3D Geometry Reconstruction of Medical Images

The project focused on using numerical tools to analyze medical images and construct a 3D model of the organs present in the image. The goal was to provide a detailed 3D model that could be easily manipulated for use in the medical field. The focus was on Magnetic Resonance Images (MRI) of the pelvic anatomy of women patients, with an emphasis on three main organs, the bladder, vagina, and rectum. To construct the 3D models from the MRI scans computer vision was utilized as the main tool for image analysis. An object detection tool was developed and trained to detect, differentiate, and label the bladder, vagina, and rectum. These same images were converted to a greyscale, which allowed the organs of interest to be isolated as well as identified by drawing contours around them. An ellipse was then fitted into each contour, and from these ellipses certain parameters could be obtained. Including the center point (x, y, z), the length (l), height (h) and orientation (n). This data can then be utilized to create a 3D CAD model of the patients’ pelvic organs.

Stack

Computer VisionPython

Deployments

Screenshots

© 2024 Christopher Morales