In this project, we are optimizing a machine learning protocol for extracting chemical information from spectral data. FTIR, Raman or surface-enhanced Raman scattering (SERS) techniques have the potential of unveiling hidden information from complex spectra of biochemical mixtures. As a result, from thousands of measurements, a large spectral data set is analyzed by machine learning methods. The methods proposed in this project will be used to detect disease-specific biomarkers at low concentrations from bio-samples (e.g. for the detection of Alzheimer’s) or to detect micro/nanoplastics in food.

Job description

• Design an experimental protocol to identify the limit of detection and the capability of the method.

• Increase the number of examples in the data set by systematic measurements for different spectroscopic techniques (Raman, SERS, …).

• Debug the machine learning algorithms that we already have, to improve the accuracy of detection.

• Identify additional machine learning models and strategies of classification/regression that fit to the main objective of the project. 

Main outcome of the project

• Optimized machine learning protocol for the detection of biomarkers from vibrational spectroscopy data.

• Description of the capabilities, performance, and benefits of the method proposed.

• A guideline for the implementation of machine learning algorithms applied to spectral data analysis process.

We encourage highly motivated candidates to carry out their final Master’s thesis in an international, multidisciplinary research environment. Preferably, the students should have the following background: Physics, Chemistry, Biochemistry, Biomedical Engineering, Computer Science, etc. For more information, please visit our website http://www.nanogune.eu/nanoengineering.

Application:

If you are a master student and you are interested in this project, please get in touch with the scientist in charge: Andreas Seifert (a.seifert@nanogune.eu)

To apply for a master scholarship fill in the form below and follow the instructions and recomendations of the general call open until 30 June 2020.