LECTURE
"Photonic Data Science: from data to information"
Abstract:
Various photonic technologies enable non-destructive measurements with minimal sample preparation, making them essential for a wide range of applications in both research and clinical settings. However, the challenge with label-free measurement techniques lies in the complexity of interpreting the raw data. By applying chemometrics and deep learning, we can extract valuable insights from spectroscopic and imaging data. In this lecture, we will explore the challenges and solutions related to experimental design, device-to-device comparability, with Raman spectroscopy as a primary example. Additionally, we will examine how deep learning techniques can enhance the interpretability of multimodal nonlinear imaging methods.

TUTORIAL
"Chemometrics for Raman spectroscopy"
Abstract:
In this hands-on tutorial, we will walk through a basic example of Raman data preprocessing, followed by both unsupervised and supervised data analysis techniques, including dimensionality reduction, clustering, and classification. The tutorial is Python-based and requires Git. However, for those who prefer not to install anything on their laptops, an online platform will be available that only requires an internet connection and a web browser.

BIO:
PostDoc, Leibniz Institute of Photonic Technology (Leibniz-IPHT)
Oleg Ryabchykov specializes in automating data processing pipelines for the analysis of spectroscopic and image data. He earned his PhD in 2019, focusing on data fusion of photonic and clinical data. In addition to his research, he develops user-friendly software for Raman spectroscopic data analysis and tools to facilitate experimental design.