The application area teaches further subject-specific skills for the practical application of Data Science and also contains modules from other subjects. In addition, the area allows students to take up to 10 cp in key qualifications from the TU Braunschweig's interdisciplinary pool, as well as optional project work in one of the application areas.
In total, there are 4 predefined application areas: (1) Image and Signal Processing, (2) Medicine, (3) Biology, Chemistry and Pharmacy, and (4) Data Science in Engineering. In addition, it is possible for students to define their own fields of application and focus from the range of subjects offered by the TU Braunschweig in consultation with the mentor (see mentoring concept) and the examination board.
(1) Image and Signal Processing
The application field of image and signal processing deals with the intelligent analysis of signals of any form, e.g. videos or images, biomedical recordings or speech dialog systems. These form special challenges due to form, mass, content and require specialized analysis and processing algorithms.
The variety of electives within the application field allows Data Science students to specialize in a targeted manner, positioning them in the dynamic environment of image and signal processing. Elective modules include general foundation modules, such as Fundamentals of Digital Signal Processing, Network Information Theory, Information Theory, and Signal Processing, as well as the targeted specialization modules Mathematical Image Processing, Biomedical Image and Signal Analysis, Computer Vision and Machine Learning, Deep Learning in Remote Sensing, and Speech Dialogue Systems.
Typical subsequent professional application fields include: Visual data analysis, application development for medical technology, as well as design and graphics programs, also for the entertainment industry (visual effects, computer games, film industry, 360° and 3D videos), medicine (medical image processing, digital operation planning), automotive industry (driver assistance systems), industrial manufacturing (visual quality control) and digital photography.
The image and signal processing application field offers excellent job prospects in growing industries such as machine vision, the optical industry, medical imaging, automotive, and media design. Graduates can work as software developers in the above-mentioned industries, but also in project management and quality assurance. There are also very good opportunities in the field of start-ups.
Module Overview:
- Biomedical Image and Signal Analysis (5 cp)
- Computer Vision und Machine Learning (5 cp)
- Deep Learning for imaging in nano and quantum science (5 cp)
- Deep learning in remote sensing (5 cp)
- Digital Signal Processing (8 cp)
- Fundamentals of Digital Signal Processing (5 cp)
- Information Theory and Signal Processing (10 cp)
- Machine Learning (5 cp)
- Mathematical Image Processing (10 cp)
- Network Information Theory (6 cp)
- Spoken Language Processing (5 cp)
(2) Medicine
The application field of medicine comprises the structured organization, representation and analysis of "medical" data. The World Health Organization (WHO) defines health and well-being in terms of environment, behavior, physiology and psychology, so that "medical" data includes not only laboratory values or blood pressure readings, but also air pollution, climate or lifestyle. Measured values and data can be collected selectively, e.g. in the context of treatment, but continuous monitoring of (vital) parameters is also becoming increasingly important. The aim is to detect the development of disease and to prevent it through early intervention.
In this environment, the application field of medical informatics offers a variety of choices. Medical subjects focus on anatomical, physiological, and biochemical data from both healthy and sick individuals. Here, Data Science students have the opportunity to experience the professional field of physicians and nurses. In the clinical and methodological majors, students can generate and understand 3D ultrasound images of their own bodies, and learn how to handle large amounts of data using digital microscopy images from the Whole Slide Imaging process.
Assistive Health Technologies primarily collects motion and environmental data to draw conclusions about health status. Likewise, students learn to apply the strict regulations for conducting clinical drug and medical device studies to their own, possibly purely scientific, work, formulate clear hypotheses, design adequate experiments, identify relevant influencing factors, and methodically control them. Finally, accident informatics encompasses the automatic exchange of information between alarm-issuing systems, the emergency medical services, and first responders. Data analysis with machine learning can be deepened with many examples of medical signal and image analysis.
Module overview:
- Health-Enabling Technologies A (6 cp)
- Health-Enabling Technologies B (5 cp)
- Selected Topics of Representation and Analysis of Medical Data (5 cp)
- Biomedical Image and Signal Analysis (5 cp)
- Medical-methodological specialization module 1 (5 cp)
- Medical-methodological specialization module 2 (5 cp)
- Accident Informatics (5 cp)
(3) Biology, Chemistry and Pharmacy
Due to technological progress in the field of various measurement techniques, more and more "high-throughput methods" are being used for data collection in the fields of biology, chemistry and pharmacy. This has resulted in a flood of molecular data that can no longer be analyzed without the support of computer-based methods. Data Science is becoming increasingly important in these fields.
In the corresponding application area, students gain insight into the generation of molecular data in biological, chemical, and pharmaceutical contexts. The understanding of data generation is indispensable both for a meaningful analysis of the data and to be able to classify the possibilities of the corresponding data collections in the first place. Although many data would be very helpful from a data science perspective, their collection is often hardly or not at all possible, as for example in the case of biopsies from healthy subjects for comparison with corresponding tissue samples from patients. Beyond data generation, the application area focuses on teaching analytical methods for understanding large-scale molecular data and biological networks. This includes machine learning techniques to identify biomarker signatures and graph-based approaches to enter the field of systems medicine.
There is currently a severe shortage of personnel in both research and industry who can adequately analyze the extensive molecular data that is being collected. In particular, there is a massive need for sound, robust evaluation in order to derive generalizable insights from the data that can lead, for example, to new treatment methods. This is repeatedly reflected in unexpectedly high "Numbers needed to treat", i.e. the phenomenon that a medication is often only actually effective in a fraction of patients. This situation results in excellent job market prospects for graduates with a correspondingly chosen specialization.
Module Overview:
- Applied Bioinformatics (10 cp)
- Elucidation and Modeling of Biological Structures (8 cp)
- Introduction to Chemometrics for Pharmaceutical Engineers (6 cp)
- Immunmetabolism (10 cp)
- Metabolic modelling (Flexi-Modul) (5 cp)
- Network Biology (5 cp)
- Numerical Ecology (5 cp)
- Applied Bioinformatics (10 cp)
- Theoretical Biophysical Chemistry (8 cp)
(4) Data Science in Engineering
In the area of "Data Science in Engineering", students learn to apply and evaluate methods of Data Science in different application contexts from engineering. Students can take modules that demonstrate the use of Data Science methods in different engineering contexts as well as modules that deal with basic engineering methods in order to become familiar with the conceptual world and methodological approaches of engineering.
Students should be enabled to understand, evaluate, and appropriately apply modeling concepts and methods in engineering in order to competently select and develop models in engineering problems. Possible areas in engineering sciences are fluid mechanics or product development in mechanical engineering or modeling and analysis problems in civil and geoengineering.
Module Overview:
- Data-Driven Material Modeling (6 cp)
- Deep learning in remote sensing (5 cp)
- Fundamentals of Turbulence modeling (5 cp)
- Automotive Software Engineering (5 cp)
- Fundamentals of Coastal Engineering (6 cp)
- Fundamental Measurement Methods in Fluid Mechanics (5 cp)
- Introduction to Finite Element Methods (5 cp)
- Machine Learning (5 cp)
- Ecological Modeling (6 cp)
- Railway Timetabling & Simulations (6 cp)
Presentation "Application Area: Data Science in Engineering"
