Front Cover page in the Journal of Biophotonics
We are thrilled that the Journal of Biophotonics selected our publication as front cover page for the latest issue. In that work we showed how label-free optical technologies might benefit diagnosis of lung diseases in the future.
For that paper, Dr. Lucas Kreiss and colleagues at MBT and the University Hospital Erlangen (UKER) developed an advanced optical system that can reveal microscopic tissue structures, as well as the biochemical composition of biological samples. This system elegantly integrates Raman spectroscopy to our multiphoton microscope. Since both of these techniques can solely rely on natural optical contrast, it does not require extensive staining or preparation of samples.
We tested this ‘label-free’ setting, in a large experimental study on chronic inflammatory lung fibrosis, in close collaboration with our medical partners at the UKER. This experimental model mimics several key features of chronic lung diseases, like chronic obstructive pulmonary diseases (COPD) or idiopathic pulmonary fibrosis (IPF). The careful analysis of our data showed that label-free features from Raman spectroscopy and multiphoton microscopy correlate very well with the severity of the disease. Thus, these features might represent potential biomarkers that can be accessed all-optically from un-processed samples.
Finally, we used data-driven machine learning models to automate the detection of inflammation and fibrosis. Together with Alexander Mühlberg, our expert for Artificial Intelligence (AI) and machine learning at the MBT, we used random forest models to differentiate fibrosis from healthy control samples automatically, based on our label-free tissue features. When combining all features, these models were even able to predict the severity of the disease.
In the future, this combination of label-free optical technologies supported my AI could support doctors for easier and faster diagnosis of several inflammatory diseases.