A sampling of senior theses and independent work

Brendan Kavanagh, Mechanical and Aerospace Engineering “Design of a Ski Tread for Variable Terrain Ascent” Brendan designed a tread that can be attached to the bottom of a ski to increase traction on the snow and allow the skier to ski uphill. He designed five different patterns and milled them onto pieces of low-density polyurethane, approximately two feet in length. He used Pro-Mechanica to analyze the interaction between the snow and the various treads, and later did field tests on snowy slopes in Utah. One of the treads made it possible for the skier to ascend slopes up to a 25-degree angle of inclination.

Sharam Fouladgar-Mercer, Computer Science “AEMER: Advanced Electronic Medical Record” AEMER is a patient record system developed for orthopedists and athletic trainers to keep more extensive, sophisticated information about patients’ ailments and injuries. It allows the physician to highlight localized pain centers on a diagram of the body with a simple point-and-click mechanism. The visualization of the ailment helps physicians track trends and analyze the injury. The user interface resembles that of the popular Windows to make the software feel more familiar and encourage use. Doctors consulted on the specifications and design of the AEMER program, and a number of doctors, medical students, professors, and trainers have tested it.

Alison Weingarden, Operations Research and Financial Engineering “Improving the Lots: Actuarial Testing, Survival Analysis and Megan’s Law.” Megan’s Law requires law enforcement agencies to notify the regional community when a dangerous sex offender is released from prison. About a dozen states use a formula called an actuarial test to determine whether a person is “dangerous.” The formula includes a variety of personality characteristics and details about their prior offenses. Alison obtained access to Arizona’s state data on convicted sex offenders and used survival analysis to find weaknesses with the actuarial tests and then suggest ways to alter the test to improve its predictive ability. Survival analysis assesses and compares the recidivism (arrests, parole violations, etc.) of “dangerous” and “nondangerous” convicts. She found that her changes to the test did improve its predictive capabilities. She suggests that survival analysis should be used to cross-validate actuarial tests.

Aaron Ellerbee, Mechanical and Aerospace Engineering “Design of a Knee Brace for ACL Support” Aaron’s thesis was inspired by Michael D’Emilio ’88, a former member of the Princeton varsity football team. D’Emilio tore the anterior cruciate ligament (ACL) in his knee during his senior year, rendering him unable to play the rest of his final season. He suggested that Aaron design a knee brace to prevent ACL tears or support injured ACLs without inhibiting the wearer’s ability to play their sport. Aaron found that by using composite materials, a brace could resist loads from directions that may injure the ACL while allowing other loads crucial to normal sports activity. The knee brace design is not yet ready for manufacture.

Sasha Rao, Chemical Engineering “In Silico De Novo Design of Human Beta-Defensin 2”Because protein mutations can cause disease, the ability to “design” proteins has major applications in drug design and disease treatment. Pathologists and engineers would like to reproduce or improve upon normally functioning, naturally occurring proteins. But first they must identify the protein’s analog—the sequence of amino acids that will produce the same specific three-dimensional structure that gives the protein its function. This task is called the inverse protein-folding problem and is nigh impossible to solve experimentally because the number of possible amino acid sequences is astronomical. Thus, Sasha attacked the problem computationally. He developed two optimization models—one that minimized energy, and one that maximized entropy—and used them together. The hypothesis that the models can identify analogs that will function better than the original protein can be verified experimentally.

Ryan Teising, Computer Science “Game of Hex”
“Game of Hex” is a board game created by two mathematicians (one of whom was John Nash), in the 1940s. Ryan aimed to create a computer player that could beat advanced players, who can “see” and predict maneuvers 15 to 20 moves into the future. Ryan developed an algorithm that blocks an opponent’s ability to find the best path across the board, thus thwarting some of the most advanced players. Play the game at www.princeton.edu/~rteising/Hex/hex.html.

[ contents ]   [ previous story ]  [ next story ]   [ top of page ]