Project Details
Projekt
Infrared-based hyperspectral microscope
Subject Area
Medicine
Term
Funded in 2025
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 552405662
Treating chronic inflammatory diseases and cancer is complex and the effectiveness of available therapies is limited, while relevent adverse events can occur. Despite initial successful approaches, precision treatment concepts guiding the selection of the individually most effective and/or safest therapy and thereby substantially improving the quality of care, have so far only been established for a few selected conditions and constellations. Here, we apply for funding to acquire a infrared hyperspectral microscope (wave number 1900 to 950/cm) with two objectives (12.5x and 4.0x), which we want to use to develop algorithms to predict disease course and response to treatment in different inflammatory and malignant diseases. Whereas previous attempts to make such individual forecasts based on the expression of single molecules have only been partially successful, we aim to exploit the direct correlation of the infrared tissue spectra obtained by hyperspectral infrared microscopy with the overall molecular composition of the tissue leading to a "molecular fingerprint". The central hypothesis of our project is that individual disease progression and response to therapy depend on the entirety of molecular processes in the diseased tissue and that molecular fingerprints detected by hyperspectral infrared microscopy thereby allow a personalized prediction of these aspects. To this end, we will obtain infrared spectroscopic data on unlabeled tissue section and analyze them with artificial intelligance. Neuronal networks will be trained to identify spectra and locations specifically correlation with defined clinical endpoints. The perspective for the future is, that following thorough validation of such algorithms, hyperspectral infrared microscopy can provide individual predictive information within less than 20 minutes. In addition to the correlaton of spectral information with molecular tissue composition, a key advantage of the technique is that it uses unlabeled tissue samples allowing further downstream analyses. E.g., spatially resolved spectra can be correlated with spatially resolved single cell transcriptomics or laser capture microdissection can be used to isolate tissue areas, where spectra specifically correlating with the respective endpoints have been identified, for further transcriptomic or proteomic analyses. Collectively, we aim to use the requested device to improve treatment results for patients by bringing forward the individualized treatment of inflammation and cancer with a novel approach that is directly related to clinical problems and, thus, of highest clinical relevance.
DFG Programme
Major Research Instrumentation
Major Instrumentation
Infrarot-basiertes Hyperspektralmikroskop
Instrumentation Group
5730 Spezielle Laser und -Stabilisierungsgeräte (Frequenz, Mode)
Applicant Institution
Friedrich-Alexander-Universität Erlangen-Nürnberg
