Week 6: Eddie Wei - Confocal Fluorescence Image Collection

Figure 1: Confocal Fluorescence Image of Extracted patient’s Tumor Region in Bladder (Picture taken from phone of screen)


Data Collection - Confocal Fluorescence of Ex Vivo Bladder Tumor Samples

This week, I learned how to operate the SamanTree Histolog Scanner and use it to acquire confocal fluorescence image data on extracted tumor bladder samples. Like typical confocal fluorescence, the scanner uses 488 nm excitation light and optical sectioning to rapidly image freshly excised tissue. Before imaging, each specimen was briefly dipped in fluorescein and then rinsed in saline. With guidance from SamanTree specialists, we collected confocal fluorescence images from 10 bladder tumor specimens obtained during cystoscopic procedures. Once I receive the necessary hospital approvals and access to the required software, the next steps will be to export and organize these images, incorporate them into my current image-classification model, and begin training the model on clinically relevant bladder tissue data. Going forward, I will continue collecting additional tissue images to expand the dataset and improve the model’s performance. While awaiting approval to access the images, I have also been practicing how to interpret representative confocal images using an online training platform provided by the SamanTree representatives. This has helped me become more familiar with identifying tissue structures and image features that may be relevant for future classification. Next week, once granted access to export, I plan to analyze and label the tumor grades of these images with pathology experts.

Patient Interview and Surgery Shadowing


On July 7, I shadowed a couple patient interviews. One thing that I learned was that a conventional biopsy is generally not performed for a suspected testicular tumor because of the risk of disrupting the tumor and spreading malignant cells. Instead, the affected testicle is typically removed through a radical inguinal orchiectomy for both diagnosis and treatment. I also observed discussions involving May-Thurner syndrome, CT imaging in a patient with a neurostimulator, and the evaluation of hematuria for possible causes such as infection, stones, or urinary tract malignancy. On July 8, I observed a robotic procedure to remove a 3.3 cm adrenal mass located near the vena cava. Through a standard open procedure, I also observed the removal of the left kidney due to an arterial aneurysm, chronic thrombosis, and impaired blood flow caused by clot formation. What was interesting was the relatively excessive amount of blood that clotted and stored up in the extracted kidney. On July 9, I observed part of a ureteral reimplantation procedure performed to address a kidney obstruction. Because I arrived after the procedure had begun, I mainly observed the stage in which the ureter was reconnected to the bladder. I also observed an attempted robotic prostatectomy involving a patient with high-grade prostate cancer. Unlike the other robotic procedures I observed, Dr. Scherr decided to stop the operation because the prostate could not be safely identified through the extensive surrounding adipose tissue. Continuing would have created excessive risk of injury to nearby structures, bleeding, or other surgical complications.


Comments

Popular posts from this blog

Week 3: Eddie Wei - Multimodal Endoscope

Week One: Eddie Wei

Week 1- iqra