Medical Visualization

Medical imaging, also known as medical visualization, is a field of strong interest to me. Images of the human body are created by means of radiography, endoscopy, thermography, photography, microscopy, and even through illustration. The images may be created for clinical purposes in order to diagnose or examine a particular disease, or for the study of the human anatomy and physiology. I’ll discuss a few imaging techniques that are of particular interest to me, as well as an approach to medical visualization that does not result in any specific imaging.

One process I find interesting is magnetic resonance imaging (MRI), a technique that uses a magnetic field and pulses of radio wave energy to image organs and structures inside the body. An MRI scan can be done for the human head, including the brain, the chest, blood vessels, the abdomen and pelvis, bones and joints, and the spine. MRI creates many two dimensional pictures of “slices” of the body. It is the preferred imaging technique for detecting tumors within a person, and therefore is of personal interest to me due to my family history.

Another medical imaging technique is ultrasonography, which uses sound waves to produce two to three dimensional images. This is most often the technique to image fetuses in pregnant women, however, ultrasounds can produce images of many organs within the human body. I had the opportunity for a Visualization project to interview a cardiologist and tour the visualization laboratories at the Central Texas Heart Center in College Station. He showed us several approaches used to image the heart, including 3-D echocardiograms (ultrasound of the heart) and PET scanning (positron emission tomography).

I would like to conclude with the argument that medical visualization is not limited to the creation of images. Is not the dissection of the body or its parts a means to visualize the human system? A very unique approach to medical visualization is the Body Worlds traveling exhibition. This exhibition features human bodies and parts preserved through plastination, a technique which replaces water and fat with plastics. The display is a fascinating one, revealing inner anatomy of both healthy and ill bodies. It is incredibly informative through its visual, and at times artistic, depiction of the human body. It features specific organs, muscle systems, the central nervous system, and helps people to visualize how these systems and parts practically function.

The reason I chose the topic of medical imaging is because after my graduation in December 2012, I intend to pursue an MS in Prosthetics and Orthotics. The field is a fascinating blend of biomechanics, art, psychology, and medicine. I will likely use two and three dimensional visualization techniques in order to help design the prosthetics and orthotics. There are so many uses for visualization techniques besides artistic, architectural, and entertainment purposes, and medical imaging is one example. 

References: Dorland's Medical Dictionary for Health Consumers, 2007; McGraw-Hill Concise Dictionary of Modern Medicine, 2002; WebMD "Medical Imaging and MRI Scanners"

The Language of Amélie

The cinematography of the French film Amélie is quirky and whimsical. In order to achieve this mood, the director and cinematographer establish interesting compositions, close shots, and playful camera movement throughout the entire movie. I have chosen a short clip that exemplifies the overall style:

http://www.youtube.com/watch?v=mmllotLUU38

The first shot is a fairly quick camera tilt that scans the character Amélie from head to hand. The tilt ends abruptly, but not harshly, with a view of her hand dipping into a sack of grain. The close-up view is constant for the entire shot.

A cut transitions to the next scene, which simply is an extreme close-up of Amélie’s face, and a spoon that she holds up in front of her. A cut to the next shot is simply an extreme close-up.

The next scene is less straightforward than the previous ones, as far as camera movement is concerned. A bird’s eye view reveals a river dam waterfall, and the camera begins a fairly slow clockwise rotating/tracking motion. Once Amélie comes into view, she becomes the origin of the rotation, which speeds up slightly. Then, the camera pivots down until it is level with the water level, and she is level with the horizon line. 

The filming is rather effective, though I think that the bird’s eye view feels rather out of place and unnecessary. Perhaps a close up of her stones hitting the water would have sufficed. 

A cut to the next scene reveals Amélie in her home, and the camera follows her by tracking from the left to right side. Once the tracking motion stops, the camera zooms in slowly on Amélie, whose attention has been captured by something not in the frame. This psychic line of her gaze helps the viewer to understand the following shot, which is an over the shoulder view of the neighbor she is observed from her window. A cut to the next shot depicts Amélie closing her window from outside of her home, and the camera slowly zooms in again. A cut to the next shot shows her walk again towards her window, this time with a spying telescope in hand. Again, the director uses the concept of a psychic line to lead into the next shots.

Through the use of a vignette, which likely was added in post-production editing, viewers feel as if they are looking through Amélie’s spying glass.  The camera even toggles as if it were being held by hand, and pans as if she is directing her attention elsewhere. Two more cuts reveal shots of her spying through her window, and the other from her perspective. These cuts in the window peeing scene are based on Amélie’s movements and are always forward.

Throughout this short clip and the rest of the film, straight cuts and extreme close ups are implemented. Much of the action is centered on Amélie, rather than many extraneous outside characters, though through parallel editing viewers are introduced to a counterpart later in the film. Time is compressed significantly in the entire film, and the scenes would be unrelated visually would they not be describing and involving the main character. 

Modeling a Human Hand

The human hand is challenging to represent, whether the medium be a 3D modeling program, traditional pencil and paint, or clay. I personally modeled a human hand for my VIST 305 course, and will discuss the steps I took in approaching this geometry.

It is important in this process to assign which part of the hand goes in which coordinate direction. My z-axis would represent the depth of the hand, my x-axis the width, and my y-axis the height. I want my fingers to point in the positive y-direction, and my palm to face the positive z-direction.

The starting object is a cubic primitive, which can be extruded three times–this will result in four vertical devisions, which will ultimately become our four fingers. The first step in separating the fingers is to bevel the edges at the top. Then, subdivide the block horizontally, making sure to situate a division at the vertices of the beveled triangles, and twice more. The number of subdivisions is up to the modeler, however, I am choosing less for now to be as efficient as possible. There can always be more subdivisions as we progress through the modeling process. 

The rest of the process of modeling a hand consists of geometric extrusions, subdivisions, and the translation of vertices. To accurately render the curvature of the palm, I would consider its topology–moving vertices and subdividing as needed. I would reference my own hand and images of the hand.

The fingers are detailed and involve more subdivisions. They must first be extruded, then subdivided vertically at least once to attain roundness, then horizontally at least six times. The horizontal subdivisions are important for modeling the joints. Many modelers choose to model a single finger, and then duplicate it three times, adjusting size and shape accordingly. I would personally choose this method, but would be careful to situate the joints in their proper height hierarchy, and to take care to model natural imperfections of the human hand. We have to make sure that the modeled hand doesn’t appear to be a robot glove. Another consideration for the fingers is to delete any unnecessary faces–specifically, the ones that lie where the fingers will weld with the hand.

The thumb is fairly difficult to model. First, it requires that a large face on the side of the palm be rotated to the proper angle. The face should rotate towards the positive z-axis or away from the center of the palm. Then, the face is extruded and subsequently subdivided to model the details. Once extruded, the thumb rotates slightly around its single joint. 

The description of the process of modeling the human hand using a 3D software program could occupy pages and pages. Essentially, it involves a cubic primitive, vertical and horizontal subdivisions, five extrusions, and the translation and rotation of vertices. 

References: Modeling of Joan of Arc by Michel Roger