Current undergrads working on data science research with faculty

Dr. Jesse Wilson

Dr. Jennifer Mueller

Dr. Nikhil Krishnaswamy

Project Abstracts:

Jesse Wilson and his lab produce molecular-level biological images. These images contain noise – a product of the method and equipment. A clean image can never be produced by the method. The objective of this research is to de-noise these images. Typical deep learning de-noising methods have access to clean images, but we do not. Methods that do not require clean targets do exist (Noise2Noise, Noise2Void, DnCnn, etc.). These architectures are explored and expanded upon. Another issue is we don’t know how good the model is actually doing. With no clean images, there is nothing to compare the models output to. Methods for understanding the quality of the output image with no ground truth are investigated.

Dr. Jennifer Mueller’s project has the student working on deep learning based methods to improve image resolution in electrical impedance tomography images.  A 2-D training set of pediatric lung image data has already been created using archival CT scans.  The student will work on two objectives in parallel.  One is to create a 3-D training set from 3D segmentations that have already been created using the software ITK-Snap. The other objective is to use the 2-D training set to develop a novel end-to-end reconstruction method for 2-D images from EIT data collected on 16 electrodes in a circumferential geometry.  The method will be validated on a reserved portion of the training set, and potentially applied to human subject data collected on infants at Children’s Hospital Colorado. 

Within Dialogue Modeling research in AI and NLP, considerable attention has been spent on “dialogue state tracking (DST)”, which is the ability to update the representations of the speaker’s (user’s) needs at each turn in the dialogue, by taking into account the past dialogue moves and history. This is used to help arrive at the most useful dialogue policy (in an ML model)  to predict next state/actions that should be performed in the dialogue. Less studied but just as important to dialogue modeling, however, is “common ground tracking (CGT)”, which identifies the shared belief space held by all of the participants in a task-oriented dialogue. Carlos will be annotating multimodal classroom interactions, and examining important features to train into CGT models to both identify the current set of beliefs, as well as predict upcoming common ground, e.g., where the dialogue will go (the “questions under discussion,” or QUDs). The goal is to provide a more informative snapshot of the dialogue situation, after each action in the task, to develop a policy incorporating shared beliefs in addition to past dialogue history. Annotations performed over video will be operationalized as features to use in deep neural network models for recognizing and predicting common ground.

Student Bios

Hello! My name is Sam and I am a senior at CSU studying Data Science with a Computer Science concentration. I am fascinated by everything to do with Data Science, specifically deep learning and computer vision. CSU has been an amazing place to grow and learn, especially through undergraduate research. Last summer, I did statistics research with Dr. Kayleigh Keller, where I investigate spatial confounding through simulation. In October, I began research with Dr. Jesse Wilson. I investigate image de-noising methods using deep learning. The CSU research body has been an amazing gift and has shaped my journey through academics.

Hi! My name is Carlos and I am a student with a specialized focus in Statistics and a passion for problem-solving. Currently in my third year at Colorado State University pursuing a B.S. in Data Science.

In high school, I worked with district officials to create an internship program to provide opportunities for students in my small town.

When I am not working on school, I enjoy volunteering for HOBY and meditation. 

Carlos is a fourth-year student from Pueblo, CO, pursuing a degree in Data Science with a concentration in Statistics. He is excited to join the SIGNAL lab and work on the iSAT project, gaining valuable hands-on experience in the field. He is looking forward to challenging projects, an intellectual environment, and helpful peers, which will all contribute to his success as a Data Science student.