COMP 523: Project Teams

Project Proposals

After all the presentations have been made in class, please send me an email
(to stotts.comp523@gmail.com) with title "COMP 523 PROJECT PICKS".
Email is due Tuesday evening, 1/15/2013 by 5pm.

In this email give me the LETTERS (from the list below) of your top 6 project
choices... something like this: 1Q 2L 3H 4Z 5C 6K

I will do what I can to maximize happiness but I give no guarantees
other than you will end up assigned to some project.

You may self-select a team if you wish... simply send me an email with
all three student names as a team along with your 5 project preferences.
If you do not self-select a team, I will assign you individually to a team
based on project preference matches as best I can.

Summaries of project proposals

A) Multi-player Classroom Software
Diane Pozefsky, UNC Comp Sci
My first year seminar on the business of games is run as a multiplayer classroom and I need some software to help me manage it. Players (students) earn points based on health (attendance) and individual and group quests (assignments).
more details

B) nuTRUtion: Getting towards the truth about our food environments
Shu Wen Ng and Hye-Chung Kum, UNC Comp. Sci
The Population Informatics Research group in the computer science department and the UNC Food Research Program at the Population Center are working on a project to generate and share information about the nutritional quality of our food environments via smartphone apps and social media platforms. Data (digital images, text and numeric) from foods people encounter everyday can be used to create a dynamic database that includes up-to-date true information on the nutritional quality of food environments and allows users to learn about nutrition. Importantly, it will also form the basis for US nutrition monitoring. For such a system, we need good phone apps, a good image processing engine to turn images into nutritional facts, and then finally a dynamic website to check the data and feed nutritional information out to the users. Come work with us and make a different in the health of the US population!

C) Create a Web API for Natural Science Data
Brian C. Russel, NC Museum of Natural Sciences
Museum web site
In 1879 the North Carolina Museum of Natural Sciences was founded, in part, as an agricultural and geological collection. One hundred and thirty three years later, the Museum's collections continue to grow and currently contain over 3 million specimens. The collections have expanded beyond the original agricultural and geological scope to include mammals, birds, amphibians and reptiles, fishes, invertebrates, geology, and paleontology. These collections hold a myriad of data that can be used for education, resource management, research, and much more.

Help us create a language-independent web application programing interface (API) to connect a test subset of our data to the public Internet. We would prefer that the API allow for growth in eventually connecting all of our data because our goal is to allow public access to our data and thus increase the number of applications and mash-ups that can be made. Hopefully this will lead to more scientific discovery and educational opportunities. Examples using the API with python and ruby would be great and version control is a must. More specific details will be determined during the process.
The current science API

D) MESHVALMET: Object Match Evaluator extension
Stephen Pizer, UNC Computer Science
Use of MeshValMet is by developers or users of methods to extract anatomic objects from medical images. It allows them to compare their answer to the right answer. This is done by measuring the overlap of the interiors of the two objects or by measuring distances between corresponding points on the surfaces of the two objects and then displaying the objects with coloring according to distances to corresponding points, histograms of distances, and various numerical measures of agreement.

The project objective is to revise, refine, and document for public, web-resident distribution the MeshValMet code, which evaluates the difference between two objects in 3-Space. The revisions needed are
```
1) Improving the user interface, including better defaults for 
   the control parameters. 
2) Adding a capability for a user to add a code module specifying 
   how to compute point correspondences, to be used in place of the 
   default "closet point" mechanism presently in the code. 
3) Refactoring the code, in particular to allow revision 2. 
```
Full project description.

E) Shape Population Viewer
Beatriz Paniagua and Martin Styner, UNC Computer Science
In one of the major projects at the UNC Neuro Image Research Analysis Lab, Drs. Martin Styner and Beatriz Paniagua develop novel methodology for the computation of Statistical Shape Analysis. Statistical shape analysis is a geometrical analysis from a set of shapes in which statistics are measured to describe geometrical properties from similar shapes or different groups. Shape analysis has become of increasing interest to the medical community due to its potential to precisely locate morphological changes between healthy and pathological structures, i.e. the difference between male and female Gorilla skull shapes, normal and pathological bone shapes, etc.
Full project description.

F) PainTracker
Deborah Barrett, UNC School of Social Work
This project is an interactive web-based app for individuals with chronic pain. It allows users to collect customized self-survey data so that they can experiment and improve how they fare.
presentation slide

G) STENCIL
Julian Wooten and Cameron Musler, Social Venture Chapel Hill
STENCIL is cloud-based student information software that allows teachers to more easily manage student data. We are looking for students interested in helping us design the student management portal that would eventually integrate the algorithm that detects the likelihood of a student dropping out.
STENCIL presentation slides

H) Wave Generator Package
Leandra Vicci, UNC Computer Science
Wanna make waves???
We have recently received a grant from NSF to furnish a novel wave generator for the UNC Joint Fluids Lab wave tank in Chapman Hall. See the abstract for details. An important component of this system is an interface between a Simulink (or possibly Matlab) waveform generator and streaming digital outputs from multiple sound cards. One or two channels would be trivial, just like stereo sound output. To be useful however, up to sixteen parallel channels are needed which maintain synchronization across all channels. This would be best run on a Linux platform with multiple sound cards having suitable output capabilities. I would love to have a student team build a first prototype of this interface.

I) Adventure Game Creator
Gary Bishop, UNC Computer Science
In 2005, Eden Kung and Matt Clark developed a blind-accessible audio game Descent into Madness. It is so popular that it is used as a reward of Governor Morehead School for the Blind. Sadly, the original version no longer works on Windows 7.

Coming to the rescue, Peter Parente has begun development of a browser based version. He plans to show it at Maze Day 2013 on April 29th. As is his way, Pete didn't just hack out the game, he built a game boilerplate that is quite general and can be used as the basis for many similar games.

Meanwhile, I will be teaching my FYS Comp 80 Enabling Technology this semester with 24 eager freshmen who have 30 hour service commitments. If we had some simple tools, we could have groups of freshmen creating content for new browser-based adventure games.

The goal of this project is to develop tools that are usable by non-computer programmers to create games using Pete's dim-game boilerplate. They don't have to be pretty, or even GUI they just have to be sufficiently usable for non-CS students to be able to create and play content they create.

The FYS students won't need the tools to begin developing the storyline for their games. You'll use them as testers for your tools as they develop and they'll give you ideas for what else they need. I anticipate a feedback process that will work for both you and them.

J) Smarts and Stamina
Kathryn Britton and Marie Josee Shaar
Droid and iPhone app for sleep, diet, and exercise wellness.
Full project description.
In-class presentation.

K) Interactive Functionality for Celtic Culture Website
Dorothy Verkerk, UNC Art Department
My web site
Created in 1997, I am hoping to change it from a static web site to an interactive site that would offer the opportunity to have a world-wide conversation about Celtic cultures -- that includes anything from ancient artifacts, LOTR, Dungeons and Dragons, Druids, music, etc...

M) Transfer Support Ware
N.V. Stammes, UNC
Currently North Carolina does allow easy access to students and faculty to transfer classes between 2 year and/or 4 year institutions. Each institution has its unique course offerings and names, which makes it difficult for transfer students and advisors to navigate. To fill this gap I propose developing an advising program that with interconnectivity between the universities and recognize individual schools and their particular agreements. This should streamline the transfer process for both the students, advisors and schools.
Full project description.

N) Point-of-care mobile medical app
Ryan Madanick, UNC School of Medicine
We would like to develop a "point-of-care" mobile application (meaning real-time, at the point that medical care is taking place) for use on tablets & smartphones that could assist in a faculty member's assessment of trainees' progress in gastrointestinal endoscopy training. The instrument itself has been published, but it isn't widely used because there is no good way of implementing it at other training sites. We believe an app would break the barrier to use and could allow it to be used nationwide.

As we envision it, the app could be used by a faculty member to record data about the trainee's skills. These data would then be transmitted to a server and eventually used by the training director, with data reports, etc. Because all of medical training is moving toward using similar types of evaluation methods, we foresee this project as being the pilot study to a much larger research, and even business, opportunity.

O) Accessible iPad Games
Diane Brauner
There is a HUGE need for accessible materials as students with low vision and totally blind are successfully using the iPad in mainstream classrooms. Ed Summers, who heads up the accessibility team at SAS Institute and I have trained just about all the active Teachers of the Visually Impaired in NC on the native accessibility features of the iPad. This spring we are offering the next series of iPad accessibility trainings called "Reading, Writing and Research" geared to train Teachers of the Visually Impaired so that they can then teach their students. While working directly with VI students, I am finding areas of need. Some things we are passing on directly to Apple's accessibility folks, but other things can be resolved through apps.

We are encouraging blind and low vision students to use Voiceover (VO) and either a Bluetooth keyboard or Refreshable Braille Display to drive the iPad. (We do teach the gestures associated with VO, but it is more efficient and easier for VI students to use the unique keyboard and Refreshable Braille Display (RBD). As long as VO works with the app, the keyboard and RBD will also work. For these projects, your students do not need to know braille or how to use the keyboard commands/RBD. I'll be happy to show them how to use VO and gestures (as needed).

I can give specific examples of these, demo VO, etc. and explain the projects in detail -- please read the project ideas below to get an idea. Ed can also provide information on where to find the accessibility standards.

There are two BIG needs -- I am not a computer programmer, but I am assuming these two projects will be more involved. You'll have to judge if they are appropriate for this class or not.

Teacher Apps to be used with VI students:
1. Easy way to mark up materials: This is needed for ALL students/teachers, not just VI. Teachers are using teaching materials from the web and are creating their own materials on their computers. They need an easy way to combine PDF and textboxes/radio buttons so that the teacher's directions, reading materials, and questions can not be accidently changed by the student. In order to insert their answers, students need to have textboxes and radio buttons (for multiple choice questions). That way the student can insert their answers without accidently changing the body or question of the document. Most teacher made and on-line materials were created to be printed and copied -- fill in the blank lines are just underscore lines, so when the document is used on the iPad, the student can not type on the line. I'm looking for something like Google surveys but easier for teachers or braillists to "mark up" the document.
  
  Long term, I'd like to see an app that took a document and automatically interpreted the questions, replacing blank lines with textboxes and multiple choice answers with radio buttons. Short term, I'd like to see an app that the teachers/braillists can use to manually "mark up" a document. Teachers would be able to manually insert textboxes and radio buttons.
2. HTML mark up -- low vision and blind students use the keyboard commands and RBD to navigate through documents, books, websites, etc. VO has the "rotor" which allows the user to chose move to or go back to the next/last Heading, link, textbox, etc. However, teachers do not know how to create HTML documents. I'd like an easy template that teachers/braillists can insert their documents into (or create new documents using the HTML) This new HTML document could also be opened in iBooks. AND/or, I'd like a simple app that would allow a teacher/braillist to highlight a heading and tap on "heading" (which would then insert the correct HTML code).
  
  There are an incredible amount of preschool, k-12 education apps available for sighted students; only a handful of the commercial educational apps are truly accessible for blind students. It is a simple process to include the few labels, etc. that would make these games accessible. We particularly desperately need beginning letter recognition, reading, writing and beginning math game apps, as well as other concepts/games. Here are a few ideas and I am open to other ideas if your students have something particular in mind. Any VI app WILL be used, not only be students in NC and the US, but world wide. Our classrooms are becoming all digital, as are the high stakes testing -- we need games and materials that will allow blind students have access to digital materials and games at all levels!
EASIER game projects:
1. math apps - counting, addition, subtraction, multiplication, number line, etc. Hungry Fish is an example of wonderful math app that is frequently used in classrooms -- a few minor changes would make this accessible.
2. Letter recognition, beginning words, rhyming words, reading, writing, vowels, phonics, etc. Word Wagon is a great example of a beginning matching letters and spelling game -- a few minor changes would make this accessible.
3. Matching -- (remember the old card game Concentration?) Blind kids love fun sounds. This game could use pictures AND sounds for matching. I'd like to see simple beginning level building to complex levels. The simple level can be 4 cards, say a pair of ducks and a pair of cows. The next level might have 8 cards and the cards are identified by number (1-8). So the card would automatically give the number as the student moves across the card. When the student selects the card, the number then sound (1. quack, quack). The more complex level might have 16 cards and might be identified by number and letter (rows are 1-4 and columns are a, b, c, d). Teachers would be able to select the level or number of cards. The different sounds can be organized by category (farm animals, transportation sounds, household sounds, musical instruments). This game would teach spatial concepts, category, matching, and more!
I have more ideas -- and these game topics can certainly be expanded!

Ages 3-5

1. My older low vision preschoolers really need to practice scanning and mazes are a wonderful way to do that. I use apps (like Whizzy Kids) to practice some mazes, as well as take pictures of some I've created, have in books or find on the web, and then use them on my iPad. I would love to see a series of mazes really designed for younger kids that give some auditory feedback while they are using the maze, besides applause or cheering at the end. For example, low vision kids need to learn about boarders and boundaries and still have a hard time with them. Auditory or haptic feedback when they go off the path would be great.

2. Some neat apps I've discovered recently have been "Seek and Find" apps by Wonderkind and Baby Learns Simple objects. Both of these apps involve finding and activating/identifying everyday objects at home and in varying community environments. It would be great to have an app that began with a simply designed theme and then levels that added objects or events to the scene, i.e., more clutter. This might look similar to Christine Roman-Lantzy's CVI cards that gradually add more clutter as the pictures progress.

If a student team chooses any of my app ideas, I have the apps (such as Whizzy Kids, Peak-a-Boo Barn) on my iPad that I can show them.

P) Open Source Platform for Interactive Publications
Patrick Reynolds (KitWare)
An emerging problem in the transmission of scientific research is the growth and diversification of scientific data. A 2-dimensional figure or chart can rarely convey the nature of input data or a result. Using existing open source tools you will build an Android or iOS application for interactively displaying 3 dimensional data alongside text describing that data. The mobile application would deal with the display of the publications and a companion desktop or web application would deal with their creation.