Saint Louis University
Dept. of Computer Science

Computer Science 4961/4962
Capstone Project

Spring 2018


The first phase of the capstone experience is the selection of a project and the creation of teams. This page provides a summary of potential projects for this semester. Students must provide their personal preferences via Google form by 5:00pm Tuesday, September 4th, 2018. You may wish to take a look at past project descriptions that have been selected in recent years.


Menu of potential projects

Table of Contents:
  1. Digital History Displays
  2. Referral Generation and Tracking for Pediatric Clinics
  3. Low-Cost Spacecraft Attitude Detection
  4. Universal Injury/Illness Assessment Record
  5. Predicting Water Chemistry via Hydrology
  6. Shakespeare Lexicon
  7. Collaborative Telepathology
  8. Knowledge-Defined Networking
  9. Evaluation of Machine Translation Approaches for Closely-related Languages
  10. Reboot of Accentuate.us
  11. Spellchecker Development
  12. Self-contained Search Engine Solution for Irish Language Websites
  13. Easy-To-Use AI for CSCI 1030 - Game Design
  14. Computer Vision Website for Segmentation Using OpenCV.js
  15. Clinical Next-Generation Sequencing Analysis Pipeline for Precision Medicine in Cancer at SLU

Digital History Displays

Organization: SLU History
Client: Dr. Tom Finan
Supervisor: TBD
Description:

The computer science department in conjunction with the SLU History Department and the Ong Center for Digital Humanities will be conducting a joint CS/History capstone experience. Students will create a content management system that will allow historians to create digital content corresponding to historic places and artifacts at SLU and around Saint Louis. SLU has an extensive collection of historical data and artifacts, as well as having a historic campus and is located in one of the oldest cities in the region. SLU History students hope to expose this latent, unseen history that is all around us.

Students will work concurrently with history students on this project in order to identify what kinds and how best to present content to end-users. History students will spend the Fall 2018 semester preparing for integration and testing of our system in the Spring 2019 semester.


Referral Generation and Tracking for Pediatric Clinics

Organization: SLU Public Health
Client: Dr. Ellen Barnidge
Supervisor: TBD
Description:

Pediatric clinics are a key engagement point for social workers in under served communities. The clinic presents an opportunity to screen families for social challenges such as insecurity in food, housing, utilities, diapers, etc. and then to refer those families to local organizations such as food pantries or Parents As Teachers organizations. However, it is difficult to connect families with these local resources and then keep them engaged after they leave the clinic. Researchers working with SLU's pediatric clinic would like a student group to help better connect local providers to families through improvements to the screening and referral system. The researchers are open to suggestions, but have identified several limitations of the existing referral system. (1) There is no system in place for tracking referral success (the percentage of people who end up going to providers after referral). (2) The screeners themselves could benefit from computerized prompting of possible referrals, as these are often volunteer or part-time individuals. (3) Multiple methods of communication such as email, text, and paper documents, or additional communications such as reminders, could reduce lost or missed referrals.

This project proposes a system where a social worker can define local providers of resources, and patient screening criteria that trigger a referral to that provider. Patients are screened in the clinic and a referral information document is generated for them, and this information is also passed via text message and email, if applicable. Finally, the system would allow providers to visit a website in order to indicate referral success.


Low-Cost Spacecraft Attitude Detection

Organization: SLU Space Systems Research Laboratory
Client: Dr. Keith Bennett
Supervisor: TBD
Description:

Spacecraft attitude determination is the process of understanding how a spacecraft is pointed in space. SLU's Space Systems Research Laboratory is planning to launch a mini-spacecraft called a CubeSat in Fall 2019. Unlike many traditional space systems, SLU's CubeSat will be constrained in two areas: (1) the spacecraft is planned to tumble through space without active control, and (2) power availability for computation is not guaranteed. This project proposes for students to create or adapt techniques for attitude determination when spacecraft orientation can change rapidly and the interval between successive opportunities to run code is unknown and may range between seconds or weeks and months. This spacecraft will have two cameras positioned opposite each other at both ends of the spacecraft as primary sensors. Attitude determination techniques that might be applied by students include star tracking (finding known relationships among stars), coastline mapping, Earth edge detection, measuring solar panel power output, etc.


Universal Injury/Illness Assessment Record

Organization: SLU Health Sciences - Athletic Training
Client: Kitty Newsham
Supervisor: TBD
Description:

Students will create a medical note-taking system with an emphasis on computer security to be used by athletic training students during their clinical rotations. The purpose of this system is to replace an existing paper-based ad-hoc system of note taking to make notes more consistent and accessible for course assignments. This system will also enable the systematic collection and reporting of patient outcomes. Such a system will need to identify individual students so they can be tracked over time, but also anonymize records when they are used and reported. Students will need to identify and implement to their ability the best practices for storage and retrieval of medical records, such as database encryption, confidential transmission of data, creation of a system risk analysis, etc.


Predicting Water Chemistry via Hydrology

Organization: SLU Earth and Atmospheric Sciences
Client: Dr. Elizabeth Hasenmueller
Supervisor: TBD
Description:

This project would entail using machine learning to predict water quality responses in streams on short (i.e., flood response) and long (i.e., seasonal) timescales. We would like to understand what meteorological and hydrological variables are most important for predicting water chemistry response. A network of stream guages and stream sensors across the St. Louis region currently collect and have collected data on hydrology and water chemistry at five minute intervals for some time, so this would be a great opportunity to apply modern data analysis techniques to a real data set.


Shakespeare Lexicon

Organization: SLU Theater
Client: Nancy Bell
Supervisor: TBD
Description:

The Shakespeare Lexicon is an exhaustive documentation of any and all words found in Shakespeare's plays. For example, it provides appropriate context-dependent definitions for every different usage of a word. However, using this resource can be tedious due to it's size and exhaustiveness: some common words have hundreds of entries. This project aims to create a a searchable, indexed form of the Lexicon that is better suited to academic and classroom study at SLU.


Collaborative Telepathology

Organization: SLU Medicine
Client: Dr. Grant Kolar
Supervisor: Dr. Flavio Esposito
Description:
Telepathology is the practice of digitizing histological or macroscopic tissue images based on a glass slide for transmission along telecommunication pathways for diagnosis, consultation, or continuing medical education. In the majority of non-trivial pathology cases, to minimize the response time to the surgeon and the probability of incorrect assessments, pathologists ask for second opinions from nearby experts (if available) by physically carrying glass specimens. Especially to complement underserved geographical areas, expert pathologists should have remote access to difficult case assessments via telepathology. Today however, telepathology is practically unused for the applications that would need it the most: fast and reliable consultations as well as multi-students live teaching sessions. Moreover, pathology is nowadays mostly taught via offline methods or via one-to-one mentor-student specimen analysis. Remotely recreating the effect of a microscope locally handled would allow multiple pathologies across federated sites to collaborate on non-trivial diagnoses. Best-effort Internet connections are simply not enough to support such applications.

Students in this project will focus on the following aspects of Internet of Things applied to Telepathology systems:


Knowledge-Defined Networking

Organization: SLU CS
Supervisor: Dr. Flavio Esposito
Description:

The objective of this capstone is to apply notions of data science to computer networking. In particular, the student(s) will focus on building models and prototype systems on real networks or devices that use artificial intelligence, e.g., machine learning or statistical inference, to improve the state of the art of (mobile) edge computing.

To know more, skim this reference: https://arxiv.org/pdf/1606.06222.pdf


Evaluation of Machine Translation Approaches for Closely-related Languages

Organization: SLU CS
Supervisor: Dr. Kevin Scannell
Description:
This would involve training a number of MT engines, including phrase-based models like Moses and, say, character-based models using RNNs (TensorFlow) which could possibly work well for very closely-related languages. I have data for the Gaelic languages and for Irish language standardization. Other pairs are possible: Zulu/Xhosa?


Reboot of Accentuate.us

Organization: SLU CS
Supervisor: Dr. Kevin Scannell
Description:
In a capstone project from many years ago, a student developed a web-based platform (previously hosted at accentuate.us) along with a corresponding firefox add-on. Its purpose was to use machine learning to automatically restore diacritics when entering ASCII-only texts, to make it quicker and easier to type in more than 100 world languages without extra keystrokes or a special keyboard. That service is no longer active and there were scaling challenges since the statistical models needed to be stored in RAM and this limited the number of languages that could be supported. With this project, a team will rely on more modern approaches to scalability to revive and extend the original tools.


Spellchecker Development

Organization: SLU CS
Supervisor: Dr. Kevin Scannell
Description:
Dr. Scannell has raw data consisting of word lists crawled from the web for 2000+ languages. It would be nice to have a website which could allow a language community to crowdsource the editing (voting, etc.) of these wordlists to turn them into spell checkers. Part of this would be an interactive tool for developing so-called "affix files", without having to be trained on the technical details, and also a tool for exporting spellchecking addons for LibreOffice, Firefox, etc. automatically.


Self-contained Search Engine Solution for Irish Language Websites

Organization: SLU CS
Supervisor: Dr. Kevin Scannell
Description:
The idea would be to build this on top of existing open source search engines like Apache Lucene, or anything derived from Lucene (Elastic Search, Solr, ...). The issue is to allow indexing according to eight possible combinations of (standard/non-standard, mutations/stripped, stemmed/unstemmed), using software developed by Dr. Scannell, and to package the result in a way that's trivial for site maintainers to deploy.


Easy-To-Use AI for CSCI 1030 - Game Design

Organization: SLU CS
Supervisor: Dr. Jason Fritts
Description:

The Unity game engine tools used by CSCI 1030 offer built-in tools for AI that are rather complex, and presently too difficult for students in CSCI 1030 to effectively use. There are also additional AI tools available on the Unity asset store, but their effectiveness and complexity is uncertain. I'm interested in a student project to develop some versatile but easy to use AI that students in CSCI 1030 can add to their games in Unity. The tools should offer a few different AI types (e.g. enemy AI like guard, shooter, chaser, etc. and friendly AI, that can "speak" to player), while remaining easy to use and parameterize.


Computer Vision Website for Segmentation Using OpenCV.js

Organization: SLU CS
Supervisor: Dr. Jason Fritts
Description:

OpenCV is the predominant set of tools used for computer vision. It's long supported C++, Python, and Java, but recently a JavaScript version has been released, enabling the development of interactive computer vision tools in websites. My research studies image segmentation, which attempts to classify the pixels in an image to regions corresponding to real-world object. I'd like to use OpenCV.js (and appropriate other tools, like node.js) to create an interactive website for performing, evaluating, and visualizing image segmentation.


Clinical Next-Generation Sequencing Analysis Pipeline for Precision Medicine in Cancer at SLU

Organization: SLU Medical School
Client: Dr. Huazhang Guo
Supervisor: Dr. Tae-Hyuk (Ted) Ahn
Description:

Rapid evolving of next-generation sequencing (NGS) in genomic medicine has been driven by low cost, high throughput sequencing and rapid advances in our understanding of the genetic bases of human diseases including cancers. Today, the NGS method has dominated sequencing space in genomic research, and quickly entered clinical purpose. In this project, we will build a genomic clinical data analysis pipeline from lung/colon/melanoma cancer specimen. We plan to use diverse open-source programs and database to build this pipeline. Our goal is reducing significant amount of time and efforts to generate accurate pathology reports from the genomic data using the proposed pipeline software package. This will ultimately make an effect to Saint Louis University Hospital and precision medicine in cancer.