Note: I submitted this in April 1998, and waited until it had gone out to everyone in the department before revealing that it was a joke. It has nothing to do with my real Ph.D. thesis.

Robotics Thesis Proposal

An Application of Limbless/Legless Locomotion and Telepresent Imaging to Gastrointestinal Endoscopy

Martin C. Martin

Friday, April 10, 1998
2:00 p.m.
FRC 100

Committee:Hans Moravec (chair)
Illah Nourbaksh
Simon Penny
Peter Cariani, neurobiologist from the Eaton
    Peabody Laboratory of Auditory Physiology,
    Massachusetts Eye & Ear Infirmary.

Abstract:

Continuing the recent Ph.D. work of both Kevin Dowling (Limbless Locomotion: Learning to Crawl with a Snake Robot) and John Murphy (forthcomming, on the panospheric camera), this thesis explores the union of these ideas, applied in the medical realm. Specifically, I propose to use a panospheric camera mounted on an autonomous "snake robot" to image the human large and small intestine. Both applications are novel. Current endoscopes, which are tethered and must be controlled by the physician, can only penetrate about 10% of the gastrointenstinal tract. And current endoscopic imaging systems, with their poor resolution and field of view, force the physician to spend a long time getting an adequite image, which he or she becomes reluctant to adjust for fear of losing the view.

In contrast, the proposed system will allow the snake to propel itself through a large part of the patient's intestinal tract, perhaps all the way to the mouth, completely autonomously and without the need for human intervention. Issues to be addressed include propulsion in a non-rigid medium, quantifying and detecting "interesting areas" inside the intestines, and locomoting without damaging the patient. It is hoped that all of these can be achieved before in situ trials begin. As well, the 360 degree field of view panospheric optics, when mated with the appropriate telepresent display, will allow the surgeon a much more immediate and intuitive experience, creating the illusion that the surgeon's head is actually inside the patient's rectum. Such an experience, while disconcerting at first, quickly becomes second nature for those with similar previous experience, and will certainly aid the accurate determination and correction of any gastrointestinal misfunction.

Martin C. Martin:  martin at metahuman.org