Journal Notes
Week 6
Day 27 – Sunday, July 27, 2003
I met with Zhenrong today, and the meeting went great. I was able to get all of my questions answered and now have a clear idea of what I need to work on. For example, there are some issues with some of the data Zhenrong had originally used being corrupted. I had tried to regenerate the data, but it took quite a bit of time, and in the end it still was not generated properly. I am going to work on fixing some of these errors so that I could run the program smoothly. I am also going to work on extracting the data for the left lung. Zhenrong said that she had set certain parameters in order for the program to work properly on the data from the right lung and that she is not sure how well it will work on the left lung. In any case these are some of the problems that I am hoping to solve soon.
Day 28 – Monday, July 28, 2003
Today, I tried to regenerate the corrupted data twice, from different points. Each took over an hour to run, and still I could not regenerate proper data. I am going to take a look at the errors I got and see if I can correct this. I am also trying to get the data I need in order to test the program on the left lung. In order to do so I need to take get the set of images from the Visible Human Data set which are in RAW format and covert them to JPEGS. I am able to do so, however the have not yet been able to set it to the correct size, and they images do not look exactly like the previous data that was used by Zhenrong. Right now, I am working on correcting this, and hope that I can do so quickly. I have also written about the first two pages documenting how to get this program up an running so that the next student working on it can do so quickly and smoothly.
Day 29 – Tuesday, July 29, 2003
I did not try to regenerate the corrupted data of Zhenrong’s again today. I decided instead decided to try and find a way to get the data for the left lung. In order to do this, I first need to find a viewer that will be able to convert RAW format to JPG format. The two options that I have come up with are IrfanView and Photoshop. I tested a sample image using IrfanView and found that the results were not as good as the data set that Zhenrong had prepared for the right lung. I then test an image using Photoshop. I found that the vewer in Photoshop is better, but the result image was the same. I can not figure out what Zhenrong used to prepare the data, but decided that IrfanView was good enough for what I need to do this summer. Linda and I decided against Photoshop and chose it IrfanView because is a free download.
Day 30 – Wednesday, July 30, 2003
Today, I attended the weekly SIRS meeting. In about two weeks, Cornelius would like all of the summer students to present what they have been working on. I also found out that I have to give a formal presentation of what I have been working on/learned on August 20, along with other students who are working in the bioinformatics area. I spent the rest of the day trying to figure out why I was not about to regenerate the corrupted data, and also figuring out how much I can complete by the end of August before I leave. I am hoping to be able to regenerate the blood vessel trees in both the left and right lung, and also testing the data on organs similar in shape. I would also be able to create an interface in which the results of the different steps of the EM algorithm can be displayed. Other than that, I would also like to complete a thorough documentation of the program incorporating what I have learned about it.
Day 31 – Thursday, July 31, 2003
Today, I spent a good portion of the day getting the data set for the left lung. This required me to download the full set of images (over 150), which are in zipped and in RAW format. Therefore after downloading the images, I unzipped them all. After that I had to take each individual image from the data set, and open then up in IrfanView (an image viewer), set certain specifications and saved them all as JPEGS because that is the format that Zhenrong’s program inputs. This however had to be done separately for each image. I also attended the weekly BISTI meeting in which John Gennari spoke about a conference on Protégé that he attended, and how it will fit in with the BISTI project.
Day 32 – Friday, August 1, 2003
Today I spent the morning running the set of images that I prepared yesterday through Zhenrong’s program. The purpose of this was to crop each of the images so that the left lung is visible in the new image. This took a while seeing as there are over 150 images and each can be run through the program individually. After that I tried to select in one of the images what I felt was an artery, so that I could trace it through the set of images and see if I could create the branch. The artery I had picked to trace/regenerate took about three hours, and the result was not good. One of the classes that is supposed to be automatically generated when running the program is not getting created and I am having a difficult time finding out why. However, while the program was running, I did use a laptop and continued working on my document.