This repository contains the official PyTorch implementation for the paper Depth to Anatomy: Learning Internal Organ Locations from Surface Depth Images
"Depth to Anatomy" is a new AI-based method that can predict the 3D shape and position of internal organs using just a single 2D depth image of a person’s body. Unlike traditional methods, it doesn’t need extra steps like building a 3D model of the body’s surface or aligning everything to a standard reference. This makes it much faster and easier to use in real medical settings. The system uses a deep learning model called Pix2Vox and has shown high accuracy in locating different organs. This could help improve how CT or MRI scans are planned - by automatically positioning the patient correctly without relying on external markers or extra preview scans. As a result, it could save time, reduce errors, and make the scanning process more consistent across different technicians.
