This summer, I worked on a project to study the differences in bloodflow in normal and diseased hearts. I was given several DICOM files of coronary angiograms, which I had to figure out how to analyze. Under the mentorship of Dr. Eleni Katifori, I began analysis by capturing several frames in sequential order from each video. Once this was done, I was left with thousands of files to sort through and analyze– this was a daunting process which was impossible to do manually in a timely manner. I used ImageJ, an image processing tool, to simplify the files to a binary and skeletonized form so that they could be analyzed using MATLAB. With ImageJ, I was able to equip both the built in functionality as well as plugins, which were posted online for open access, to process each image file. I also had to learn how to code macros in ImageJ so that I could batch process my images rather than manually applying the process to each individual file (which would have taken hours!) Once I had all of the processed image files, I simply ran the image sequences from each video clip through a MATLAB program and ended up with several graphs to compare and analyze.
Working on this project this summer was a really educational experience– I had done research before, but this was the first time I had this much control over the design and process. It challenged me to hone my problem-solving skills in order to get past the issues that came up throughout the process. I learned how to solve problems in unconventional ways, often using tools that seem low-tech or overly simple. I think that even more than learning how to use new tools and a new coding language, learning how to design solutions and solve problems on my own was the most valuable thing I learned during my research experience. I am excited to take these skills and apply them to situations I encounter in the future.