Week5

For my project, I did a summery of what I've done and made comparison between different algorithms I have tried for this project. I also read another paper of an interesting algorithm. Apart from my project, I also had a chance to go to OR and shadowed two surgeries, although I didn't make it to the end because I wasn't feeling well that day.


Fig 1a: Neuron outlines detected by watershed algorithm;
Fig 1b: Neuron outlines detected by region-growing based method. 
  1. Project updates:
    By far, I have tried three methods for automatic image segmentation: watershed, watershed-based neutrosophic method, and regrowing-based method. The first method, watershed, works very well for most of the neurons, but neurons in the region of low signal cannot be detected. A neutrosophic methods is developed based on watershed. In this algorithm, data is converted from image domain to a neutrosophic domain. However, several non-linear transformations amplify the noise, and therefore degrade segmentation results. The last methods is based on region growing. It can work well to separate overlaying neurons, but still cannot detect neurons with low signal magnitude. Fig 1 presents results of the first and third method, since the second method doesn't work very well.
  2. Clinical experience:
    I also had a chance to go to OR this Thursday, and shadowed two surgeries in the afternoon. One of the surgeries was to fix the leakage of cerebrospinal fluid. In order to locate the origin of leakage, the doctors injected a kind of dye which can turn the cerebrospinal fluid to green, so the origin can be traced by the green dye. What is also impressing to me is that the doctors burned the tissue to stop the leakage. Although it seems to be a commonly used technique in the surgery, I haven't seen that before.

Comments

Post a Comment

Popular posts from this blog

Week7

With the end of our summer immersion approaching, I feel sad to say goodbye to all the persons I've met in the hospital during this summer. The past 7 weeks has been very fulfilling and exciting to me, and I will never forget my experience here. As a student will electrical engineering background, I never thought that I can be able to go to a real operation room in the hospital and observe doctors operating on surgeries. But in the past 7 weeks, I was able to go to OR every week, and there's always surgery techniques that are impressing to me. Beside clinical experience, I also had a research project, automatic segmentation of zebra fish neuron images. As I have presented in my previous bloggers, for my project, I have tried several different algorithms. The first is simple thresholding. It is usually used as a preprocessing step in image analysis. The second method is watershed algorithm. It works pretty well, but, it's very sensitive to noise, resulting in over segmenta
Read more

Week 5: Spine Surgeries and Excel Files

Image
On the context of my clinical research project, I continue extracting the different demographic and skeletal specific factors for the patients of our study. I was also able to attend an additional spinal fusion surgery under the supervision of Dr. Kim at HSS. During this week, I was particularly interested in the surgical equipment that he uses for the procedure. Dr. Kim uses the Zimmer Biomet Vitality Spinal Fixation System for his spine surgeries. The components of the system are made from medical grade titanium alloy and medical grade cobalt chromium alloy. In summary, these were the main steps of the process for installing the screws into the patient spine which, personally, was the most interesting part of the procedure.  Initially, the pedicle area of the vertebrae is localized, and an insertion is made in the cortical shell. Then, a probe, included with the system, creates a pathway for the polyaxial pedicle screw to be inserted into the pedicle and the vertebral body. It i
Read more

Week 6 (Jason Chang)

Image
With only a week of immersion left, I have been focusing all my time on developing a preliminary proof-of-concept protocol for automating EEG artifact removal in MATLAB. Admittedly, it has been very challenging at times, especially given my limited neurology and programming experience, but I have made significant progress as of recently with the assistance of Dr. Mangat and Dr. Forgács. Last week, I began using EEGLAB, an interactive EEG analysis toolbox in MATLAB, to perform independent component analysis (ICA) on selected datasets from the subarachnoid hemorrhage (SAH) study. After ICA is applied, EEGLAB has different built-in signals processing functions to mark and remove components containing any artifacts. With several methods to choose from, I consulted with Dr. Forgács on Thursday to seek his clinical expertise. Figure 1 . Individual component analysis (ICA) to reject artifacts by visual inspection of continuous EEG signals  Figure 2. Removing ICA components using
Read more