Hello everyone, I am from team 10 for Senior Design from Cal State LA, my name is Christopher Morales, the project is 3D Geometry Reconstruction of Medical Images. Where we're proposing to our clients who are gynecology organs specifically in the bladder, vagina and rectum. Where we will use MRI scans to determine if any patients have a prolapse; which consist of the pelvic floor muscles and ligaments detected being weak and stretched. To identify medical images, we are going to use a method called Computer Vision. Where computer vision is a technique to identify and understand an image. With the use of computer vision, we are able to detect the bladder, rectum, and vagina using MRI Images. Where we will draw an ellipse inside the wanted object in the image to gather information that will be sent to a CAD program. By using the data obtained from the MRI scans through computer vision, we are creating a three-dimensional CAD model of the pelvic floor. In our case, we are utilizing Free CAD because it is an open source parametric 3D modeler which we can feed the information necessary to generate a unique model.
It is important to note that while I have access to MRI images, I cannot share them due to confidentiality agreements with both the CSULA project and the patients. Therefore, as an alternative, I will be utilizing a black and white filtered image of my cat.
Input picture here!!! - Arty Chan
The research relevance is based on the notion of 3D Medical Images to grasp an internal representation of manipulating images and give a better observation tool for medical doctors. Implementing Computer Vision to MRI images allows the freedom to interpret MRI images with a 3D render applied, in this case, to the pelvic cavity. Every pelvic organ will be reconstructed with lofted ellipses identified on MRI images that constitutes the 3D volume. Applying a morphological transformation to the MRI images will result in finding the wanted organ to determine the ellipse that contours it to be used for a 3D model.