Our Proposal

DSP BORDER

Objective

To reconstruct a two-dimensional cross-section of an actual body from one-dimensional data taken from a tomography machine.

Method

The first step in our project is to understand the analytic processes required in going from 1-D to 2-D - the understanding of the equations and their actual physical meaning.

Our project is going to be based on real biomedical data obtained from Baylor Medical School. The data we will be getting will be from tomography machines. It might be helpful for us to get a basic understanding of how the machine works in taking data.

We will have to determine what file format the data is in. Will we have to separate out each signal from one vector of data? We will need to know the granularity of each 1-D signal itself and how many there are. We also need to know what density the end-product 2-D picture should be. Is the data real or complex?

To reconstruct the signals into a 2-D picture, each point will have to have a magnitude, an angle, and a color. This color can be either gray-scale or full-color, so our picture could be either black and white or color.

We will have to create a filter specific to our data type to remove noise from the machine and any singularities in the data.

Steps

1. Understand the Math
2. Implement Code in MATLAB
3. Create simple test cases to fine-tune our program : take a circle to a 1-D vector
4. Send the actual data through
5. Write up the report and give our presentation

DSP BORDER
Back to the front page.