All posts by Merle Losch

Neural network for tissue classification

The spine is the central support structure of the body that helps us humans sit, stand up and walk around, twist and bend. Age-related degeneration, but also congenital deformities can cause back pain and spinal instability, requiring surgical fixation of the spine. To guide the surgeon’s drill during spine surgery, the tissue in front of its tip is assessed with a fiber-optic sensing system, but we currently don’t know how to evaluate the recorded data.

The focus of this graduation project in collaboration with Philips Research will be to train a neural network with data from different tissues found in/around the spine to develop a classifier that will guide surgeons during spine surgery. This project does not require preliminary AI knowledge or programming skills, although knowing some python basics will be of use.

Interested? Contact Merle Losch, m.s.losch@tudelft.nl

Design of a steerable device for spine surgery

Spinal fusion is one of the most common surgical procedures in the world. At the BITE group, we are developing a novel drill that allows for the surgeon to steer through the vertebra along a secure drilling trajectory, avoiding nerves and blood vessels that run along the spinal column. To help the surgeon find and maintain the right trajectory, a fiber-optic sensing system will be integrated into the drill to provide the surgeon with positional feedback in real time.

For this graduation project, inspiration will be drawn from nature to design a steerable device for spine surgery. A 3D-printed model will be built, and its usability for steering through the vertebra will be assessed.

Interested? Contact Merle Losch, m.s.losch@tudelft.nl.