Neue Sensorik für die multispektrale klinische Bildgebung

New sensor technology for multispectral clinical imaging

Duration: July 2023 - June 2026

Project Partner

  • HS Pforzheim
  • Fraunhofer HHI
  • Klinikum Darmstadt

Motivation and Objective

In the operating room (OR), surgeons rely on being able to diagnose and, if necessary, remove altered or diseased tissue. This detection is often done by visual color analysis with the eyes. However, different spectral compositions can lead to the same color impression. The differentiation of different tissue types, which are associated with different spectra, can therefore be significantly improved by analyzing numerous wavelength ranges. This is achieved through hyper- or multispectral imaging.

The overall goal of the NeoSpek project is to realize hyperspectral systems that allow real-time imaging, highlight tissue and 3D structure contrasts in an appealing and automatic manner, are insensitive to optical interference, and neither disturb nor interrupt the clinical process.

The Fraunhofer HHI will provide support with the following sub-goals in order to implement the overall project described above:

  • Evaluation of 3D and spectral information in real time
  • Automatic contrasting and visualization of spectral information
  • Implementation of improved tissue differentiation and perfusion analysis for clinical use
  • Demonstration of the above points using a demonstrator


At the beginning of the project, a system will be realized that records hyperspectral data with lighting correction. In a second step, automatic contouring will be developed to provide surgeons with an optimal image of the hyperspectral information. 
In addition, development work is being carried out on upsampling the light field data, i.e. the spatial resolution of the hyperspectral image data is to be scaled beyond the resolution of a few hundred pixels by utilizing the different viewing directions. 

Once high-resolution image data is available, tissue differentiation and perfusion analysis will be developed. In addition, there will be continuous clinical monitoring with intraoperative images and evaluation of the measurement data.


E. L. Wisotzky, A. Hilsamnn, P. Eisert,
3D Hyperspectral Light-Field Imaging: a first intraoperative implementation.
Current Directions in Biomedical Engineering, 9(1):611-614, 2023. doi:

E. L. Wisotzky, J. Triller, A. Hilsmann, P. Eisert,
Multispectral Stereo-Image Fusion for 3D Hyperspectral Scene Reconstruction.
In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP, pp. 88-99, 2024. doi:


NeoSpek is funded by the Carl-Zeiss-Stiftung.