Liver Retractor Demo

This is a video demonstration that I recorded for the retractor described in the previous two posts. You may notice that the "wrist" portion of the retractor is significantly shorter by the end of the video. This is because I broke the retractor while recording this demo and had to shorten it to get it working correctly. The shorter version functioned much better and allowed for a tighter bend in the wrist section.

The overall retractor worked fairly well, but unfortunately was not the ideal design for the purpose it was supposed to serve. I fell prey to the common problem of overestimating the scale at which the surgery took place. I had taken measurements and looked at videos and it seemed that the abdomen was a great cavernous space with room for a 5 inch long robot. During testing I realized that realistically, the "ledge" on the liver that I would be wrapping this retractor around was very narrow. A better approach would have been to think of the retractor as pushing the liver out of the way rather than lifting it up. A better design would have been more like a tent frame that supported the liver while staying out of the way of the surgeon operating on the stomach.

Liver Retractor Control

The first complaint that I got from the surgeon when I let him test the prototype retractor shown in the previous post was that the interface didn't make any sense. He moved the joysticks while watching the retractor, expecting it to imitate the motions of his fingers, while I had designed it so that the joysticks themselves were representations of the retractor fingers and the movements of the retractor would mimic the position of the joysticks.

The interface just wasn't set up for the kind of control that the surgeon wanted. He was correct though, the surgeon wouldn't be looking at the joysticks when he was working, he would be watching the retractor inside the patient. The interface would have to translate his hand motions into motions of the retractor. So I set out to make a new interface that would hide the joysticks and just focus on getting the surgeon's hand into a natural position.

This diagram shows a part of the new interface, the joysticks that control the finger portion of the retractor. It is shown upside down in an attempt to make the joysticks recognizable for what they are. The surgeons hand would wrap around the top (bottom in this diagram) of the shell containing the joysticks and his fingers would rest in a series of loops protruding from the shell. Each loop stem would pivot around a pin in a slot, allowing side to side and forward and backward motion. These motions would be transferred to tiny joysticks inside the shell which were wired to the motors that controlled the retractor. I tried to get at least the basic idea of this across in this section view which was drawn based on a CAD model I used to fabricate the shell.

However, I didn't show this diagram to the surgeon as that would have defeated the entire purpose of my redesign: to hide the joysticks so that he didn't have to think about them. When I let him control the retractor, it was much more intuitive for him. He did complain that the range of motion for each finger was too limited, but at least it was a step in the right direction!

Liver Retractor

The project was a retractor to support the liver during minimally invasive robotic surgeries. The goal of the diagram was first to convey the concept and basic function of the first retractor prototype to the surgeon sponsoring the project and second to show enough action and complexity that he would feel that his money had been well spent and we weren't wasting his time.

I had it in my head that it should be an entirely visual poster with minimal text. Most engineering posters look like walls of text, with a few graphs and small pictures squeezed into the times new roman masonry. I think that people who look at a poster will pretend to read the words while actually just darting their eyes from one picture to another. If the pictures don't tell the story, they probably won't take the time to read the text.

So I decided to try to give a flow to the diagram, by wrapping a semi-exploded rendering of the device around some more explanatory pictures of the device in surgery to give some context to the poster. The focal point at the center would explain to the viewer what the research was about: "robotic surgery". More specifically: "visualization of the surgical field" the viewer should realize as he or she sees the surgeon's POV off to the right of his head. This is all encircled by the device itself, in an attempt to guide the viewer's eye around the poster.

The first part of the retractor to catch the viewer's eye would probably be the end effector, with its three bright blue fingers. The control of these fingers is explained as the eye moves counter clockwise to find the motor array in the box at the bottom left. The method by which these motors are controlled is explained as the eye continues to the bottom right to the joystick interface. Just to clear up any ambiguity, a top view of the joystick interface is overlaid on the retractor end effector to show the relationship.

I wanted the images in the poster to all have a uniform and simple but tidy look. To do this I drew on source images from photos of surgeons in the OR, CAD drawings that I had made of parts of the system and an endoscopic view taken during a surgery. I took all of these source images and redrew them using an open source paint program (paint.net). This program allows the use of layers which made it easy to move the components of my drawing around.

The poster ended up working very well and the surgeon was impressed. I also submitted this for my engineering department's poster contest and won first place. I think the benefit of a visual poster like this in a poster competition is that it allows the presenter to do all of the talking without the viewer being distracted by trying to read text on the poster while listening. The presenter can then point to aspects of the design as he or she discusses them to lead the viewer through the research. This makes the poster more of a handy collection of visuals to be used in presenting an idea.

One thing that I wish I had done better, was to show the proposed device in surgery. The prototype retractor is barely visible in the surgeon's POV image at center right just under the liver. The image is dark and the retractor is barely recognizable. This requires a little imagination on the part of the viewer to figure out how the retractor is helping the surgeon.