Register now After registration you will be able to apply for this opportunity online.
Master’s Thesis Project: Revealing Trabecular Bone Architecture in 3D with Deep Learning based Super Resolution
The resolution of clinical CT images is not sufficient to characterize the complex network of trabecular bone. The student will develop a Super Resolution pipeline to enhance CT images and enable a characterization of the microarchitecture of human bone.
Keywords: Deep Learning, AI, Biomedical Imaging, Computer Vision, Bone Mechanics
Three tasks are foreseen: i) analysis of multi-resolution CT images of trabecular bone (3D registration, segmenta-tion, microstructural measurements). ii) Development and training of a generative model for 3D super resolution imaging of trabecular bone. iii) Characterization of model performance at predicting trabecular bone mor-phometric and microstructural properties.
Three tasks are foreseen: i) analysis of multi-resolution CT images of trabecular bone (3D registration, segmenta-tion, microstructural measurements). ii) Development and training of a generative model for 3D super resolution imaging of trabecular bone. iii) Characterization of model performance at predicting trabecular bone mor-phometric and microstructural properties.
This project’s aim is to implement a DL pipeline for the generation of Super Resolution images of trabecular bone from CT images of the human hip. The final goal is to enable a characterization of the architecture, fabric and micromechanics of trabecular bone from clinical CT images. We will utilize state-of-the-art diffusion models to push the physical spatial resolution limit of this type of scans.
This project’s aim is to implement a DL pipeline for the generation of Super Resolution images of trabecular bone from CT images of the human hip. The final goal is to enable a characterization of the architecture, fabric and micromechanics of trabecular bone from clinical CT images. We will utilize state-of-the-art diffusion models to push the physical spatial resolution limit of this type of scans.
Please send your CV and transcript to Gianluca Iori (giiori@ethz.ch) and Daniel Vera Nieto (daniel.vera-nieto@psi.ch). Links to previous work (e.g., your GitHub profile) are highly appreciated.
Please send your CV and transcript to Gianluca Iori (giiori@ethz.ch) and Daniel Vera Nieto (daniel.vera-nieto@psi.ch). Links to previous work (e.g., your GitHub profile) are highly appreciated.