Data-Driven Modeling for Light-Based Diagnostics and Analytics

Our Research

We study the entire lifecycle of photonic data—from generation and analysis to long-term archiving. Our goal is to systematically harness complex measurement data from diverse photonic techniques and to establish robust, data-driven foundations for analytics, diagnostics, and therapy. Our work bridges methodological research with direct applications in medicine, the life and environmental sciences, and pharmacy.

A central focus is the development of holistic data pipelines. These include experimental and sample size planning, structured data preprocessing, and the integration of chemometric methods with model transfer techniques and machine learning. This enables data of different origins and quality levels to be made comparable and jointly analyzed. Particular emphasis is placed on the transferability of models across measurement systems and studies.

The Photonic Data Science department develops methods for data fusion of heterogeneous sources as well as for the simulation of photonic measurement techniques to specifically improve correction and optimization strategies. In parallel, we investigate approaches for interpreting data-driven models to ensure analytical results are transparent and reproducible. The methods developed are translated into software packages and validated in real-world application scenarios, such as clinical studies. Another key focus is the development of data infrastructures for photonic measurement data in accordance with the FAIR principles.

Prof. Dr. Thomas Bocklitz

Head of Department
+49 3641 948328
thomas.bocklitz@leibniz-ipht.de

Ina Jahn

Assistant
+49 3641 206248
ina.Jahn@leibniz-ipht.de

Research Focus Areas

Machine Learning for
Photonic Imaging & Spectral Data

Development and application of AI methods for the analysis of photonic imaging and spectral data in analytics and diagnostics

Chemometrics &
Data-Driven Analytics

Chemometric methods and applied data analysis for spectral data from chemistry, physics, and the life sciences

Data Fusion &
Model Transfer

Correlation and fusion of heterogeneous measurement data as well as transfer of analysis models between different photonic techniques and studies

Data Infrastructures &
Software for Photonic Data

Development of FAIR data structures, analysis pipelines, and software solutions covering the entire data lifecycle of photonic measurement data

Working Groups

Photonic Data Science

Prof. Dr. Thomas Bocklitz

AI Laboratory for IR

Dr. Shuxia Guo

Collaborations and Networks

The Photonic Data Science department is closely integrated into interdisciplinary research networks and collaborates with partners from photonics, the life sciences, medicine, and environmental research. Its methods for data analysis, modeling, and data integration are applied and further developed in numerous application-oriented projects, particularly where complex photonic measurement techniques intersect with clinical or biological research questions.

At the Leibniz Center for Photonics in Infection Research (LPI), the department contributes data science methods for the analysis, interpretation, and integration of photonic measurement data, for example in the context of clinical studies and translational research approaches. The close coupling of methodological development and application strengthens the use of data-driven methods across the entire innovation chain.

Through collaboration with other research departments at Leibniz IPHT as well as with external academic and clinical partners, the department contributes to the standardization, comparability, and sustainable use of photonic data. In doing so, it supports the establishment of Photonic Data Science as a key competence for data-based analytics and diagnostics.