Automated Image and Data Analysis in NACIP
- Contact:
- Project Group:
Machine Learning for Time Series and Images (ML4TIME)
Machine Learning for High-Throughput Methods and Mechatronics (ML4HOME)
- Funding:
HGF
- Partner:
Helmholtz Program (PoF IV): Natural, Artificial and Cognitive Information Processing (NACIP)
(BioInterfaces in Technology and Medicine (BIFTM), PoF III)
Description:
1. Development of automated image and data analysis algorithms for high-throughput processes in microscopy as part of the Screening Hub and Imaging Hub of the “Biointerfaces in Technology and Medicine (BIFTM)” program.
- Main focus areas in the danio application scenario: Analysis and visualization methods for cell membranes in SPIM data, motion video data, analysis of the resulting time series, analysis of small molecule screens. In cooperation with ITG (AG Strähle, AG Geisler, AG Grabher, AG Dickmeis, AG Foulkes), APH (AG Nienhaus), University of Ulm (AG Rottbauer, AG Just), University of Birmingham (AG Mueller), Howard Hughes Medical Institute, Janelia Farm Research Campus (AG Keller).
- Main focus areas in cell type detection, classification, and description for high-throughput toxicology and genetic screens: Comparative analysis of experimental data for several cell lines. In cooperation with ITG (AG Weiss, AG Davidson, AG Levkin).
2. User support in analysis of biological, biomedical, and chemical images and data, among other things, as part of the screening hubs and bioinformatics platform of the BIFTM program, including statistical consultation for applications to experiment on animals.
In cooperation with (a selection) ITG (AG Strähle, AG Schepers, AG Rudolf), IAR (AG Schultz), University of London (AG Hilpert), University of Leipzig (AG Hoffmann), University of Bonn (AG Perner), University of Bochum (AG Rosenhahn), Aberystwyth University (AG Huws), University of Heidelberg (AG Rupp, AG Wolf), University of Warwick (AG Levi), Bauerfeind.
3. New methods for feedback-based segmentation and visualization of complex 3D image data sets, for use on electron microscopy images – Heika Project, duration 1/1/2015 – 12/31/2015. In cooperation with AG Leitte, University of Heidelberg, Project supervisor: Dr. Reischl.
4. Development of a new method for the control of human-machine interfaces:
- Automated bilateral calibration for myoelectric signals
- Final testing, validation, and documentation (BMBF project, TELMYOS, completion on 06/31/2015)
In cooperation with University of Göttingen (AG Liebetanz), University of Heidelberg (AG Rupp)
Project supervisor: Dr. Reischl
5. Development of a new method for describing human reactions to stimulus signals in energy systems (data analysis, modeling, simulation, FE 5420.0051.6381) (DFG project, completion of the 1st phase on 04/30/ 2015):
- Analysis of Residens data
- Simulation in closed control loop
- Data analysis as part of EnergyLab 2.0 (under construction)
- Integration in the MATLAB toolbox, Gait-CAD energy extension, and in MATPOWER. In cooperation with: TU Ilmenau (AG Westermann), FhG AST Ilmenau (AG Bretschneider)
Project supervisor: Prof. Mikut
6. Software development (open source):
- Provision of an operational version of the cluster-enabled n-D image processing software (XPIWIT) based on C++, ITK; in cooperation with Large Scale Data Facility (LSDF), SCC (AG v. Wezel), IPE (AG Stotzka) Form version – 01/2013
- Gait-CAD: Main toolbox extension to match data quality dimensions, compatibility with the new MATLAB versions, Gait-CAD version 2015a/b on SourceForge
7. Lectures in the Mechanical Engineering Department (responsible for: Computational Intelligence (new) in the 2015/16 winter semester, Computational Intelligence III in the 2015 summer semester).