The goal of the proposed project is to develop a differential mobility spectrometry-based technology for label-free molecular imaging and to pilot its applications in biological tissue analysis. Successful project would lay the foundation for widespread use of label-free molecular imaging that would revolutionise analysis of tissues by allowing large-scale adoption and realise unforeseen applications also outside the medical area.
Conventional molecular imaging comprises of two stages: In the first stage, a label substance that adheres to compounds of interest, is administered to the sample. In the second stage, the label is detected by e.g. magnetic resonance or optical imaging. Label-free molecular imaging simplifies the process by omitting the use of label substances and directly analysing the molecular content of the sample. It is an increasingly important method for investigating and visualising spatial distributions of biomarkers, metabolites and other molecules in life science field and can be used, for example in analysing tissue samples to detect molecular changes caused by diseases or to quantify pharmaceutical agents. The barrier of large-scale adoption of label-free imaging has been the requirement for mass spectrometry that is an expensive and maintenance-intensive technology. Differential mobility spectrometry is an emerging technology that has been demonstrated to match or exceed the performance of mass spectrometry at a cost of less than 10%.
The goal of the iDMS project is to develop new, robust, cost-effective method for label-free molecular imaging that is based on differential mobility spectrometry technology. The main objectives of the project are 1) to construct a molecular imaging system that utilises novel hopping-mode differential ion mobility spectrometry (HM-DMS) technology together with laser-desorption sampling, 2) to collect test data with the constructed system in pilot applications, 3) for developing qualitative tissue identification and classification methods and quantitative molecular concentration analysis methods.
The data analysis techniques developed in the project are based on advanced machine learning methods such as convolutional neural networks. The qualitative analysis development is focusing on spatial mapping of regions of cancerous and healthy tissue and quantitative analysis in estimation of concentration profiles of specific drug molecules in tissue. These use case examples will provide a proof of the concept of affordable DMS-based label-free molecular imaging.
The breakthrough of the proposed technology is the ability to dramatically reduce the cost of molecular imaging by substituting frail mass spectrometry technology by robust DMS that in most applications will provide equal performance. This will enable widespread adoption of label-free molecular imaging for the European scientific community and enables realisation of unforeseen commercial applications.