Artificial Intelligence global team

Experience with deep learning architectures for tasks such as semantic/instance segmentation, image classification, and image synthesis using GANs, applied to medical images (CT/MRI) - PyTorch.

"Deep Learning for Markerless Surgical Navigation in Orthopedics"The thesis focused on navigation for TKA and presented the following solutions:- Markerless video-based surgical navigation system;- Proof-of-concept for automatic detection of anatomical landmarks in image-free navigation.Part of this work has resulted in the paper "Towards markerless computer-aided surgery combining deep segmentation and geometric pose estimation: application in total knee arthroplasty", presented at the AE-CAI | CARE | OR 2.0 Joint MICCAI Workshop 2020 and published in the journal "Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization".

Summer Internship focused on using Deep Learning for Image Segmentation.- Introduction to Machine Learning and Deep Learning;- Trained and tested a deep learning semantic segmentation network (U-Net) in images from video of TKA surgery; - Trained and tested a deep learning instance segmentation network (Mask-RCNN) with multiple datasets of general scenes (BOP datasets); - Participated in the BOP Challenge 2019 (https://bop.felk.cvut.cz/challenges/) with the method "Félix&Neves-ICRA2017-IET2019".

NEDF/AAC provides support to students, organizing main events in academic qualifications and pedagogy areas.