it locks automatically. So mine was just straight teeth. And it wouldn't work that way. So I went on and made a rough implementation of the original. I'm not showing you the code here, which is a bit too excessive to show you. Again, my GitHub link will be on the second to last slide. The next thing what I did is I made the arm shape differently. This was not that much of a big mod. So I just did that in a hurry, I must say. And I printed out another prototype. So if you look closely at the arms shape, you might see that there's something wrong with that. So I messed it totally up. What it resulted in was that it was bending in the direction of where I'm pushing, which resulted in an incredibly weak grip on the tool head. So OK, I was frustrated. On this day, I fixed the arm shapes. I have to move some stuff around. And well, I wanted to make another prototype at this day, but actually I was too tired to actually fix the locking pins. Because by making the arm shape differently, I broke the locking pins on this stage. So if you push that down, if you try to squeeze it, it will not match up. The teeth wouldn't just not match up. And this will no way grip at all. Well, next day was an easy fix, wasn't it? Well, there's a wall missing. Easy fix. So just added a piece of wall, and basically it was ready for prototyping again. So we had number five. So I had a working hinge, somewhat working tool head. Now we have an arm shape that's in the right shape and some somewhat working locks. So I was saying somewhat working tool head. The way I tested the tool head was it applies for still the grips. That's the function of it. So I took a plastic bag. I put the original hammer stud in, locked it, and tried to pull it out. I did repeat that with my printed model. And it was much inferior than the metal model. So I was thinking, yeah, how can I fix that? It's probably just making the arms bigger so there's more material. And well, that included a few changes to the code, basically just a few settings. But it broke something else. So I didn't notice it this time. Just printed it out with a big arm shape. So I've got number six. I've got a working hinge. I've got a somewhat working tool head, a good arm shape. And the locking pins weren't working with this. Noticed this problem before. So yeah, I should have looked actually at the code that this produced. I needed to retake these along. So there wasn't a big problem, just a bit of time loss. And there was number seven. So I had a working hinge, I had a working tool head, or a somewhat working tool head. I had a working arm shape. And well, OK, it locks actually. It worked. It locked. But I did the plastic-backed thing again and tried to actually test it with a hemostat. It was much better than before. But I managed to get the same group that I got on one tooth of the actual metal tool compared to three, pulling it all the way down to three locking pins on the plastic tool. So it was still inferior. So I was thinking about my pipe library that I made. It accepted squares only at this point. And if I try to make it a rectangle, it will show something like this. This doesn't look really right, does it? So can I have this rectangle pipe thing that I need? Well, the answer was yes. Just required a bit of a little hack. And this was not the complete hack. I had to do a small hack on my crystal scat. It was about two lines changes to change. And just needed to add this. And I had my working pipe library for rectangle pipes. So great, I thought. Print it out and see what happens. So I had a working hinge. I now had a good working tool hat. I had a good arm shape. And I had, well, not a good locking because it broke. It broke under its own force. So the least thing you want to have when a surgeon operates a patient is that stuff gets in the patient. So by breaking off a part of the tool, this might actually get into the patient without being detected. So that would be a total disaster. Well, what do I do? Make it bigger. I don't know. I might have missed slightly, but actually just made it a bit bigger. And just needed a small hack to do that, which I did. And that was number nine. So I had a working hinge, a working tool hat, a good-looking mechanism. And I had a working or somewhat working grip. I will tell you about one thing now. And I will say I have a somewhat working gripping mechanism. So I sent this to my employer and said, have a look at this. This works quite well for me. So I actually printed it, and it has it, and it worked for him, too. And I was talking about him, about the gripping mechanism. What do you think about it? He said, oh, it's great. I was like, hmm, but this doesn't look that perfectly when I print it out. I was like, yeah, but it's fine. So I was like, if it ain't broken, why fix it? So I left this subtle four bits. And this was, for a while, this was my last prototype. So I had some other tools to work on. I had some other projects to work on. And after months, three days ago, my employer came back to me and says, hey, I talked to the surgeons. So I went to a hospital. That was stage three of the testing versus I was talking about earlier. And he said, yeah, there's some of the things that are still problematic on these. Locking mechanism is one thing. I was like, yeah, I told you so. But OK, it doesn't help. So I was thinking about it in the right. And on the next day, which was two days ago, I was looking at it and saying, what do we actually need three prints on every side? If we just have one pin, we could rotate the pins in the correct shape. So they would lock according to the position of the actual opposite part. So the goal was to have both walls that connect to each other to be in the same angle. Well, in the same angle, so both the walls have the most contact area. So there was no air gap in between, which was the case. There was just one problem that made it bigger. So I was like, yeah, tried out. I printed it out. And I wrote a working hinge. On this, this is number 10, by the way. Out of working hinge, I had a working tool head out of good looking R-shape. And I had a, for me, perfectly working locking mechanism. So I have to say, I haven't reached my employer yet. And probably I don't have access to a 3D printer. So this, at the moment, is my final prototype, number 10. And well, for me, it works. It needs to be tested by my employer by someone independent yet. Later on, you can, after the talk, you can actually have a look at this. It's here on the table with the original. And well, I've got no overview about all the prototypes I need to make. So you can see on the top one, I was focusing on making the shape. Then I wanted to make it bigger, making it stronger, and fixing the last thing, like, to work. So I was going through the other tools that I need to make. So to make now the work. For that, I love my name. I love my crystal scat thing, because I could just inherit the things that I need to make. And copying most of the stuff. And like this, I just override the tool hat and have a different tool with the same stuff in it, which was actually true about the needle driver. So at this point, I wanted to talk about what this thing actually is. So maybe you can zoom in a bit on the stream. So this thing is used to make searches. And it holds a needle, a surgical needle. And it needs to be held in different grips. But that's not the only function, which I didn't know at first. So I needed a few more prototypes. So you wrap a yarn around this thing to make knots. So the most important part is that, actually, the tool hat is flush with the body. And my previous prototypes for the hinge, I just put a bolt through and then bother about this issue. So for this model, I had to redesign the hinge to make the bolt flush with the surface. So you could actually put a thread around and it will just go on to the shaft. The other tool hat I made is called Total Clamp. And I only needed one prototype. So how did I do that? I told you earlier that this project was done before. And they published the paper. And they had the photo of all their tools they had. And they said in the text that they had problems with making the Total Clamp. I haven't explained what it actually does. It's fairly simple. It basically clamps two pieces of Total together or cloth and locks it in place so it doesn't fall to anywhere. So I actually, inspired by the design they did, I just made a model. Didn't make a lot of changes in my model from the MSDAT. I printed it out. And I got a working hinge, working tool hat, working body, and the somewhat working lock that I haven't fixed on this one to the last one that I have on the MSDAT. Yet, but this will be done in a bit. So actually, I will need one more prototype. But actually, this one came out perfect for me. Came out perfect for the doctor. And he gave it to other doctors as well. So it just passed all three test stages on one try. Well, excellent. So this is unrelated to this particular project, but it's also really, really cool stuff. This is a 3D printed status scope. It uses zero parts of the original. And I have to say, we made it like a copy of the so-called gold standard of the industry, which costs about 150 euros. We printed it out. We tested a lot of different status scope parts. I will come to that in a bit. We attached some tubing to that. I've molded the air plugs. I will show you that in a bit as well. And it works better as the original gold standard in the industry. And guess what this costs? It's under 1 euro of production costs. So the 1 euro parts, which looks a bit like a toy, works better than the 1 out of a 50 euro part. And we couldn't get the reason for this, but I'll show you how we tested it actually. So we filled a water balloon with water. And we attached some headphones to it. And we attached a tube to the status scope part. This is the original Lidman. And inside this tube, there was a microphone. This was our setup to record the sound that was going into the status scope. We compared it to our printed ones. And the latest version of the printed one is actually performing better than this metal one. So this is how I made the earplugs for the status scope. There was actually two parts, a silicone mold. I put in silicone form by pushing it down a syringe. Let it harden, get the mold off, and get working earplugs. As simple as that. So that was the status scope project. There's a status scope here. You can have a look at it later. So this one I don't have on the table. This is my current work in progress. This machine is, I picked this up from Klim Yanov. This prototype, I've already developed this prototype to be a semi-automatic machine, which is used to produce bandages. So you put in the yarns, all the coms, and a yarn to a shuttle and then produce bandages for also for companies that don't have access to medical supplies. So they can make themselves. All right, cat tags. This is basically the end of my presentation. These are all my links for my GitHub-related stuff. All the work that I've done is on GitHub. You might want to check out. If you browse through the code, you want to check out the crystal scut as well. It's on crystalscut.org. Yeah, the code for the other projects are on there. And if you want to contact me, my email is webmaster at joaz.d. And you can contact me if you have any questions to the project, if you have any work for me, engineering work. Also, I have a web shop that I've been working on. Also, I have a web shop that is web-websource.com. I sell 3D print. I do that for a few years right now. So if you need any of these, go on this website and help me out there. OK, we have most of the time for questions, I think. OK. No? Yeah, OK, it's working. In the beginning, you said that you had to obviously change from metal to plastics. Did you have any feedback on the doctor you're working with on disinfecting and disinfection? Is that a problem for plastic? Or does it just work the same way than with metal? That might be a bigger problem than on the metal part. But not this disinfected part. If you try to disinfect this part, it's probably much better than having no parts at all. So the point is to have access to these things. And by not having access to these things, the patient will probably just die anyway. So you can possibly try to disinfect it with disinfectants. Or ABS can withstand up to 105 degrees. So you could steam it. I'm not sure how effective this is for 3D printed, but as this surfaces are not perfectly smooth. So I'm not an expert on this. Maybe a microbiologist can tell me about this. But as I said, it's probably much more value right now than having no tool at all. Are there other questions? Do you use any special material? Sorry, just the regular plastic that you can use for any 3D printer? For my testing and just use normal ABS. And my doctor or employer used normal PLA for printing these out. So it was a requirement of the job to have normal materials to print these, as they are most available in countries that don't have medical equipment. So what the temperature or the melting temperature of this material? Do you know that? What do you mean by melting temperature? Well, in the 3D printer it's melted, right? The extrusion temperature, you mean. Oh, yeah. OK. Yeah, ABS melts up like 238 degrees in my printer. PLA would melt about 190 to 210, depending on type and printer. Have you thought about rounded corners and tips for the tool head? They look a bit edgy. OK. Possibly. They get a bit rounded through the fact that the 3D printer cannot make perfect edges. So for my feedback so far, it looked OK. So I might get feedback that it says it should be more round. OK, then I might get more round. Are there other questions? So I see no more questions. When there are questions, I think we'll be around. And now you can see the instruments on the desk at the front. Please come forward and see the tools with the printer ones.