Uncertainty-Aware Stenoses Segmentation in Digital Subtraction Angriography using Probabilistic Deep Learning

Overview

• Stenosis segmentation is essential for optimal stent choice

• Expert opinions vary depending on experience, education and skill level

• Probabilistic segmentation can support quantifying uncertainty in stenosis segmentation

  • https://arxiv.org/abs/1906.04045

  • https://arxiv.org/abs/1907.01949

• Quantify expert label uncertainty based on multi-expert labeled data

Skills

• Python programming experience (keras or pytorch, scikit, pandas, numpy,…)

• Deep Learning knowledge (CNN, U-net)

• Image processing knowledge

• Basic medical or anatomy knowledge is a plus

Tasks

• Literature research

• Data preparation and pre-processing pipeline implementation

• Probabilistic segmentation and/or detection model implementation, training and evaluation