KI-Morph

KI-Morph: Artificial Intelligence for Automated Segmentation of 3D Image Data for the Analysis of Morphological Structures

“KI-Morph: Artificial Intelligence for Automated Segmentation of 3D Image Data for the Analysis of Morphological Structures”. By harnessing the power of artificial intelligence, KI-Morph project automates the segmentation process, saving valuable time and resources while ensuring accurate results.

Objectives & Goals

The rapid advancement in imaging techniques for X-ray tomography has expanded its applications across various domains within the life sciences. X-ray tomography is utilized to generate 3D data of tissues, ranging from individual cells and their components to tissue clusters, organs, and complete organisms. Subsequently, this 3D data is employed for constructing models and scrutinizing the morphology of specimens.

Despite the automated nature of 3D data generation through tomographic methods, which typically only takes a few minutes, the subsequent image processing and analysis remain predominantly manual, consuming significant time even for experienced researchers. Consequently, the image analysis phase constitutes a significant bottleneck in the entire processing workflow. In our innovative research project, KI-Morph, we aim to tackle this challenge by providing a framework tailored for petabyte-scale image processing and analysis.

Areas of Research

Our focus lies on three primary components:

  • Establishing infrastructure for processing petabyte-scale imaging data.
  • Developing AI algorithms tailored for segmenting large-scale 3D tomography data.
  • Assessing the processing pipeline's performance using data from various fields within the life sciences.
Project team at KIT-IPS
  • Dr. Alexey Ershov
  • Dr. Tomas Farago
  • Xiaoying Tan
In collaboration with
  • Engineering Mathematics and Computing Lab (EMCL), Prof. Dr. Vincent Heuveline, IWR, Heidelberg University
  • Service-Bereich Future IT - Research and Education (FIRE), Dr. Martin Baumann, University Computing Centre, Heidelberg University
  • Centre for Organismal Studies (COS), Prof. Dr. Joachim Wittbrodt, Heidelberg University
  • Competence center for biodiversity and integrative taxonomy (KomBioTa), Prof. Dr. Lars Krogmann, Museum of Natural History Stuttgart
Funding

The project is funded by the Federal Ministry of Education and Research (BMBF), 2023-2026.

Project website

(coming soon)