AI-based segmentation of diseased skin areas as part of a monitoring app for patients with chronic skin illnesses

HiWi position at the FZI.

AI-based segmentation of diseased skin areas as part of a monitoring app for patients with chronic skin illnesses

Dematologie FZI

Context

The aim of the HybridVITA project is the contactless dermatological assessment of chronic skin diseases using a patient smartphone. Among other things, video-based photoplethysmography observation is to be used to measure the pulse at an affected skin area and to draw conclusions about perfusion based on the time offset to a reference measurement in order to observe disease progression. This measurement requires reliable automated segmentation of the skin areas to be observed.

Tasks

The following challenges await you within the project

  • You will research the current state of the art and science.
  • You will develop a data management system and automatically create ground truth data for machine learning.
  • You will design and train intelligent algorithms for the segmentation of diseased skin areas in video recordings.

Requirements

  • You are motivated, proactive and conscientious.
  • You have experience with Python and machine learning.