students doing research in lab

CEMS has created a research experience for undergraduates (REU) to promote opportunities for students to engage in research projects (CEMS-REU). The program provides stipends intended to incentivize undergraduate engagement in research and funds up to 100 opportunities for one semester REU’s during each academic year. The most important quality you will bring to this position is your eagerness to learn. Our faculty do research in a variety of areas. Reach out to them to learn more about their research and any REU opportunities they might have.

What CEMS-REU's provide

Each student selected will be provided up to $500/semester, for a maximum of two semesters. This includes the fall, spring, or summer semesters. Stipends will typically be provided as hourly wages, for work performed on a project defined by you and your research mentor.

Who is eligible

Any CEMS undergraduate student is eligible for this research program, including students already earning credit for Undergraduate Research or an Honors Thesis. Note that you must be working with a CEMS faculty mentor as part of your research (see "How to apply" below).

How to apply

Students are required to have identified a CEMS faculty member who has agreed to be their research mentor. The faculty mentor's home department can be the same or different from the department of the student's major. Applications will be reviewed by the home department of the faculty mentor. To apply, send an email to the appropriate contact shown below, for the department of your proposed faculty mentor, and please cc your faculty mentor. The subject line should be "CEMS-REU APPLICATION". Include in the e-mail your name, a brief description of the project you are proposing to work on, and the faculty member you are proposing to work with. Individual departments may later request additional material for evaluation.

Academic Year 2018-2019 REU Participants

Dept Last Name First Name Faculty Project
EE Tran Khang Eva Cosoroaba Battery Management using a Buck-Boost Converter for Microgrid Applications
EE Sychykov Valeriy Jim Modular Analog Computer Design with Control System Applications
EE Berne Sophie Eva Cosoroaba Analysis of and Strategies for Vermont's Energy Transition
EE Peterson Karl Tian Xia Automobile collision detection using radar
BME Warren Rose Ryan McGinnis Developing a Musculoskeletal Kinetics- and Kinematics-Based Algorithm for Fall Prediction in Patients with Multiple Sclerosis
BME McGill Madeleine Drs. Menon and Bates Ventalect, a communication tool for patients on mechanical ventilation
BME Carey Allison Drs. Menon and Bates Ventalect, a communication tool for patients on mechanical ventilation
BME Baraky Grace Drs. Menon and Bates Ventalect, a communication tool for patients on mechanical ventilation
BME Fisher Isabelle Drs. Menon and Bates Ventalect, a communication tool for patients on mechanical ventilation
M&S Myhaver Vanessa Dorais Formalizing theorems of linear algebra: The project consists of building computer verifiable proofs of theorems in linear algebra using the Lean Theorem Prover, a new proof assistant developed by Microsoft Research and Jeremy Avigad (Carnegie Mellon) https://leanprover.github.io/. This is intended as part of a long-term REU program for mathematics and computer science students. The goal is to have students contribute to the growing library of computer-verified proofs. The three students team will contribute by formalizing theorems of linear algebra and building on earlier work done by Vanessa over the Summer. This work is especially important since linear algebra was identified as an area of need at a recent meeting in Orsay.
M&S Stupinski Anne Marie Dorais Formalizing theorems of linear algebra: The project consists of building computer verifiable proofs of theorems in linear algebra using the Lean Theorem Prover, a new proof assistant developed by Microsoft Research and Jeremy Avigad (Carnegie Mellon) https://leanprover.github.io/. This is intended as part of a long-term REU program for mathematics and computer science students. The goal is to have students contribute to the growing library of computer-verified proofs. The three students team will contribute by formalizing theorems of linear algebra and building on earlier work done by Vanessa over the Summer. This work is especially important since linear algebra was identified as an area of need at a recent meeting in Orsay.
M&S Thomas Helene Dorais Formalizing theorems of linear algebra: The project consists of building computer verifiable proofs of theorems in linear algebra using the Lean Theorem Prover, a new proof assistant developed by Microsoft Research and Jeremy Avigad (Carnegie Mellon) https://leanprover.github.io/. This is intended as part of a long-term REU program for mathematics and computer science students. The goal is to have students contribute to the growing library of computer-verified proofs. The three students team will contribute by formalizing theorems of linear algebra and building on earlier work done by Vanessa over the Summer. This work is especially important since linear algebra was identified as an area of need at a recent meeting in Orsay.
M&S Kotzen Ben Taras Lakoba A stability analysis of a numerical scheme that simulates so-called coupled-mode equations by a Method of Characteristics: Recent numerical experiments by a colleague of mine revealed an unstable behavior of this scheme when noise is being added during the calculations to simulate the physically occurring noisy field. We are trying to come up with an analysis that would explain these observations, and at the end hope to propose a modification of the numerical scheme that will not have such a numerical instability.
M&S Hirsch Laura Maggie Eppstein (CS),  Donna Rizzo (CEE),  collaboration with Dr. Robert Gramling, (Medicine) Assessing the quality of medical caregiver communication is a time-sensitive endeavor and may infringe on patient privacy.  To address these issues, the Vermont Conversation Laboratory is conducting research into automated assessment of quality of conversations using machine learning. Over last summer, Laura helped us to quality-control on transcripts of audio recordings of real palliative care consultations and conducted some preliminary data analysis. During this academic year, Laura will be conducting exploratory research using natural language processing and machine learning from these transcripts, to help us try to identify signatures of effective conversations. For example, among other things she is exploring whether patterns in duration of clinician/patient speaker turns are associated with the degree to which the patient later reported feeling heard and understood.  This work involves significant creating thinking, programming in Python, and principled application of various statistical and machine learning approaches.
CS/CSYS Williams Blake Laurent Hebert-Dufresne This project will evaluate the resilience of real-world influenza A (H3N2) hemagglutinin genotype networks to host immunity and compare resilence with a space of simulated networks.  Real-world networks will be generated from NCBI database protein sequences.  Features of the real-world networks will be analyzed, and features of resilient networks will be identified through multi-strain compartmental modeling. Utilizing networks will offer a complementary view to the phylogenetic trees commonly used in analyzing the evolutionary dynamics of influenza subtypes.
CS Stuntz Lindsey Christian Skalka The student will explore new technologies for improving security in modern networks, including networks that underly the Internet. In particular, the student will explore techniques for automatically guaranteeing that network routing configurations adhere to stated policy, and that any changes to configurations are properly authorized. A secondary goal of this project is to synergize UVM research with the student's internship work at Amazon Web Services.
CS Duval Ben Robert Erickson The responsibilities for an undergraduate student will be to conduct fact-finding meetings with the client to document their needs.  The current side is working but needs several additions that will require more coding. This is a great opportunity for one of our students to apply existing knowledge as well as branch out into areas of Development they have not explored yet. Languages needed: HTML, CSS, PHP, MySQL
CS Do Edmund Byung S. Lee This project aims to build a software program that converts 12-lead ECG data to multi-dimensional VCG data and visualizes the VCG data.  Public data available from PhysioNet will be used as the ECG data, and Python libaries will be used for VCG visualization over a selected set of dimensions.
CEE Hancock Kevin Priyanthan Wijesinghe Kevin will study new bio-inspired structural forms that can be utilized at the organism level as freestanding structural forms. He will also research the demanding material properties and analysis techniques and tools required to make these structures a reality.
CEE Gillies Merrick Mandar Dewoolkar About thirty percent of world's housing is made of adobe, air-cured large bricks made out of soil often mixed with some straw. These structures are heavy and unreinforced making them vulnerable in seismic regions. This project will assess strengths of adobe and grouts individually and as a unit.
CEE Thorne Heidi Mandar Dewoolkar We are conducting research developing and applying machine learning algorithms to help us understand what factors contribute to the effectiveness of clinical conversations with seriously ill patients.  Using audio recordings of 363 palliative care conversations, we are currently focusing on two sub-projects, as follows. (1) Connectional Silences: Human connection is an essential component of palliative care conversations. Some moments of silence amid such conversations indicate such human connection. Our preliminary work demonstrates capacity to use machine-learning to identify pauses in conversation and human coding to reliably classify pauses as “connectional” or not. (2) Temporal Attribution & Affect:  Palliative care conversations involve reflection on the past, understanding the present and contemplation of the future in a highly emotional context of medical decision-making. These two elements—time and affect—are conceptually “findable” using text processing methods and are likely to mark different clinical “types” of conversations that are likely to lead to different outcomes (eg. treatment choices, quality of life).  The three undergraduate research assistants will assist in both of these projects through various tasks including data pre-processing of conversation transcripts, audio screening of pause clips identified by our machine learning algorithms, helping to compile a corpora of temporal reference words and phrases, and exploratory data analysis.
CEE Romero Eric Scott Hamshaw Research will involve GIS and Matlab analysis of sediment data several watersheds throughout Vermont. These watersheds include but are not limited to: Lewis Creek, Shepard Brook and Mill Brook watersheds. The goal of this research is to analyse turbidity-TSS relationships in the various watersheds, characterize the land cover, geomorphology, geology, road network and hydrology upstream of sensor monitoring locations and to process and analyze time series of river discharge and turbidity/TSS (Total Suspended Solids) collected at other monitoring sites throughout Vermont.
CEE Jobin-Davis Eliza Dewoolkar, Rizzo, Garcia Educational modules using augmented reality sandbox

 

Interested? There are many opportunities for undergraduate research in CEMS!