After my wife had some medical tests done (everything was OK!) she was sent home with set of MRI images. I realized that I could use these 'slices' of my wife's skull to create a 3D model. After a little experimentation, including a sloppy foamcore model and a flipbook, I decided to paint each layer onto an acetate sheet and stack them in a box with LED backlighting. Here we see that original lamp with blue lights and the new one in green that I will describe in this instructable.
The basic idea is to paint each 'slice' of the MRI onto a pane of glass and stack them up in a frame. LEDs mounted in the top illuminate the layers creating a glowing 3D effect. My approach to design is to look around my workbench and scrap pile for usable stuff and make it work. Everything used in this project was either recycled or laying around my shop. My goal here isn't to teach you how to build a lamp but to show you how I built a lamp. Feel free to do with it what you will.
If you don't have an MRI of your skull laying around ask friends and family. Chances are someone you know has a set. I found a low resolution scan online by doing a GIS. Perhaps a medical student or friendly medical professional could help you out.
An MRI basically presents a series of images of 'slices' of the body part that was scanned. Bone shows up particularly well in white. What we want to do in this step is to create and print an image for each slice of the skull which we can place under each pane and trace with paint. Each pane will be stacked in order to create the illusion of three dimensions.
In this case the MRI came on old fashioned film. More and more doctors are saving and sharing this information in digital forms. Either way we want to import these images to a computer, open them in the photo editor of your choice and make them usable for our purposes. I used a piece of white paper for a backing and scanned my MRI films with a standard desktop scanner.
I used Photoshop, but you could easily use Gimp, Corel, Paint, or any other image editing software. Specifics vary from program to program and user to user, but here's what you want to do-
1. Crop each frame of the MRI into a separate image. Each image should be the same size, carefully aligned and numbered sequentially. I create a duplicate layer, use the rectangle tool to highlight a frame, crop the image and save it. Next, use the move tool to slide the copied layer to the next frame so that it lines up to the canvas edges and save it with the next sequential number. Repeat this for each frame of the MRI.
2. Adjust the brightness and contrast of each image to accentuate the bone structure and darken the rest of the image. It may seem like more work to process each image separately, but making adjustments and applying filters to the whole image first makes it harder to align each frame properly since a lot of detail is lost after the adjustment. Use batch processing to automate the tedious task of adjusting each image.
3. Convert each image to greyscale, invert the colors and simplify it. The idea is to make each layer clear and easy to trace later. In Photoshop I convert the image to greyscale under Image>Mode and then invert it under Image>Adjustments. Automate this if possible. Now you should have a set of black images representing the cross sections of the bone. Go ahead and clean up the image, eliminating any black marks that are not bone. Take your time. Sometimes dental fillings or surgical screws cause a 'flare' effect. These spiky artifacts are kind of cool looking but they can also be carefully erased layer by layer if you choose. The more time spent cleaning up and refining your image, the better your final product will be.
When all the images are cleaned up, I use the cutout filter to smooth and simplify the image. Set the layers to '2' in the filter's options and you will get a black image of each bone cross-section on a white background. Play around with the settings to get a good balance between simplicity and detail.
4. Give each image a thin black border and a number. This will make aligning each frame to a pane of glass later much easier. Open a new blank layer. Use the wand tool to select the entire canvas area. Use Edit>Stroke to add a thin black border to the image. Copy the border layer into each image and use the text tool to add the layer number to each image.
5. Resize each image and print. Determine the dimensions of glass you want to create and resize each image appropriately. Set your printer to print in black and white only and use a low quality/draft setting to conserve ink. Check the prints and if everything is acceptable, set them aside for later.